Community Supervision

Transcription

1 Community Supervision PROTECT THE PUBLIC SUPPORT RECOVERY ASSESS AND EXTRACT THREATS Helen Harberts

2 Probation clients need help! Community supervision is critical to their success!

3

4 We use evidence based practices to improve outcomes. Risk-Needs-Responsivity. [RNR] RISK: who to target with increased time and services NEED: What things to target with services and interventions. Allows change in dynamic factors that are proven to be associated with criminal behavior. RESPONSIVITY: Matching services and needs to improve outcomes. How to best meet the varied needs of our participants.

5 Core Correctional Practices Using validated risk assessments. Normed and validated to the population you are working with. DUI Population is different! Follow, and repeat the assessments! Develop and adjust case plan based on RNR principles Treatment plans should be incorporated and reinforced Spend time modeling and working in a positive manner with probationers.

6 Who has seen this? Do you have it?

7 Dosage matters Not years, dosage. Dosage is driven by assessment +professional judgment. Assessment give us the risk level and needs of the offender required in order to intervene in criminal behavior. Too much intervention does harm Too little intervention does harm

8 What is risk? Risk to fail probation without more structure. It does NOT refer to the ultra high risk public safety folks. (Nor do these recommendations)

9 DUI vs Regular Assessment The population is not homogenous DWI will have an entire spectrum of offenders Criminogenic Needs Assessment, Ongoing Not one size fits all Often low on risk assessments-wrong! entitled, educated, sophisticated and VERY SLIPPERY Resistant to supervision Team may be too comfortable with them

10 Definitions Community Supervision is: The effort to monitor the offender s behavior and program compliance outside of the court room and to support change by addressing good and bad target behaviors quickly.

11 Who Can Do Community Supervision? Badges-extend to 24/7 Do NOT send unsafe and untrained people into the field to do home visits and searches. Do NOT do it.

12 Is this important? No, it is critical. Including law enforcement on the drug court team is practiced more rarely but is clearly associated with more positive outcomes. Working on the street, law enforcement can contribute a unique perspective to the drug court team. Law enforcement can improve referrals to the program and can extend the connection of the drug court team into the community for further information gathering and monitoring of participants (e.g., in the form of home visits). This all contributes to positive outcome costs. Exploring the Key Components of Drug Courts: A Comparative Study of 18 Adult Drug Courts on Practices, Outcomes and Costs NIJ March 2008 Shannon Carey PhD

13 Law Enforcement Matters

14 What does this mean? We don t fully know the value of probation and community supervision via research. Probation varies too much. We DO know that probation helps in staffing. We DO know that field services, home visits, and non governmental hours matter.

15 What is your target? High risk, high need participants May present a lethal risk to the community (DUI) Are often different to supervise in a number of ways. Require intense supervision and sharp knowledge base that evolves constantly. Rules differ for crime categories! (DUI)

16 QUADRANT MODEL FOCUS here for public safety High Needs Low Needs High Risk Accountability Treatment Pro-social habilitation Adaptive habilitation 300 hours of tx (combo) Accountability NO TREATMENT! Pro-social habilitation (Adaptive habilitation) 150 hours of criminal thinking Low Risk Treatment (Pro-social habilitation) Adaptive habilitation 200 hours of tx (combo) Secondary prevention Diversion hours of Ed

17 Purpose of Community Supervision 1. Protect public safety 2. Provide accountability 3. Protect internal and external program integrity 4. Support the progress of the participant 5. Provide early intervention 6. As an adjunct to treatment 7. Extend the team into the community

18 What are we looking for? Threats to the recovery environment Signs of negative changes Signs of pending lapse Signs of unhealthy or good relationships Signs of LLPOF ALL CONCERNS FOR FAST INTERVENTION! AND-to catch them doing something right, and build on that.

19 But is that all? NO! Critical Goal: Catch them doing something right! Apply RNR practices Assess, assess, assess Report it to the team immediately for incentives.

20 ASAM Recovery Environment An essential part of the treatment assessment and placement criteria.(treatment gets self report) Who on the team is performing this function? How many hours per week does the team spend with clients? What are they doing the rest of the time?-somebody better know!

21 LOOK for mental health issues co-occurring-cannabis-and-nicotine-dependenceincreases-the-odds-for-dui.html Look for anxiety, trauma and depression. They will need to be treated. ASSUME they are there and rule them out, not in. ASSUME poly substance abuse.

22 Supervision: you are the eyes of treatment and the Court! Your information sharing should be ultra high speed Your communication should be constant Your application if sanctions and incentives should be rapid and effective Your activities improve outcomes and prevent disasters.

23 Supervision Issues Supervision must be proactive and aggressive in monitoring participants and as a support to recovery at a level not generally required of standard probationers Field work, home visits, searches, aggressive testing are necessary. This is not a program wish, it is a necessity

24 Common tasks at field visits Breath test, drug test Assess for dimensions of recovery-engage Search for signs of substance abuse Garbage cans, drawers, bags, cars, fridges, garages, ice chests, grocery and convenience store receipts,etc. Search for signs of driving or criminal conduct Check & mark tires, odometer readings, locks on storage areas, etc.

25 Search Full 4 th A. waiver! Social media Cell phones Computers Internet caches Texts Phone photos It is AMAZING what you will find!

26 Get on those cell phones!

27 THE WORLD CAN BE TERRIBLY CRUEL TO PERSONS IN RECOVERY: SIMPLE ERRANDS BECOME A DIFFICULT CHALLENGE.

28 Focus: what are they doing? How many hours per week are there? How many hours per week are participants in services, or under your visual supervision? What hours does the government work? What hours does addiction work? What are the hardest times for clients to remain sober?

29 What is going on at home? IS IT SAFE FOR THEM TO LIVE THERE? PHYSICALLY SAFE? RECOVERY SAFE?

30 Remember: you MUST do field work to truly know your client

31 Recovery is not for wimps! It is hard work. Folks want to quit It takes a LONG time It hurts Folks have tried to quit before They need help from you. It takes lots of time and work to change brains.

32 This means: Extreme risk for trigger based relapse Alcohol is a disinhibitor, making it a primary cause of relapse on drug of choice (NIJ-20 times more likely to relapse if you use alcohol) Remember: just because you caught them with one or two drugs, that may not really be the drug of choice

33 Your participants: Are working on other issues than you How can they use and not get caught Who knows the secrets of how to beat the system That is called addiction. It is expected at first. So they consult friends and they check the internet for advice.

34 I hid my alcohol. In my windshield washing fluid. New container, filled with vodka, some blue food coloring, and drilled a hole into my truck cab for the tube. Vodka on command. My wife bitched at me about drinking so I filled one garden hose in the back with booze and capped it off. Gardening became drinking.

35 I hid my drugs: Inside big scented candles on my table Above the hall closet door in the trim Ask your participants!

36 Basic Field Services

37 Consider this. There is great benefit in knowing that your probation officer may show up without notice at any time That your probation officer cares and wants you to succeed That your probation officer will hold you accountable and will acknowledge good work. That your probation officer might be a little crazy..

38 They might not know this is a problem... or don t want to tell you.

39 Are there public safety issues that exceed the client alone? Dependent adults? Dependent children? Family violence? Methamphetamine labs? BHO Labs? Cannabis grows?

40 Protect the public! Elder abuse Child abuse & neglect

41 Are there other problems? Signs of use Signs of tampering Signs of problems with relationships Signs of new occupants Signs of bad associates they need help with. Case management threats that require different approaches. Things they forgot to mention

42 And people you didn t know Unsafe or unsavory associates have to go. Officer safety is paramount. People, places and things that undermine recovery have to go. about

43 Engage the family or a reliable support system Learn about adolescent brain development and behavior-emotional growth tolls from the time of addiction forward. Adjust your directives and case plan to meet the capabilities of your participant. Clear, short directions work best.

44 For all clients, Identify and address an inappropriate home environment-domestic violence in home, associates or roommates who keep drugs around, sexual assault, unsafe conditions. Plan ahead for what you are going to do about gang membership, sometimes three generations deep. For some, that IS family. The team will need to assist with the development of a support system for the client which may include family, or assist with the creation of a new clean and sober community for the participant in recovery.

45 This is all about. Public safety Treatment Recovery- helping them succeed and learn Proximal and distal behaviors Detecting desired and undesired behaviors Addressing critical issues consistently Applying incentives and sanctions as close to real time as possible.

46 Information Gathering/ Continued Assessment It is critical to praise, identify positive work and report good news even faster than negative news. Part of the goal is to catch participants doing right! You may provide them the incentive to just make it one more day.

47 Field services protect the public Field services support recovery Field services provide help with refusal skills. Field services protect the integrity of the Court process.

48 Is this only about gotcha? NO! N The most important goal is to catch them doing something right!

49 Improved dimensions of recovery Increased organization Better problem solving House lighter, cleaner Fewer fuzzy dishes Evidence of structure Reduced chaos Treatment and medication compliance

50 What we should begin to see: LifeRing

51 Supervision Staff Competencies Essential for Drug Court Experienced supervision officers Most effective Maintain balanced view Apply RNR and CCP.

52 Effective Community Supervision

53 DOSAGE PRo x BATION: Rethinking the Structure of Probation Sentences Center For Effective Public Policy Prepared for the National Institute of Corrections January 2014

54 Madeline M. Carter, Principal, Center for Effective Public Policy The Honorable Richard J. Sankovitz, Milwaukee County Circuit Court Acknowledgements We are grateful to the National Institute of Corrections generally, and the following individuals in particular, for their vision and leadership around effective correctional practice: n Jim Cosby, Chief, Community Services Division, National Institute of Corrections n Greg Crawford, Correctional Program Specialist, Community Services Division, National Institute of Corrections n Lori Eville, Correctional Program Specialist, Community Services Division, National Institute of Corrections Special thanks to our esteemed colleagues who took the time to thoughtfully review and critique this monograph: n Guy Bourgon, Ph.D., Senior Researcher, Corrections Research, Public Safety Canada n Natalie Jones, Ph.D., Director of Research, Orbis Partners n Edward Latessa, Ph.D., Interim Dean and Professor, College of Education, Criminal Justice and Human Services, University of Cincinnati n Christopher T. Lowenkamp, Ph.D., Instructor, University of Missouri, Kansas City, Department of Criminal Justice and Criminology n David Robinson, Ph.D., Chief Executive Officer, Director of Assessment, Orbis Partners n Ralph Serin, Ph.D., Associate Professor, Department of Psychology, Carleton University n Kimberly Sperber, Ph.D., Chief Research Officer, Talbert House n Marilyn Van Dieten, Ph.D., C.Psych., Senior Partner, Director of Program Implementation and Development, Orbis Partners And to those who made important contributions to its development: n Kurt Bumby, Ph.D., Senior Manager, Center for Effective Public Policy n Mark Carey, President, The Carey Group n Rachelle Ramirez, Program Associate, Center for Effective Public Policy This project was supported by Cooperative Agreement No. 12CS09GKM1 awarded by the National Institute of Corrections. The National Institute of Corrections is a component of the U.S. Department of Justice. Points of view or opinions in this document are those of the authors and do not necessarily represent the official position or policies of the U.S. Department of Justice. 2014, The Center for Effective Public Policy. The National Institute of Corrections reserves the right to reproduce, publish, translate, or otherwise use, and to authorize others to publish and use, all or any part of the copyrighted material contained in this publication. 2

55 INTRODUCTION, Isaac Newton was among the first modern scientists to recognize that new discoveries depend heavily on science that is already established: If I have seen further, he wrote, it is by standing on the shoulders of giants. 1 Giant strides have been made in the fields of public administration and criminal justice by applying science to practice. Evidence-based decision making asserts that public policy and practice should be informed by the best available research and enhanced through ongoing performance measurement and evaluation. Scientific study has demonstrated that recidivism can be reduced when three key principles are followed: n The risk principle suggests that justice system interventions should be matched to offenders risk level, focusing more intensive interventions on moderate and high risk offenders. n The need principle asserts that justice system interventions should target those factors that most significantly influence criminal behavior. n The responsivity principle demonstrates that interventions are most effective when they are based on research-supported models and tailored to the unique characteristics of individual offenders. In this paper, we propose to take this knowledge one step further: to link the duration of probation supervision to the optimal amount of intervention an offender needs in order to reduce risk of reoffense. The proposed dosage model of probation suggests that the length of supervision should be determined by the number of hours of intervention necessary to reduce risk, rather than an arbitrarily (or customarily) established amount of time (e.g., 3 years, 5 years). For many offenders, the research shows that correctional intervention is analogous to treating a patient: too little intervention and the patient receives little or no benefit; too much, and the treatment is ineffective or even harmful. 2 Given this, we postulate that the length of supervision should depend on how long it takes an offender to achieve the dosage target the type and amount of intervention that research tells us he or she needs in order to maximize the potential for behavior change and that is necessary in order to minimize risk to the public rather than a fixed term of supervision. 1 Letter to Robert Hooke, 15 February 1676, quoted in The Correspondence of Isaac Newton, Volume 1, , ed. H. W. Turnbull (Cambridge: Cambridge University Press, 1959), Although the framework described in this paper is focused specifically on offenders placed on probation, the principles seem to apply similarly to offenders on parole and offenders who are incarcerated. The types of services and conditions under which dosage is provided may vary, but the underlying principles are believed to be constant. 3

56 Section I of this paper offers a review of key research about reducing an offender s risk of reoffending evidence-based approaches to reducing recidivism in our communities. Section II builds on these approaches and reviews recent research on dosage, its applicability to sentencing and community supervision practices, and its promise for reducing recidivism even further. Taken together, this research establishes a policy and practice framework upon which a new model of supervision dosage probation 3 can be constructed. Section III outlines the dosage probation model, an approach designed to build upon the existing research and advance community supervision with the goal of increasing community safety through recidivism reduction, as well as reducing the fiscal impact associated with extended periods of supervision. 3 The term dosage probation is drawn from NIC s Evidence-Based Decision Making in Local Criminal Justice Systems Initiative in Milwaukee County, Wisconsin. The term was conceived to name a demonstration project engineered by criminal justice system and community partners in Milwaukee, working with their NIC technical assistance provider. This project is featured later in this paper. 4

57 SECTION I, THE PRINCIPLES OF EFFECTIVE INTERVENTION, Over the past three decades, researchers in the U.S., Canada, and abroad have conducted studies of probationers and parolees, juveniles and adults, and programs and services of all varieties, including those administered in institutions and provided in the community. While there remains much to understand about the pathways to criminal and delinquent behavior and the strategies that will result in desistance, there is much we know now about what makes interventions effective in reducing recidivism. Who We Target for Intervention Matters: The Risk Principle, One of the key tenets of effective intervention is the risk principle. It holds that offender programming should be matched to the offender s assessed level of risk. The links between the two have been demonstrated over decades of research (Andrews & Bonta, 2010; Lowenkamp, Latessa, & Holsinger, 2006). Conversely, considerable research has shown that offering services to offenders without regard to risk level typically fails to reduce recidivism and, particularly for low risk offenders, may result in increased recidivism (see, e.g., Andrews & Bonta, 2010; Lowenkamp, Latessa, & Holsinger, 2006; Lowenkamp, Latessa, & Smith, 2006; Lowenkamp, Pealer, Smith, & Latessa, 2006). Lowenkamp, Latessa, and Holsinger, in their 2006 meta-analysis of nearly 100 correctional programs, found compelling evidence of the importance of the risk principle. The study sample was comprised of 13,676 offenders who had been placed in halfway houses, in community correctional facilities, and under probation or parole supervision. For the purposes of the analyses, the offenders were categorized in two groups, low/low moderate risk and moderate/higher risk. 4 The researchers found that recidivism was lower among those placed in facilities and programming that adhered to the risk principle and, conversely, that recidivism increased when programs and services were delivered without regard to risk. A key to putting this research to work is the ability to reliably assess and classify offenders according to the risk of reoffense they present. Researchers have studied whether offender risk to reoffend can reliably be predicted. It can be. Research demonstrates that risk to reoffend is most accurately assessed when structured, empirically based tools are used in combination with professional judgment. Furthermore, research has conclusively demonstrated that matching the level of intervention to offenders 4 For purposes of this paper, offenders who present the highest risk of reoffense are excluded. In the literature, the term high risk offender typically does not refer to that small portion of offenders who are better described as extremely high risk. The principles and practices suggested in this paper are of questionable efficacy with this sub-population. assessed level of risk produces the greatest reductions in recidivism, with minimal intervention provided to low risk offenders and greater intervention provided to moderate and high risk offenders. 5

58 Studies demonstrate that structured, research-grounded risk assessment tools in combination with professional judgment are reliable, in contrast to unstructured assessment methods and/or professional judgment alone (Campbell, French, & Gendreau, 2009; Grove, Zald, Lebow, Snitz, & Nelson, 2000; Hanson & Morton-Bourgon, 2004). 5 Intervention programs that employ tools such as these are more effective in reducing recidivism than those that do not (see, e.g., Lowenkamp, Latessa, & Smith, 2006). What We Target for Intervention Matters: The Need Principle, A second cornerstone of effective correctional intervention is the need principle. Research demonstrates that although offenders typically have many needs, some of them result in criminal behavior but others do not. These traits are referred to as criminogenic needs and represent the changeable, crime-influencing risk factors that must be the targets of risk reduction efforts (Andrews & Bonta, 2010). The criminogenic needs that most strongly predict recidivism are antisocial cognition (thoughts and beliefs) that support antisocial behavior; antisocial temperament, which is often characterized by poor decision making skills, anger management difficulties, and impulse control deficits; and antisocial associates (see Andrews & Bonta, 2010; Gendreau, Little, & Goggin, 1996). Other dynamic risk factors that influence crime, albeit to a lesser degree, include family/marital stress, substance abuse, employment instability, educational attainment and engagement difficulties, and lack of prosocial leisure activities. Offenders present other needs, such as low self-esteem, depression, anxiety, or general health concerns, but research has not demonstrated a clear link between these factors and recidivism (Gendreau et al., 1996). As such, they are considered non-criminogenic 6 and, according to the need principle, should not be the emphasis of correctional interventions. Research shows that interventions that target criminogenic rather than non-criminogenic needs consistently lead to superior outcomes (Andrews & Bonta, 2010; Gendreau, French, & Taylor, 2002; Gendreau & Goggin, 1996; Lowenkamp, Latessa, & Smith, 2006). Furthermore, targeting a greater 5 It is important to note that, in the field, risk to reoffend can be considered a relative term. To be clear, in our discussion here we intend that likelihood to reoffend is determined using structured, empirically based assessment methods; that quality control mechanisms ensure accurate assessment results; and that stated risk levels are not adjusted as the result of overrides or policydriven cut points. 6 While these risk factors are well-established for the general population of male offenders, studies identify additional and/or other factors or contexts for females and some sub-populations of offenders (e.g., sex offenders). Detailing the risk factors associated with sub-groups of offenders is beyond the scope of this paper. The distinction around risk factors is not intended to imply that the dosage model would be inappropriate for sub-groups of offenders. 6

59 number of criminogenic needs (e.g., 3 4 more criminogenic than non-criminogenic needs) results in more substantial recidivism reductions as much as 30 percent lower on average than is achieved when targeting fewer criminogenic needs (e.g., 1 2 more criminogenic than non-criminogenic needs) (Gendreau et al., 2002). These principles apply to both treatment programs and to interventions by probation officers themselves (Bonta, Rugge, Scott, Bourgon, & Yessine, 2008; Lowenkamp, Flores, Holsinger, Makarios, & Latessa, 2010; Lowenkamp, Pealer, et al., 2006). When an officer uses face-to-face time with an offender to address criminogenic needs, better outcomes are achieved, including reduced recidivism (Bonta et al., 2008, 2011; Robinson et al., 2012). Thus, it is not surprising that when officers target non-criminogenic needs and spend more time monitoring conditions of supervision during their contacts with offenders, reductions in recidivism rates are not positively affected (Bonta et al., 2008). Studies have demonstrated the factors that are correlated to recidivism and isolated those that have the greatest influence on future criminal behavior. Research has also demonstrated that when these criminogenic needs are effectively targeted with moderate and high risk offenders, recidivism is reduced, particularly when multiple needs are addressed. These results hold steady whether the agent of change is a program, service, or intervention by a corrections professional such as a probation or parole officer. How We Intervene and Interact Matters: The Responsivity Principle, The effectiveness of interventions also depends on delivering them in ways that are most likely to engage offenders and facilitate meaningful change, and by matching the right program to the offender and his or her individual traits. This is known as the responsivity principle, which is comprised of two aspects: general and specific responsivity (see, e.g., Andrews & Bonta, 2010). n General responsivity refers to the fact that some types of programming are generally more effective than others. In particular, offenders respond better to cognitive behavioral strategies than to other kinds of programming (see Andrews & Bonta, 2010; Lipsey, Landenberger, & Wilson, 2007). Cognitive behavioral strategies assist offenders with changing harmful thinking patterns and attitudes, as well as developing prosocial skills. Studies show that well-implemented cognitive behavioral interventions can reduce recidivism by as much as 30 percent on average, particularly with moderate to high risk offenders (Andrews & Bonta, 2010; Lipsey et al., 2007; Lowenkamp et al., 2010; Lowenkamp, Latessa, & Holsinger, 2006). Other strategies are not as effective, in particular traditional surveillance- and enforcementoriented approaches to supervision, designed primarily around imposing, monitoring, and enforcing conditions of supervision, and sanctioning noncompliance (see, e.g., Bonta et al., 2008; Cullen & Gendreau, 2000; Drake, Aos, & Miller, 2009; Lowenkamp et al., 2010). 7

60 Multiple meta-analyses demonstrate that such strategies fail to reduce recidivism and, in some instances, are associated with increases in recidivism (see Drake et al., 2009; Gendreau, Goggin, Cullen, & Andrews, 2000; Lowenkamp et al., 2010; Lowenkamp, Latessa, & Holsinger, 2006; Lowenkamp, Pealer, et al., 2006). For instance, in one meta-analysis, intensive supervision, electronic monitoring, and day reporting were found to be significantly less effective in reducing recidivism than cognitive behavioral strategies (Lowenkamp, Latessa, & Smith, 2006). Far more effective are blended or balanced supervision frameworks in which risk management and accountability functions (e.g., monitoring and enforcement) are paired with rehabilitative or risk-reducing strategies (Bonta et al., 2008, 2011; Bourgon, Gutierrez, & Ashton, 2011; Drake et al., 2009; Gendreau et al., 2000; Lowenkamp et al., 2010; Lowenkamp, Latessa, & Smith, 2006; Taxman, Yancey, & Billanin, 2006). In addition, as will be explored further, an important aspect of general responsivity is the nature and degree of professional alliance between the offender and the professionals with whom they work. n Specific responsivity speaks to the individual characteristics of the offender (e.g., gender, culture, cognitive or developmental functioning, motivation to change, etc.). This aspect of the responsivity principle recognizes the diversity of the offender population and that individual differences among offenders influence how they engage, internalize, and respond to interventions. It suggests the The research underlying the responsivity principle ineffectiveness of a one size fits all approach. grounds correctional interventions in models and specific techniques that are shown to reduce General and specific responsivity are best addressed through recidivism. It highlights the powerful influence of a constellation of evidence-based elements referred to as corrections professionals in addition to service core correctional practices (Andrews & Bonta, 2010; Dowden providers as change agents and identifies the & Andrews, 2004). These risk-reducing strategies include, strategies and approaches that are most effective for example, developing a strong professional alliance with in reducing offender recidivism. offenders; modeling and reinforcing prosocial attitudes and behaviors; creating opportunities to teach concrete skills such as problem solving, impulse control, and anger management; allowing for practice and rehearsal of newly learned skills; using reinforcers and responses to noncompliant behavior effectively; and providing advocacy and brokerage to support offenders participation in needed programs and services. Implementing core correctional practices has been shown to reduce recidivism (Andrews & Bonta, 2010; Bonta et al., 2008, 2011; Dowden & Andrews, 2004; Robinson et al., 2012). Research highlights the pivotal role probation officers in particular can play in steering offenders away from illegal behavior (Bonta et al., 2008, 2011; Kennealy, Skeem, Manchak, & Eno Louden, 2012; Paparozzi & Gendreau, 2005; Robinson et al., 2012; Skeem, Eno Louden, Polaschek, & Camp, 2007). Worth noting is a recent study indicating that offenders who perceived their corrections professionals to have a balanced orientation (i.e., trusting, caring, fair, and authoritative) 8

61 reoffended at lower rates (Kennealy et al., 2012). This is consistent with earlier research identifying better outcomes among offenders who are supervised by officers who see themselves as taking a balanced approach to supervision (Paparozzi & Gendreau, 2005). As is described above, adherence to any one of the three intervention principles risk, needs, and responsivity yields better results for probation supervision. When supervision strategies are designed to incorporate all three, even better results are achieved (Andrews & Bonta, 2010; Lowenkamp et al., 2010; Lowenkamp, Latessa, & Smith, 2006). Research demonstrates that the approach and skills of corrections professionals can influence recidivism outcomes. Offenders on the caseloads of corrections professionals who successfully build professional alliance, focus on criminogenic needs, work with offenders to identify and address skill deficits, and use rewards and responses to noncompliant behavior effectively are less likely to engage in future criminal behavior. How Well Interventions Are Implemented Matters, Unfortunately, studies of adherence to these principles and practices reveal that the fidelity of implementation is less than optimal (see Gendreau, Goggin, & Smith, 1999; Lowenkamp et al., 2010; Lowenkamp, Latessa, & Smith, 2006). Fidelity and Integrity of Correctional Programs and Services, Lowenkamp, Pealer, and colleagues (2006) conducted a meta-analysis of 66 supervision-based programs (including day reporting, electronic monitoring, intensive supervision, work release, and some treatment interventions such as substance abuse, domestic violence treatment, and residential programming) designed for diverting offenders from jail or prison. The authors examined the degree to which the programs were faithful to the risk and needs principles. They found the following: n A majority of the programs failed to reduce recidivism (and, indeed, slightly increased it). n Those programs adhering to the risk and need principles achieved lower recidivism rates. n When supervision length was a function of risk, recidivism reductions resulted; likewise, no riskreducing effect was found for supervision lengths that were not in accordance with the risk principle. n Recidivism rates were lower when higher risk offenders were referred to more treatment programs. n When more referrals were made to risk-reducing programs (i.e., interventions that targeted criminogenic needs) rather than to programs addressing non-criminogenic needs, recidivism decreased. n Increases in recidivism were observed for offenders who were referred to more non-criminogenic than criminogenic interventions. 9

62 The analyses also indicated a linear association between recidivism rates and the cumulative adherence to four measures of the risk and need principles. These measures were (1) prioritization of higher risk offenders, (2) longer supervision periods for higher risk offenders, (3) more treatment referrals for higher risk offenders, and (4) higher density of referrals for criminogenic than noncriminogenic needs. Programs adhering to 3 4 measures achieved significant recidivism reductions; those adhering to 1 2 measures achieved minimal reductions; and for programs that adhered to none of the measures of the risk/need principles, increases in recidivism were found. Fidelity and Integrity of Corrections Professionals Interventions, While some of the research focuses on program design, other research focuses on the role corrections professionals can play in implementation. Five studies (or sets of studies) are worth noting. In 2010, Lowenkamp and colleagues published a meta-analysis of 58 intensive supervision programs with over 11,000 offenders, exploring the extent to which supervision philosophy (i.e., human service-oriented versus deterrence-oriented) and program integrity (i.e., alignment with the evidence-based principles of correctional intervention) were related to outcomes. The researchers concluded that, n supervision was most effective in reducing recidivism when it followed a balanced model and when the program aligned well with the principles of effective correctional intervention; and, n recidivism increased for supervision programs that were deterrence-oriented and that did not adhere well to intervention principles. To gain an understanding of the extent to which probation officers contacts with offenders aligned with the principles of effective correctional intervention and the core correctional practices, researchers reviewed the risk/needs assessments, case files, and audiotaped sessions of 62 probation officers (Bonta et al., 2008). Findings included the following: n Consistent with the risk principle, officers generally had more contacts/sessions with higher risk offenders than lower risk offenders, although the frequency of these contacts may have been insufficient for optimizing risk-reducing interventions with high risk offenders. n Criminogenic needs were identified through assessments, and indicators of these needs surfaced during the course of the sessions, although few of these needs were acknowledged or addressed during these contacts, nor were core correctional practices utilized routinely. 10

63 n When criminogenic needs were targeted in sessions, the density was high (i.e., multiple criminogenic needs were targeted), making it difficult to adequately address them and resulting in less productive sessions. Paradoxically, under these circumstances, the more criminogenic needs that officers attempted to address within a single session, the higher the recidivism rate. 7 n Many officers devoted more time during sessions to addressing conditions and compliancerelated matters than to criminogenic needs. When this was the case, higher recidivism rates were identified. In an effort to understand the link between caseload size, the presence or absence of evidencebased practices, core correctional practices, and recidivism, a multi-site study (Jalbert et al., 2011) was conducted, with the following results: n Officers with smaller caseloads made more frequent contacts with offenders. n Officers had more time to spend during those contacts. n Officers were more likely to utilize effective correctional interventions. n Offenders supervised by officers with smaller caseloads had lower recidivism rates, if the supervision agency had implemented evidence-based practices. n When probationers were supervised on reduced caseloads in agencies in which evidence-based practices had not been implemented, recidivism rates were higher. A fourth group of studies focused on training programs developed to incorporate evidence-based principles and core correctional practices into the day-to-day efforts of supervision officers, particularly in their face-to-face contacts with offenders. Such training initiatives are designed not only to promote skill acquisition on the part of officers, but also to sustain these skills, and thus fidelity of implementation over time, through coaching, supervision, and mentoring. Promising examples include the Effective Practices in Community Supervision (EPICS), Strategic Training Initiative in Community Supervision (STICS), and Staff Training Aimed at Reducing Re-arrest (STARR) models (see, e.g., Bonta et al., 2011; Lowenkamp, Holsinger, Robinson, & Alexander, 2012; Robinson et al., 2012; Smith, Schweitzer, Labreque, & Latessa, 2012). These approaches have proven effective: as a result of officers focusing more on criminogenic needs and skill building, supervision failure and recidivism rates were significantly lower (Bonta et al., 2011; Lowenkamp et al., 2012; Robinson et al., 2012). 7 This study determined that 67% of probation officers dealt with an average of 5.2 criminogenic needs during a supervision session. Researchers determined that the more topics that were covered during the session, the higher the recidivism rate. 11

64 Studies have demonstrated that while the principles of effective intervention are key to risk reduction, attention must be paid to implementation and fidelity in order for recidivism reduction potential to be realized. Furthermore, researchers have shown that corrections professionals, including but not limited to probation and parole officers, can have a significant impact on offender outcomes; however, skills training and coaching are critical to their success in this regard. Finally, Bonta and colleagues (Bonta et al., 2008, 2011) have demonstrated a link between recidivism and the amount of time officers spend face-to-face with an offender. They found that recidivism rates among officers who spent minutes with offenders per session were lower than the recidivism rates of officers who spent less than 16 minutes. The Relationship Between Early Termination of Supervision and Recidivism, In 2009, the Administrative Office of the Courts, an agency of the U.S. Federal Courts, conducted an initial study of the impact of early termination of supervision among federal probationers. This study demonstrated that such practice, when limited to appropriate cases, did not adversely affect public safety. An expanded study was subsequently conducted using a larger sample and a matched-sample design. Subjects in the early termination and the full-term groups were followed for three years after release, and recidivism was measured on the basis of arrests for new crimes. Although the subjects of the study were predominately low risk offenders, moderate and high risk offenders were represented as well. Researchers determined that the offenders in the early termination groups, regardless of risk level, had lower rates of recidivism than their full-term counterparts (Baber & Johnson, 2013). 12

65 SECTION II, ADDING DOSAGE TO THE EQUATION, How Much Dosage We Deliver Matters, In the health care field, determining the appropriate dosage is an empirical venture: n Conduct an assessment to identify the extent and nature of a presenting concern, including its root causes and the patient s unique characteristics. n Identify the range of potential interventions with demonstrated effectiveness in producing positive outcomes. n Determine a course of intervention, including the optimal amount, frequency, and duration of the intervention. Research in the corrections field, and in particular research concerning intervention principles, suggests that a similar approach can be taken to determining the type and amount of intervention an offender should receive to minimize recidivism and increase public safety the dosage. Studies examining differential dosage are quite limited, but generally support this concept (see Sperber, Latessa, & Makarios, 2013a, 2013b). For example: n Gendreau and Goggin s (1996) post-hoc analysis of the effectiveness of correctional interventions revealed that programs of 3 4 months in duration were associated with better outcomes than shorter programs. n In a meta-analysis of 200 juvenile programs, effectiveness was linked to duration, with programs that lasted a minimum of 6 months yielding larger effect sizes than those of shorter length. The findings also revealed that roughly 100 hours was needed to reduce recidivism (Lipsey, 1999). n A meta-analysis of more than 40 cognitive behavioral programs revealed that effectiveness was greater for programs that targeted higher risk offenders who also received greater frequency and total hours of programming (Lipsey et al., 2007). n Lowenkamp, Latessa, & Holsinger s meta-analysis (2006) revealed that simply providing the proper model of programming (i.e., cognitive behavioral) was not sufficient to maximize risk reduction. Rather, effectiveness was enhanced by differential dosage more units of service or referrals to risk-reducing programs and longer duration of interventions. The researchers found that this approach was more effective for higher risk offenders than for lower risk offenders receiving the same dosage. n An empirical examination involving over 600 adults in a prison setting (Bourgon & Armstrong, 2008) concluded that, for moderate risk offenders, 100 programming hours was sufficient, whereas moderate/high risk offenders required 200 treatment hours, and high risk/high need offenders may require more than 300 hours. 13

66 n In yet another study, the relationship between dosage and recidivism was explored with a sample of 69 parolees under supervision who had dropped out of a community-based treatment program targeting antisocial attitudes (Kroner & Takahashi, 2012). Dosage was measured not only in terms of number of sessions prior to the drop-out point, but also with respect to previously completed prison-based programs. Current dosage was predictive of recidivism, but the number of prior program completions was not. In addition to providing further confirmation of the importance of differential dosage, the findings illustrate the importance of engaging and retaining offenders in order to prevent drop-out. n Most recently, the effectiveness of differential dosage was examined among a sample of nearly 700 adult offenders discharged from a community-based correctional facility who were under supervision (Sperber et al., 2013b). Generally speaking, greater treatment dosages were associated with reductions in recidivism across risk levels, and were most pronounced with high risk offenders: high risk offenders receiving high dosage (200 or more hours) compared to those receiving a moderate dosage ( hours) recidivated at markedly lower rates. Further Study Needed, Further evaluation of dosage and its potential is needed. For instance, some studies conducted thus far suggest that the relationship between dosage and effectiveness may not be fully linear; that is, it appears that there may be a point of diminishing returns, treatment fatigue, or dilution of program effectiveness (Kroner & Takahashi, 2012). Similarly, research demonstrates that attempting to address multiple criminogenic needs in a single encounter reduces the effectiveness of interventions (Bonta et al., 2008; Lowenkamp et al., 2012). These findings provoke questions about the specific circumstances under which the effectiveness of interventions are mitigated. Additionally, while it is clear that dosage matters, there is less than optimal empirical guidance about how much dosage is desirable and, in particular, what kinds of interactions with offenders count and how to measure dosage (Kroner & Takahashi, 2012; Sperber et al., 2013a, 2013b). If an offender attends a 90-minute treatment session but is disengaged, does this count? (Probably not.) If the offender practices a skill at home with a parent, does that count? (Possibly.) Given research showing the effectiveness of focusing interventions around the highest-value criminogenic needs, does this suggest that time devoted discussing the offender s employment, for instance, doesn t count or simply counts less? These and other areas of inquiry will provide additional guidance in determining the structure of interventions of the future. 14

67 Although further research is clearly warranted given the limited number of studies conducted to date that are specific to dosage and recidivism, the following reflects a conceptual model to guide risk-based interventions: Dosage Conceptual Model Illustration Risk Level Dosage Target Likely Duration Dosage Hours Delivered by Corrections Professional Dosage Hours Delivered through Referral Services Moderate risk 100 hours 12 months supervision (52 weeks) with 12 months services (52 weeks) 45 minutes/ 2 weeks for 12 months Total hours: minutes/week for 12 months Total hours: 78 Moderate/ high risk 200 hours 18 months supervision (78 weeks) with 15 months services (65 weeks) 45 minutes/week for 12 months + 45 minutes/2 weeks for 6 months Total hours: 49 3 hours/week for 9 months + 90 minutes/week for 6 months Total hours: 156 High risk 300 hours 24 months supervision (104 weeks) with 18 months services (78 weeks) 45 minutes/week for 24 months Total hours: 78 6 hours/week or 24 hours/4 weeks for 6 months + 90 minutes/week or 6 hours/4 weeks for 12 months Total hours:

68 SECTION III, IMPLICATIONS: THE DOSAGE PROBATION MODEL OF SUPERVISION, Although the subject warrants deeper study, there appears to be sufficient grounding for further testing and perhaps expansion of the application of dosage to justice system practices (i.e., the dosage probation model). The following summarizes the relevant research to date: n Applying evidence-based principles and practices (i.e., risk, need, and responsivity) with fidelity reduces recidivism (Bonta et al., 2011; Lowenkamp et al., 2012; Lowenkamp, Latessa, & Smith, 2006; Robinson et al., 2012). n Corrections professionals face-to-face contacts with offenders can be an effective intervention and, as such, corrections professionals play a key role as agents of change (Bonta et al., 2008, 2011; Robinson et al., 2012). Their risk-reducing interventions complement those provided by others (e.g., treatment providers) and, as such, it is reasonable to consider their interventions as contributing to the minimum dosage necessary to reduce recidivism. Despite the lack of a standard operating definition of dosage, a growing body of evidence indicates that dosage considerations are important to maximizing outcomes and reducing recidivism with correctional populations, particularly for moderate and high risk offenders (see, e.g., Bourgon & Armstrong, 2005; Kroner & Takahashi, 2012; Sperber et al., 2013b). These findings suggest that officers practices during the course of supervision can reasonably contribute toward the minimum dosage requirements needed for recidivism reduction, and that a probation model based on the risk, need, and responsivity principles has the potential to enhance risk-reduction efforts. Taking together the research summarized in this paper, the primary elements of a dosage probation model emerge: n Research-based, structured assessments are conducted to reliably differentiate higher from lower risk offenders. n Sentencing, supervision, correctional programming, reentry, and violation decisions are informed by assessed level of risk, criminogenic needs, and optimal dosage. n Probation completion is linked to achievement of a dosage target rather than a fixed period of time, thereby incentivizing offenders engagement in risk-reducing interventions. n Probation terms and conditions emphasize risk-reducing interventions that target criminogenic needs. n Officers and offenders collaborate to develop case management plans; interventions are designed to address the most influential criminogenic needs; dosage targets are set. n Offenders are referred to programs and services that demonstrate the capacity to effectively address their needs, thereby incentivizing service providers to deliver evidence-based programs. 16

69 n The amount of dosage received is tabulated over time and objective behavioral measures are used to gauge change. n Probation officers are trained in core correctional practices; they are provided with ongoing coaching; and caseloads and workloads are right-sized so that officers have sufficient time to meaningfully engage offenders face to face. n Quality assurance and continuous quality improvement strategies are implemented to ensure the integrity of these evidence-based practices. n For those who meet their dosage target and who achieve objective behavioral indicators, probation is terminated, as opposed to terminating supervision at some point further down the road when supervision time runs out. Conclusion, According to the U.S. Department of Justice, Bureau of Justice Statistics, 67% of individuals released from prison are rearrested within three years of discharge. An estimated 30% of adult probationers supervised in the community are reconvicted for a new crime. Despite changes in laws, sentencing practices, and intervention approaches, these recidivism rates have remained relatively stable for decades (Andrews & Bonta, 2010; Bureau of Justice Statistics, n.d.; Hughes & Wilson, 2005). However, research over the past two decades demonstrates that significant reductions in recidivism are possible if current knowledge is applied with fidelity (see Andrews & Bonta, 2010; Lowenkamp, Latessa, & Smith, 2006). No longer is the challenge in understanding what we need to do to positively influence offender behavior; instead, the challenge is doing it. Practically speaking, adopting an evidence-based approach means restructuring the way in which we do business in our jails and prisons, in probation and parole, and among judges, prosecutors, and others so that organizational structures and cultures enable, rather than hinder, the implementation of policies, practices, programs, and services that are known to work in reducing criminal behavior. Dosage probation takes our current knowledge of intervention principles to the next logical step. Implementation of the dosage probation model, coupled with a rigorous empirical evaluation, offers potential for a justice system double play: increasing public safety while decreasing the costs associated with offenders persistent cycle of crime. 17

70 CASE STUDY, Dosage Probation in Milwaukee County, Wisconsin, Introduction, Sometimes the principal goal of correctional supervision is to contain a known risk over a given period of time. However, in most cases, the principal goal of supervision is risk reduction. Yet the measurement of successful completion of supervision usually is tied to the offender s ability to remain trouble-free for the duration of the supervision period rather than demonstration that risk has actually been reduced. If the primary objective of the criminal justice intervention is to accomplish risk reduction, then the termination of supervision should be tied to the achievement of that goal, not merely the passage of a given length of time. Project Purpose, Through its work in the Evidence-Based Decision Making in Local Criminal Justice Systems Initiative, a multidisciplinary team of criminal justice stakeholders in Milwaukee County (i.e., courts, corrections, prosecutors, defense attorneys, probation, and treatment) has designed and is empirically testing the first dosage probation experiment. Probationers will be sentenced to a period of dosage probation, with the opportunity to earn early termination from supervision by accomplishing riskreducing objectives. Project Goals, The project seeks to determine if the dosage probation model will n lower recidivism among the target population; n reduce the average length of supervision for those who successfully complete supervision; and n align sentences with risk assessment information; align the probation agent s supervision and intervention practices, as well as community-based treatment, with needs information; and align probationer incentives with the achievement of risk reduction goals. Study Design, n Approximately 600 medium and high risk offenders will be determined to be eligible for dosage probation. n Offenders will volunteer to participate in the study. n Eligible, voluntary offenders will be randomly assigned to either the control unit (business as usual) or the dosage probation unit. n Agents in the dosage unit will be trained and coached in the principles of effective interventions and core correctional practices, and provided tools to engage offenders in behavior-changing activities. n Independent evaluators will determine the impact of dosage probation on short-term measures (e.g., violations and revocations) and long-term measures (i.e., post-supervision recidivism). Funding for the evaluation is provided under a grant from the U.S. Department of Justice, Bureau of Justice Assistance. 18

71 REFERENCES, Andrews, D. A., & Bonta, J. (2010). The psychology of criminal conduct (5th ed.). New Providence, NJ: LexisNexis Matthew Bender. Baber, L. M., & Johnson, J. L. (2013). Early termination of supervision: No compromise to community safety. Federal Probation, 77(2). Retrieved from Fedprob/ /no-compromise.html Bonta, J., Bourgon, G., Rugge, T., Scott, T.-L., Yessine, A. K., & Gutierrez, L. (2011). An experimental demonstration of training probation officers in evidence-based community supervision. Criminal Justice and Behavior, 38, Bonta, J., Rugge, T., Scott, T.-L., Bourgon, G., & Yessine, A. K. (2008). Exploring the black box of community supervision. Journal of Offender Rehabilitation, 47, Bourgon, G., & Armstrong, B. (2005). Transferring the principles of effective treatment into a "real world" prison setting. Criminal Justice and Behavior, 32, Bourgon, G., Gutierrez, L., & Ashton, J. (2011). The evolution of community supervision practice: The transformation from case manager to change agent. Irish Probation Journal, 8, Bureau of Justice Statistics. (n.d.). Adults on probation, in jail or in prison, and on parole, United States, Sourcebook of Criminal Justice Statistics Online. Retrieved from sourcebook/pdf/t pdf Campbell, M. A., French, S., & Gendreau, P. (2009). The prediction of violence in adult offenders: A meta-analytic comparison of instruments and methods of assessment. Criminal Justice, 36, Cullen, F. T., & Gendreau, P. (2000). Assessing correctional rehabilitation: Policy, practice, and prospects. Criminal Justice, 3, Dowden, C., & Andrews, D. A. (2004). The importance of staff practice in delivering effective correctional treatment: A meta-analytic review of core correctional practice. International Journal of Offender Therapy and Comparative Criminology, 48, Drake, E. K., Aos, S., & Miller, M. G. (2009). Evidence-based public policy options to reduce crime and criminal justice costs: Implications in Washington State. Victims and Offenders, 4, Gendreau, P., French, S., & Taylor, A. (2002). What works (what doesn t) revised 2002: The principles of effective correctional treatment. Unpublished manuscript, University of New Brunswick, Saint John, Canada. Gendreau, P., & Goggin, C. (1996). Principles of effective programming with offenders. Forum on Corrections Research, 8(3), Gendreau, P., Goggin, C., Cullen, F. T., & Andrews, D. A. (2000). The effects of community sanctions and incarceration on recidivism. Forum on Corrections Research, 12(2),

72 Gendreau, P., Goggin, C., & Smith, P. (1999). The forgotten issue in effective correctional treatment: Program implementation. International Journal of Offender Therapy and Comparative Criminology, 43, Gendreau, P., Little, T., & Goggin, C. (1996). A meta-analysis of the predictors of adult offender recidivism: What works! Criminology, 34, Grove, W. M., Zald, D. H., Lebow, S., Snitz, B. E., & Nelson, C. (2000). Clinical versus mechanical prediction: A meta-analysis. Psychological Assessment, 12, Hanson, R. K., & Morton-Bourgon, K. (2004). Predictors of sexual recidivism: An updated meta-analysis (User Report ). Ottawa, Canada: Public Safety and Emergency Preparedness Canada. Hughes, T., & Wilson, D. J. (2002). Reentry trends in the United States. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics. Jalbert, S. K., Rhodes, W., Kane, M., Clawson, E., Bogue, B., Flygare, C., Kling, R., & Guevara, M. (2011). A multi-site evaluation of reduced probation caseload size in an evidence-based program setting. Washington, DC: U.S. Department of Justice. Kennealy, P. J., Skeem, J. L., Manchak, S. M., & Eno Louden, J. (2012). Offender-officer relationships matter: Firm, fair and caring officer-offender relationships protect against supervision failure. Law and Human Behavior, 36, Kroner, D. G., & Takahashi, M. (2012). Every session counts: The differential impact on previous programmes and current programme dosage on offender recidivism. Legal and Criminological Psychology, 17, Lipsey, M. W. (1999). Can intervention rehabilitate serious delinquents? The ANNALS of the American Academy of Political and Social Science, 564, Lipsey, M. W., Landenberger, N. A., & Wilson, S. J. (2007). Effects of cognitive-behavioral programs for offenders. Campbell Systematic Reviews, 6, Lowenkamp, C. T., Flores, A. W., Holsinger, A. M., Makarios, M. D., & Latessa, E. J. (2010). Intensive supervision programs: Does program philosophy and the principles of effective intervention matter? Journal of Criminal Justice, 38, Lowenkamp, C. T., Holsinger, A., Robinson, C. R., & Alexander, M. (2012). Diminishing or durable treatment effects of STARR? A research note on 24-month re-arrest rates. Journal of Crime and Justice. doi: / X Lowenkamp, C. T., Latessa, E. J., & Holsinger, A. (2006). The risk principle in action: What have we learned from 13,676 cases and 97 correctional programs? Crime & Delinquency, 52, Lowenkamp, C. T., Latessa, E. J., & Smith, P. (2006). Does correctional program quality really matter? The impact of adhering to the principles of effective intervention. Criminology & Public Policy, 5,

73 Lowenkamp, C. T., Pealer, J., Smith, P., & Latessa, E. J. (2006). Adhering to the risk and need principles: Does it matter for supervision-based programs? Federal Probation, 70, 3 8. Paparozzi, M. A., & Gendreau, P. (2005). An intensive supervision program that worked: Service delivery, professional orientation, and organizational supportiveness. The Prison Journal, 85, Robinson, C. R., Lowenkamp, C. T., Holsinger, A. M., VanBenschoten, S., Alexander, M., & Oleson, J. C. (2012). A random study of staff training aimed at reducing re-arrest (STARR): Using core correctional practices in probation interactions. Journal of Crime and Justice, 35, Skeem, J. L., Eno Louden, J., Polaschek, D., & Camp, D. (2007). Assessing relationship quality in mandated community treatment: Blending care with control. Psychological Assessment, 19, Smith, P., Schweitzer, M., Labreque, R. M., & Latessa, E. J. (2012). Improving probation officers' supervision skills: An evaluation of the EPICS model. Journal of Crime and Justice, 35, Sperber, K. G., Latessa, E. J., & Makarios, M. D. (2013a). Establishing a risk-dosage research agenda: Implications for policy and practice. Justice Research and Policy, 15(1), Sperber, K. G., Latessa, E. J., & Makarios, M. D. (2013b). Examining the interaction between level of risk and dosage of treatment. Criminal Justice and Behavior, 40, Taxman, F. S., Yancey, C., & Bilanin, J. E. (2006). Proactive community supervision in Maryland: Changing offender outcomes. College Park, MD: University of Maryland, Bureau of Governmental Research. 21

74 22

75 IMPAIRED DRIVING RISK ASSESSMENT A PRIMER FOR PRACTITIONERS T R A F F I C I N J U R Y R E S E A R C H F O U N D A T I O N

76 The Traffic Injury Research Foundation The mission of the Traffic Injury Research Foundation (TIRF) is to reduce traffic-related deaths and injuries. TIRF is an independent, charitable road safety research institute. Since its inception in 1964, TIRF has become internationally recognized for its accomplishments in identifying the causes of road crashes and developing program and policies to effectively address them. This primer was contracted with the Traffic Injury Research Foundation (TIRF) and funded by the Canadian Institutes of Health Research (CIHR) Team in Transdisciplinary Studies in Driving While Impaired Onset, Persistence, Prevention, and Treatment. Traffic Injury Research Foundation 171 Nepean Street, Suite 200 Ottawa, Ontario K2P 0B4 Ph: (613) Fax: (613) Website: January 2014 Traffic Injury Research Foundation Copyright 2014 ISBN:

77 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS Robyn D. Robertson / Katherine M. Wood / Erin A. Holmes Traffic Injury Research Foundation

78 ACKNOWLEDGMENTS TIRF gratefully acknowledges the reviewers of this report who graciously assisted in its development and who provided feedback and comments on earlier drafts. Their knowledge and perspectives were instructive and enabled us to create a user-friendly and useful report that can benefit criminal justice, transportation and health practitioners across the country. Thomas G. Brown, Ph.D. Director, Addiction Research Program Pavilion Foster / Douglas Hospital Research Centre Assistant Professor Department of Psychiatry, McGill University Faculty of Medicine Wendy Schilling Program Developer, Alberta Impaired Drivers Program Alberta Motor Association Craig Staniforth Area Manager, Youth Addiction Services, Edmonton Impact Supervisor, Alberta Impaired Drivers Program Jane C. Maxwell, Ph.D. Senior Research Scientist Addiction Research Institute Center for Social Work Research The University of Texas at Austin David S. Timken, Ph.D. Director Center for Impaired Driving Research and Evaluation Boulder, Colorado Grant Duwe, Ph.D. Director, Research and Evaluation Minnesota Department of Corrections Kenneth W. Wanberg, Ph.D. Director Center for Addiction Research and Evaluation Arvada, Colorado ACKNOWLEDGEMENTS III

79 TABLE OF CONTENTS ACKNOWLEDGEMENTS EXECUTIVE SUMMARY 1. BACKGROUND 1 2. INTRODUCTION 4 3. PURPOSE AND OBJECTIVES 6 4. PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS Male Impaired Drivers Female Impaired Drivers Summary of Characteristics IMPAIRED DRIVING RISK FACTORS Male Risk Factors Female Risk Factors Summary Some Reflections on Estimating Impaired Driving Recidivism RISK ASSESSMENT INSTRUMENTS Alcohol Dependence Scale (ADS) Adult Substance Use and Driving Survey Revised (ASUDS-R) Alcohol Severity Index (ASI) Alcohol Use Disorder Identification Test (AUDIT) Inventory of Drug-Taking Situations (IDTS) Drug Abuse Screening Test (DAST) Level of Service Inventory-Revised (LSI-R) Michigan Alcoholism Screening Test (MAST) Substance Abuse Subtle Screening Inventory (SASSI) Research Institute on Addiction Self Inventory (RIASI) Biomarkers Summary TREATMENT INTERVENTIONS Screening and Brief Interventions (SBI) Motivational Interviewing (MI) 91 III VII TABLE OF CONTENTS V

80 7.3 Cognitive Behavioural Therapy (CBT) Pharmacotherapies Internet-based Brief Interventions Summary BEST PRACTICES FOR TREATMENT AND REHABILITATION OF IMPAIRED DRIVING OFFENDERS RESEARCH GAPS AND FUTURE NEEDS Gaps in Offender Research Gaps in Intervention Research Gaps in the Implementation of Interventions 105 REFERENCES 107 VI IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

81 EXECUTIVE SUMMARY Background Drinking and driving has been widely recognized as a major social problem in Canada for more than three decades. Although a general decreasing trend in the number of persons killed in a traffic crash involving a drinking driver 1 occurred in Canada between 1995 and 2008, the progress achieved since the late 1990s has been nominal and the number of persons killed and injured in crashes involving drinking drivers remains high. In 2010, (the most recent year for which data are available), 33.6% of fatally injured drivers in Canada had a blood alcohol concentration (BAC) in excess of the legal limit of.08 (Brown et al. 2013). In addition, in 2010, 744 people were killed in Canada in road crashes that involved a driver who had been drinking and approximately 2,733 drivers (excluding Newfoundland and Labrador) were involved in alcohol-related serious injury crashes in Canada (Brown et al. 2013). Similar patterns of nominal or limited declines in drinking and driving are also evident in the United States (U.S.) and Europe as revealed by an examination of crash data. In the U.S., after holding steady between 13,000-14,000 impaired driving fatalities annually for several years, new decreases have been recorded, and fatalities have since dropped to a new low of 10,136 in 2010 and 9,878 in 2011 (NHTSA 2012). It is estimated that there were 31,000 road deaths in 2010 in the European Union (E.U.) and the European Commission estimates 25% of all road deaths in the EU are alcohol-related. It is important to note that comparisons of drink driving crashes and fatalities across countries should be made with caution in light of significant differences in data collection and reporting (ETSC 2011). In light of these trends, increased knowledge and understanding of the profile and characteristics of impaired drivers, the factors that put them at risk for recidivism, available 1 Reported Canadian national data on alcohol-related crashes resulting in fatalities and serious injuries include all drivers that test positive for any amount of alcohol. This means that drivers that are below the legal limit for impairment as well as those above the legal limit are included in these counts. Hence the term drinking driver is used as opposed to impaired driver. EXECUTIVE SUMMARY VII

82 risk assessment instruments and relevant treatment options can inform the activities of transportation, criminal justice, and health practitioners to better identify, manage, and address this high-risk population in the future. Introduction There is increasing recognition of the importance and benefits of tools such as risk assessment and treatment as alternatives to complement punitive measures. Research shows that properly designed strategies and tools developed to match offenders risks and needs with appropriate programs and interventions have beneficial effects (Bonta 2002; NIDA 2006; Oglaff and Davis 2004), including reductions in recidivism as well as reductions in substance misuse that translate into long-term risk reduction and higher levels of public safety. The use of evidence-informed risk assessment tools and practices is linchpin to making the best use of available resources to achieve greater declines in the magnitude of the problem. The use of risk assessment instruments has become commonplace to help practitioners in the criminal justice and remedial driver licensing systems to differentiate among various types of impaired drivers, especially those more prone to recidivism. However, to use these tools effectively, it is important that practitioners possess a clear understanding about the most effective ways to apply risk assessment instruments to better manage impaired drivers and to direct them towards appropriate treatment interventions that are built upon best practices. Purpose and objectives The purpose of this report is to summarize available knowledge about the profile and characteristics of impaired drivers, relevant risk factors, risk assessment instruments and treatment interventions to treat impaired drivers as well as best practices in this field 2. It provides an overview of available research regarding the profile of male and female first offenders in relation to repeat offenders and highlights the inability of existing theories of behaviour to adequately disentangle the heterogeneous nature of the impaired driving offender population. It also provides a summary of relevant risk factors that have been linked to repeat impaired driving offences, while acknowledging some of the key limitations of the research in this field. In addition, the report briefly reviews some of the available tools used to assess risk, available treatment interventions that are applicable to impaired drivers, the research relating to their effectiveness, and current best practices for the treatment and rehabilitation of impaired driving offenders in remedial driver licensing programs. The intent of this report is to provide a high level review of available knowledge that can benefit frontline practitioners working both in the remedial driver licensing system and the criminal justice system. For this reason, additional resources are provided at the end of some 2 There are a broad range of other policies, programs, and interventions for impaired drivers that have been developed, implemented, and evaluated in the past three decades which are beyond the scope of this report. VIII IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

83 sections in the full report in order to afford practitioners an opportunity to review relevant research in more depth. Profile and characteristics of impaired drivers This section briefly summarizes what is known about the profiles and characteristics of adult impaired drivers and draws from the research in criminology, psychology, transportation, health, addiction medicine, and neuroscience. It first examines what is known about male offenders followed by what is known about female offenders. Key dimensions that are considered include: demographic factors, personality and psychosocial factors, substance misuse including engagement in treatment, mental health, cognitive impairment, and driver and criminal history. In all of these sub-sections, distinctions are drawn between first versus repeat offenders. Male impaired drivers Age and sex. Most impaired drivers are between the ages of 20 and 45 years old with almost half of them being between the ages of 20 and 30 years old (Simpson and Mayhew 1991; Jonah and Wilson 1986; Jones and Lacey 2001; Wanberg et al. 2005). Generally speaking, drinking and driving behaviour begins to decrease substantially after the age of 45 years (Hingson and Winter 2003), though this behaviour persists in some drivers into their 60s. Research shows that between 70% and 80% of impaired drivers are male. Ethnicity. Research spanning 30 years suggests that a majority of impaired drivers are Caucasian, although there has been less research on ethnicity relative to other demographic factors such as age and sex. However, while ethnicity is one of the factors that is linked to impaired driving (Ferguson et al. 2002; Jones and Lacey 2001), differences between populations studied and the ways in which questions have been posed have resulted in inconsistent evidence in relation to this factor (Caetano and McGrath 2005). Employment and income. Contrary to popular belief, the majority of impaired drivers are employed, although they are more likely to be unemployed relative to the general population (Wanberg et al. 2005). However, it is important to note that these offenders are more often in the lower-to-middle income range (Ambtman 1990; Wilson and Jonah 1985; Nochajski et al. 1993), and they are more apt to experience occupational instability. Marital status. Research on the marital status of impaired drivers is fairly consistent with some variations. Some studies suggest that more than two-thirds (65-75%) of impaired drivers are single, separated or divorced (Simpson et al. 1996; Wilson 1991; Nochajski et al. 1993). Thus, while approximately half of impaired drivers are in fact married (but perhaps separated), the other half are comprised of those who are currently unmarried or who have never been married. It is important to underscore that many of these studies were conducted EXECUTIVE SUMMARY IX

84 two decades ago at a time when being married was more often equated with having a stable relationship, whereas today this may be less often the case. As such, it may be more useful and practical to consider the level of stability of any co-habiting relationship as opposed to focusing on the specific marital status of this population. Blood alcohol concentration (BAC). Many impaired drivers possess BACs that are quite high relative to the legal limit in North America of.08 (Simpson et al. 2004; NHTSA 2003). In Canada, the mean BAC among fatally injured drinking drivers is.17 (Mayhew et al. 2011). In the U.S., the average BAC among drivers in fatal crashes is.18 (NHTSA 2010). There is evidence to suggest that while BAC is a good measure of level of alcohol use, it is not a reliable indicator of alcohol-related problems, involvement in impaired driving or risk of recidivism (Wieczorek et al. 1992). Personality and psychosocial factors. A wide range of personality and psychosocial factors have been examined in relation to impaired drivers including sensation-seeking, hostility, aggression, psychopathic deviance, assertiveness, antisocial personality, impulse control, risk perception, narcissistic personality, intermittent explosive disorder, external locus of control (i.e., blaming others for problems), and emotional adjustment. In particular, a comprehensive review by Wanberg et al. (2005) reported that the most salient personality variables associated with [DWI] behaviour include: agitation, irritability, resentment, aggression, overt and covert hostility; thrill and sensation-seeking; low levels of assertiveness, low self-esteem, feelings of inadequacy, and sensitivity to criticism and rejection; helplessness, depression, and emotional stress; impulsiveness, external locus of control (blame others for problems); social deviance and non-conformity, anti-authoritarian attitudes (p.23). Alcohol misuse. The role of alcohol misuse in relation to impaired driving behaviour has been studied more than almost any other factor. However, while older research has suggested that substance-related problems were a critical factor in impaired driving offending, more recent research has determined that, although substance use is strongly correlated with impaired driving behaviour, it is not a causal factor. > It has been well-established over the past 35 years that early onset of alcohol and other drug use are predictive of substance use and abuse in adulthood (Hingson et al. 2002; 2003; Grant and Dawson 1997; Wanberg et al. 2005). Generally speaking, those individuals who begin drinking at an early age (under the age of 14) often consume more alcohol as compared to those who begin drinking in their late teens or at the age of 21 (the U.S. legal drinking age). > Research shows that there are two characteristics related to family history that are the most strongly associated with number of impaired driving offences as an adult. These include: having a father with a drinking problem (Schuckit 1999; 2009); and X IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

85 having a relative who was arrested for impaired driving (McMillen et al. 1992a; Wieczorek and Nochajski 2005). > Research investigating the drinking patterns of impaired driving offenders reveals that these individuals generally consumed greater amounts of alcohol per occasion and also consumed alcohol more often than the general population of drinkers (Beirness et al. 1997). There is also research to indicate that a majority of impaired drivers are, in fact, binge drinkers (Caetano and McGrath 2005; Chou et al. 2006). These findings challenge a popular belief that alcoholism is at the root of impaired driving behaviour. > Some research suggests that a diagnosis of alcohol abuse (as opposed to alcohol dependence) is more common among first offenders than repeat offenders, suggesting that this group may generally have lower levels of problem severity relative to repeat offenders (Wieczorek and Nochajski 2005). > Many offenders, regardless of their number of prior offences, are assessed as being in the pre-contemplative stage in relation to the stages of change with regard to their drinking and driving behaviour 3. There is also research demonstrating that impaired driving offenders may be more defensive of their drinking behaviour, and more resistant to self-disclose the extent of their alcohol consumption (BHRCS 2007) than the average patient who engages in alcohol treatment. > A comparison between impaired driving offenders who completed mandated remedial programs versus those who were non-compliant indicated that the latter group possessed the following characteristics: older, lower income in last 30 days, less likely to be married or with a partner, unemployment, similar drinking patterns, more cocaine dependence, higher proportion of positives on axis 1 disorders (e.g., anxiety, depression), and higher proportion of antisocial personality features. Logistic regression further revealed that unemployment was the main predictor of non-compliance (Nadeau 2010), suggesting that cost may be a major obstacle to increased participation among poorer offenders. Mental health. A broad range of mental health and psychiatric conditions have also been linked to impaired driving offenders including antisocial personality disorder, anxiety, conduct disorder, impulse control disorder, narcissism, depression, post-traumatic stress disorder (PTSD), and bipolar disorder. Recognition of and interest in these factors has grown in the past decade, and even more recently as a result of the large number of soldiers and veterans 3 The transtheoretical stages of change model posits that individuals with behaviour problems, such as substance dependence, experience several conditions and differ in their willingness to acknowledge that they have a problem and work towards change (Alexander 2000). Interventions or treatment strategies are most likely to be successful when geared toward the stage of change that the individual client is in. Adapted from Prochaska et al. s (1992) readiness for change process stages, the various stages include: 1) Pre-contemplation (lack of awareness of a problem; no contemplation of change); 2) Contemplation (recognition of a problem; contemplation of change); 3) Preparation (consideration of behaviour change); 4) Action (taking steps to change behaviour such as participation in treatment); and, 5) Maintenance (relapse prevention). EXECUTIVE SUMMARY XI

86 that are involved in impaired driving events either overseas or upon their return to North America. A number of research studies suggest that psychiatric disorders are higher among impaired drivers (Shaffer et al. 2007; Lapham et al. 2001; McMillen et al. 1992a; Wieczorek and Nochajski 2005). Stress is also considered an important factor in relation to impaired driving behaviour (Wanberg et al. 2005). Research examining the effects of anxiety disorder in relation to substance use has also produced significant findings that may have important implications for impaired drivers (Kushner et al. 2011). Many impaired drivers have substantial histories of drug use (Beirness and Davis 2008). Rates of drug use among first and repeat offenders are not only important but also are not limited to soft drugs like marijuana. Cognitive impairment. Executive cognitive function involves the set of abilities that allows one to select behaviour appropriate to a situation, including the ability to inhibit inappropriate behaviours and to focus on a specific task in spite of distraction (Brown et al. 2008, p. 115). Deficits are linked to impulse control and self-regulation, capacity to learn and retain intervention content, problem solving, abstracting, and the speed of information processing, among other abilities. Preliminary studies of neurocognitive characteristics of first-time offenders indicate that they are more likely to suffer deficits related to executive cognitive function compared to normal drivers (Brown et al. 2010a; Couture et al. 2010, August). Driver and criminal history. Research has demonstrated that a significant proportion of impaired driving offenders may also have a history of other driving violations as well as other criminal history. In particular, the propensity for other driving and criminal offences appears to be more pronounced among repeat offenders (Simpson et al. 1996; Jones and Lacey 2001; Syrcle and White 2006; Wieczorek and Nochajski 2005). Impaired driving is likely not an isolated high-risk driving behaviour in some offenders, meaning that some individuals who drive while impaired may also have a history of other unsafe and/or high-risk driving behaviours (Beirness et al. 1997). Moreover, reliance solely on driving records to identify these drivers is problematic in light of gaps in reporting and record systems (Simpson and Robertson 2001; Nochajski and Stasiewicz 2006). Studies investigating criminal history of these offenders also illustrate that at least a portion of convicted impaired drivers have a history of other criminal offences and suggest that strengthening linkages between the criminal justice system and impaired driver treatment programs may be beneficial. Repeat and/or hard core 4 impaired drivers. This segment of the impaired driver population generally has many similar characteristics to first impaired drivers, however these characteristics are often more pronounced (Wieczorek and Nochajski 2005). > Research shows that some 90% of recidivists are male and between the ages of 23 and 45 years. 4 Hard core impaired drivers, also known as hard core drunk drivers are defined as drivers who drink and drive repeatedly, often at high blood alcohol concentrations, and have a history of prior convictions for impaired driving and or substance abuse problems. XII IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

87 > While a majority of repeat offenders can be classified as Anglo-white (Jones and Lacey 2001; Wanberg et al. 2005), it has also been suggested that ethnicity is related to repeat impaired driver status, however this varies according to region. > Repeat offenders are more often single, separated, or divorced, have less education, lower levels of income, and have higher levels of unemployment in comparison to first offenders. > Finally, among repeat offenders, arrests at higher BACs of.18 or over.20 are more common compared to first-time offenders (Wanberg et al. 2005), as is test refusal at the roadside (Robertson and Simpson 2002). > Findings from the literature exploring personality differences between first and repeat offenders are mixed. Some studies report that repeat offenders demonstrated higher levels of hostility, sensation-seeking, psychopathic deviance, mania and depression, and antisocial tendencies, as well as lower levels of assertiveness and emotional adjustment, self-esteem, locus of control, social desirability (McMillan et al. 1992a; Wieczorek and Nochajski 2005; Cavaiola et al. 2007). Other studies have failed to identify significant differences between these two groups (Cavaiola and Wuth 2002; Wanberg et al. 2005). These apparently contradictory conclusions reveal the fact that the research to date has failed to adequately disentangle the significant heterogeneity observed in the impaired driver offender population. Socially desirable responding among impaired drivers in self-report studies may also bias our understanding of personality and behavioural factors (Schell et al. 2006). > Similar to first offenders, age of onset, family history, and alcohol misuse issues play an important role in relation to repeat impaired driving offenders. A comprehensive review of the literature by Wanberg et al. (2005) similarly reported that repeat offenders have higher levels of disruptive alcohol use symptoms. > Repeat offenders have significantly higher levels of psychiatric symptoms (Wieczorek and Nochajski 2005; Wanberg et al. 2005; Jones and Lacey 2001; Simpson et al. 1996). It has been reported that there are significant differences in drug use by the number of prior offences and persistent offenders have higher levels of use than first offenders (Wieczorek and Nochajski 2005; Wanberg et al. 2005; White and Gasperin 2006). Mental health issues among impaired drivers are an important consideration given that treatment is more difficult when individuals possess emotional and psychiatric problems in conjunction with substance-related problems (Lapham et al. 2001). Hence, not only can co-occurring disorders decrease the effectiveness of treatment, but they are also considered a predictor of poorer treatment outcomes (Lapham et al. 2001; Laplante et al. 2008; Shaffer et al. 2007). EXECUTIVE SUMMARY XIII

88 > Research reveals that repeat offenders are more likely to possess cognitive impairments. The most intervention-resistant offenders have a decreased ability for self-regulation, for learning and retaining intervention content, and for exercising good decision-making even when sober. Not all of these problems are attributable to alcohol abuse severity. This suggests that new strategies in the design of remedial programs and interventions directed at some offenders with the highest risk of recidivism may be needed (Ouimet et al. 2007; Maldonado-Bouchard et al. 2012; Brown et al. 2008). > Repeat offenders are also more likely to have more traffic offences and to have been involved in crashes more frequently than drivers that are convicted of a first impaired driving offence (McMillen et al. 1992a; Nochajski and Wieczorek 2000; Wieczorek and Nochajski 2005) according to official records and/or self-report. Female impaired drivers For several decades, road safety research has demonstrated that fatalities and injuries related to road crashes (due to alcohol or other unsafe driving behaviours) have predominantly involved males (Mayhew et al. 1981; Beirness and Simpson 1988; Mayhew and Simpson 1990; Mayhew et al. 1990; Kelley-Baker and Romano 2010). In Canada, since 2002, females have accounted for 13-16% of fatally injured impaired drivers, reaching a high of 16.4% in 2006 (TIRF 2012). However, this percentage seems to have stabilized in the past four years, and, overall, females continue to account for a minority of this population. An examination of alcohol crash data from the U.S. Fatality Analysis Reporting System (FARS) indicates that the involvement of female drivers in alcohol-impaired road crashes has remained fairly stable with incremental increases from 12% in the 1980s to 14% in the 2000s. Since 2006, the percentage of women drivers who tested positive for any amount of alcohol in fatal crashes has averaged 16% annually, while in ,837 fatalities in crashes involved an alcohol-impaired female driver (NHTSA 2009). Conversely, impaired driving incident and arrest data reveal a different picture. In Canada, the impaired driving rate for females generally declined up to 1997 and remained stable through to It has for the most part increased since 2005 and in 2011, females accounted for one in every six impaired drivers, compared to 1 in 13 in 1986 (Perreault 2013). In the United States, the number of female impaired driving arrests in the U.S. rose nationally by 28.8% between 1998 and 2007 (Lapham et al. 2000; Schwartz and Rookey 2008). Thus, while in the 1990s it was estimated that about 10% of impaired drivers were female, as of the 2000s it has been estimated that women account for closer to 20% (Wanberg et al. 2005; Schwartz and Rookey 2008). XIV IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

89 There are three main hypotheses that explain these increases. Female roles in society have changed considerably (Popkin 1991; Bergdahl 1999; Mayhew et al. 2003; Robertson et al. 2011a; Tsai et al. 2008), there have been changes in social norms (Gudrais 2011; Popkin 1991), and also changes in social control mechanisms (Farrow and Brissing 1990; Robertson et al. 2011a; Schwartz and Rookey 2008; Schwartz and Steffensmeier 2007). Although much of the research investigating female impaired drivers is dated (Robertson et al. 2011b), in 2013 a series of case studies were conducted with more than 150 convicted female impaired driving offenders who participated in interview focus groups in four U.S. states (California, Michigan, Missouri and New York) (Robertson et al. 2013). In particular, three distinct profiles of female impaired drivers also emerged from this study, and it is estimated that more than three-quarters of the study participants matched one of these profiles: 1. Young women who drink in order to fit in and consume alcohol and/or binge drink at house parties and bars; 2. Recently married women with spouses who drink or who have children and drink following the birth of their children as a means for coping with loneliness; and, 3. Divorced older women and/or empty nesters who begin to drink later in life (after age 40) following a catalyst such as the death of a parent, end of a marriage, or children leaving home. Age and sex. Robertson et al. (2013) found that female impaired driving offenders ranged in age from late teens to mid-60s, suggesting that women of all ages drink and drive. However a majority of participants were an estimated 20 to 40 years of age. Generally, rates of involvement in alcohol-impaired motor vehicle crashes decrease with age, and the population of greatest concern is often young females (Peck et al. 2008). In particular, the increasing involvement of young women with alcohol, in combination with their inexperience driving and their growing propensity for risky driving (Lynskey et al. 2007; Tsai et al. 2010) warrants attention and further research. Education and employment. The literature regarding levels of education and employment among female impaired drivers is inconsistent. Female impaired drivers are generally older than men and have higher levels of education (Peck et al. 2008) but lower paying jobs (Chalmers et al. 1993; Shore and McCoy 1987). Low academic achievement in young females represents a risk factor for impaired driving comparable to that observed in males (McMurran et al. 2011). Marital status. A significant proportion of female impaired drivers are single, divorced, or separated, or are more likely to be living with a partner with an alcohol problem compared to women with no past impaired driving offences (McMurran et al. 2011; Chang et al. 1996; EXECUTIVE SUMMARY XV

90 Shore and McCoy 1987; Argeriou et al. 1986). In fact, when compared to male impaired drivers, females are even more likely to be divorced or single (McMurran et al. 2011; Chang et al. 1996; Shore and McCoy 1987; Argeriou et al. 1986). Generally speaking, female impaired drivers are more likely to be the primary caretaker of children at the time of arrest, are more likely to have experienced abuse, and are more likely to have physical and mental health needs compared to their male counterparts (Bloom et al. 2003). Personality and psychosocial factors. In contrast to the availability of research examining this issue among male impaired drivers, there have been fewer studies examining the prevalence of personality and psychosocial factors among female impaired drivers. A review of these studies suggests that psychosocial problems among female impaired drivers may not be uncommon and that, at least a portion of these women may experience depression, boredom, and problems at home and school that are related to their drinking (McMurran et al. 2011). Alcohol misuse. Alcohol use among women is a very important factor to consider in relation to impaired driving for several reasons. Research shows that women metabolize alcohol differently than men (Gudrais 2011; Greenfield 2002). In addition, females generally have less water in the body and a lower body mass. Physiological differences also contribute in part to the more rapid progression of alcohol dependence such that women often require medical intervention an average of four years earlier than males who are problem drinkers (Gudrais 2011). It is also important to note that a study by Elliott et al. (2006) found that women are more vulnerable to all types of traffic incidents following alcohol consumption. > Most recently, Robertson et al. (2013) reported that the extent of substance use varied substantially across study participants. It is estimated that almost one-half of women reported early onset of drinking with many experimenting with alcohol and/ or drugs in their early or mid-teen years; the lowest reported age of onset drinking was nine years old. Conversely, it was estimated that between one-quarter and onethird of women did not begin to regularly use or develop a problem with alcohol or drugs, or begin to drive after using these substances, until they were in their 30s or 40s. > A constellation of family history factors, including a history of alcoholism within the family, experience with abuse, anxiety and depression, and family and personal relationships that encouraged heavy drinking (White and Hennessey 2006), are associated with female impaired driving offending to varying extents, however the specific influence of each factor is unclear. > Estimates of alcohol diagnoses among female impaired drivers vary but are significant and comparable to or greater than males (Lapham et al. 2000; Maxwell and Freeman 2007; Maxwell 2011). In a study by Robertson et al. (2013) a universal XVI IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

91 theme that emerged in interview focus groups with more than 150 convicted female impaired drivers was reports that they drank for emotional reasons, or that alcohol consumption was a coping mechanism to help them manage their emotions and stress. Mental health. Findings indicate that there is a need to treat some female impaired drivers not only for alcohol misuse problems but mental health problems as well (McMurran et al. 2011). Female impaired driving offenders have significantly higher psychiatric co-morbidity relative to their male counterparts (Laplante et al. 2008). Diagnoses of anxiety, depression, and post-traumatic stress disorder (PTSD) are common among female impaired driving offenders. Histories of trauma are also not uncommon among female impaired drivers (Robertson et al. 2013). > The use of illicit and licit substances among female impaired drivers is prevalent. Some studies suggests that involvement in drug use may be more comparable among males and females (Lapham et al. 2000). However, Maxwell and Freeman (2007) reported that the use of illicit drugs was higher among females as compared to males, noting that females most likely to be diagnosed with a primary problem with sedatives or opiates, whereas males were most likely to be diagnosed with a primary problem with alcohol and cannabis (Maxwell 2011). More recently, Robertson et al. (2013) reported that, although prescription drug use was common, less than one-third of female impaired drivers reported use of illicit substances. Given that the use of drugs appears to be somewhat common among female impaired drivers, it is important that female offenders are appropriately screened, identified, and treated for all drug use disorders. Cognitive impairment. While there has been limited research into the prevalence of cognitive impairments among female impaired drivers, Brown et al. (2013) reported that executive control appears to be a feature of female first impaired driving offending and that their ability to identify goals, plan, execute, inhibit old behaviour patterns, and learn from experience is reduced. These impairments worsened with alcohol intake. As such, alcohol appeared to contribute to female first impaired driving offending through acute and chronic disruption of executive control functioning. Driver and criminal history. There are limited data to suggest that a smaller number of female impaired driving offenders relative to males have a history of other traffic offences or criminal offences, although more research into this topic is needed. Common criminal offences in females may include drug offences, theft offences, and assault (Caldwell-Aden et al. 2009). Repeat female impaired drivers. Female repeat impaired driving offenders often share similar characteristics to their male counterparts. EXECUTIVE SUMMARY XVII

92 > Older research suggests that repeat female offenders are approximately 30 years old but more current research on this issue is needed. > Similar to males, there is also evidence that this population has lower levels of education, employment, and income, and is much more likely to be single, separated, or divorced than first offenders. > Like their male counterparts, repeat female impaired driving offenders are more likely to drink more frequently and exhibit higher levels of impairment, more often abuse drugs, and utilize treatment services (Argeriou et al. 1986). > However, there are some differences between female and male repeat offenders. For example, repeat female impaired driving offenders have higher levels of psychiatric co-morbidity than male repeat offenders and are more likely to also use drugs (Laplante et al. 2008; Maxwell 2011). > Recidivism rates among male and female impaired drivers show some consistent patterns, depending on the studies consulted. Available data suggest recidivism risk may be higher for young males than females (Argeriou et al. 1986; Jones and Lacey 2001; McMurran et al. 2011; Webster et al. 2009; Wells-Parker et al. 1991), but it appears that risk of recidivism may converge as adults of both sexes age (Lapham et al. 2000). However, a comparison of rates among older offenders revealed few differences between sexes (Laplante et al. 2008; Rauch et al. 2010). As relatively few studies have specifically examined this issue, more research is needed. Summary of similarities and differences between males and females On average, impaired drivers of both sexes are generally aged 20 to 40, with many offenders being in their 30s. Relative to the general population, impaired drivers of both sexes also are more likely to have less education and lower levels of employment and income; this finding is more pronounced among repeat offenders. Similarly, impaired drivers of both sexes are more likely to be single, separated, or divorced. Again, this finding is more pronounced among repeat offenders. Alcohol-related diagnoses are very common among impaired drivers of both sexes. In particular, the age of onset of drinking and family history warrant attention. To reiterate, while such diagnoses are highly correlated with impaired driving offending, they are not necessarily a causal factor. Both male and female impaired drivers have higher levels of psychiatric symptoms relative to the general population so co-occurring disorders should not be overlooked during screening and assessment of this population. Moreover, recidivism rates for impaired driving among men and women of adult age appear similar following a first alcohol-related conviction. XVIII IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

93 There are also some important differences between male and female impaired drivers. Men appear to exhibit a higher degree of antisocial attitudes and behaviours relative to women, although research comparing these populations on this dimension is sparse. Conversely, women experience more severe psychological and mental health symptoms as well as report greater involvement in drugs. Men may be more defensive about alcohol problems and, in particular, repeat male impaired drivers may demonstrate a greater readiness for change. In addition, younger males appear to have higher recidivism rates relative to females in this age category. Male impaired drivers also have more extensive histories of driving offences and other criminal offences as well as more prior experience with impaired driving interventions. Impaired driving risk factors Risk factors are characteristics that are identified (according to sufficient research evidence) to be indicators of the potential for a group of individuals with shared characteristics to engage in a specific behaviour in the future. It cannot be underscored enough that understanding the factors associated with recidivism is critical to our capacity for better detection of highrisk offenders and our ability to orchestrate effective countermeasures (Ouimet et al. 2007, p. 743). Generally speaking, risk factors are organized in two distinct categories: 1) static factors (e.g., number of prior offences) that cannot be changed; and, 2) dynamic factors (e.g., substance abuse) which may change over time (Gendreau et al. 1996; DeMichele and Lowe 2011). Again, risk factors are relative to a group and not an individual and, subsequently, these measures are not very robust (Nadeau 2010). Risk assessment is a process that utilizes identified risk factors (usually in relation to multiple domains) to predict future behaviour. Risk assessment is not an exact science and risk factors only provide insight into the probability or likelihood of recidivism of offenders based upon existing research that is available. In this regard, much of the research around risk prediction has focused on criminal offenders and, in particular, those who have committed violent and/ or sexual offences. More recently, the quality of instruments 5 used with offenders has greatly improved (to Andrews and Dowden 2006) as our understanding of risk factors has grown. Risk assessment instruments that possess a higher degree of accuracy in prediction generally account for multiple risk factors to reach a determination as to the probability of recidivism, and place a greater emphasis on objective measures as opposed to just the reliance on professional judgment which is more often subjective. A broad range of risk factors have been noted in the literature regarding impaired drivers including: sex, age, marital status, socio-economic status, history of prior treatment, impaired driving history, criminal history of violent aggression, prior traffic offences, test refusal or 5 It is equally important that risk assessment instruments demonstrate proven reliability and are scientifically validated and standardized on an appropriate population. EXECUTIVE SUMMARY XIX

94 high-bac, and drinking patterns to name a few (Syrcle and White 2006). Yet, these studies vary dramatically in terms of the population studied, sample size, variables and measures utilized, data sources, analyses conducted, comparison groups employed, the time period used to measure recidivism, and the interpretation of results. Moreover, the number of studies that have examined the reliability of each individual risk factor is relatively small, which makes the drawing of conclusions a challenge. In light of the limitations associated with research investigating risk factors associated with impaired driving, what is currently known about impaired driving risk factors should be interpreted cautiously. At best, no single impaired driving risk factor provides a clear indication regarding the potential for future impaired driving recidivism. Collectively, however, these risk factors may provide some insight that enable practitioners to gauge the need to further explore individual cases and the need for more intensive interventions. In brief, there is some limited evidence to support the use of the following factors as predictors of impaired driving recidivism among males: > younger age; > male sex; > life history including family members or spouses with alcohol problems or impaired driving arrests; > early onset alcohol and drug use and abuse, frequency of drinking, amount of alcohol consumed; > BAC is often cited as a reliable predictor of recidivism but research findings are mixed and more recent research suggests that BAC alone is not useful and should be interpreted cautiously or in combination with other predictive variables (Caviola et al. 2007; Dugosh et al. 2013); > Instruments with some strength in predicting recidivism include the MAST, the MAC scale of the MMPI, and the subtle items of alcoholism on the RIASI. Of importance, different jurisdictions or offender samples will have higher or lower rates of failing, and agencies need to make decisions about how to balance the positive and negative predictions. That is, assessment is an exercise in prediction, and prediction has error. It is a bit of an art to balance these issues, but also a matter of agency capacity. The bottom line is that because of decisions regarding instrument precision, practitioners should be careful about comparing different assessments and even the same assessment across different populations; > Biomarkers can detect the presence of alcohol disorders fairly accurately and a number of studies have investigated the extent to which biomarkers are predictive XX IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

95 of impaired driving recidivism. More recently, there is research to suggest that biomarkers are not a good predictor of recidivism, individually or as a group. The primary reason for this is that biomarkers may not capture the drinking patterns that are most common among impaired driving offenders e.g., binge drinking (Couture et al. 2010); > A poor driving record that includes offences both prior to and following the initial impaired driving offence is predictive of recidivism (Peck et al. 1994; Rauch et al. 2002; Wieczorek and Nochajski 2005; Cavaiola et al. 2007). However, some have noted that prior impaired driving arrests may not be a good predictor as the presence of prior arrests is influenced to a large extent by the level of impaired driving enforcement as well as the length of the look-back period for counting prior arrests (Nochajski and Stasiewicz 2006); > Research investigating risk factors associated with criminal re-offending has identified a number of objective and verifiable risk indicators that are useful to distinguish between first and repeat impaired drivers. These variables are associated with an offender s criminal history and include: age at time of first arrest for any criminal act, age at time of first impaired driving conviction, having a prior summary of alcohol- or drug-related offence, having a prior misdemeanor offence, having a misdemeanor offence for a crime against persons, or having five or more prior moving violations (Dugosh et al. 2013, p.8); > Research suggests that a high rate or pattern of BAC fail readings from the alcohol interlock, particularly in excess of.02, is predictive of the likelihood of impaired driving recidivism (Marques et al. 2003; Beirness and Marques 2004). Researchers have also determined that the presence of elevated BAC tests during early morning hours can also assist in predicting future impaired driving offences (Beirness and Marques 2004); and, > A number of recent studies have identified risk factors among repeat offenders in comparison to first offenders (Nadeau 2010). Low levels of participation or involvement in treatment and treatment interventions is considered predictive of recidivism (Aharonovich et al. 2003; Crews et al. 2005; Syrcle and White 2006; Wanberg et al. 2005). Neurocognitive deficits have also been reported as predictive of recidivism among repeat offenders. More specifically, these deficits can contribute to variation in affect, impulsivity, problem solving, perception and memory (Glass et al. 2000; Ouimet et al. 2007). Finally, a reduced ability to change is also predictive among repeat offenders of future impaired driving offences (Buntain-Ricklefs et al. 1995; Glass et al. 2000; Ouimet et al. 2007). EXECUTIVE SUMMARY XXI

96 With regard to female impaired drivers, there is one key study that examined differences in risk factors among men and women. For the most part, few differences were found in terms of predictive variables with the exception that women were more likely to report a history of aggressive behaviour towards a partner than were males, and this indicator was associated with increased recidivism (Lapham et al. 2000). While it is clear that a wide range of risk factors have been examined in relation to the prediction of repeat impaired driving offences in the past two decades, the findings from this research are inconsistent in many cases and far from conclusive. There are only a small handful of common factors that have been investigated across several studies, however with regard to criminological research, more is known about risk factors among repeat drunk drivers. For these reasons, practitioners in the field are encouraged to take a broader view of and approach to the use of these factors, and focus on the presence of a number of risk factors collectively as a basis to inform decisions, as opposed to the presence or absence of individual factors. Much more research on this issue is needed before definitive conclusions can be reached. Risk assessment instruments The effective management of the many different types of impaired drivers is based upon the identification and development of a range of supervision strategies and interventions specifically geared towards those offenders who are more or less amenable to behaviour change. This is a fundamental principle of evidence-based practices. Of considerable importance, the use of valid and reliable risk assessment instruments is essential to accurately differentiate between the different types of impaired drivers that exist and ensure that they are streamed into appropriate interventions designed to address their specific risks and needs. These assessment tools are designed to identify as many potential cases as possible, while at the same time minimizing the number of false-positives (i.e., identifying someone as highrisk for re-offending when they are not). Some of these instruments are not as strong and have demonstrated limited validity and reliability in relation to the accurate prediction of future impaired driving events, including the following: > Mortimer Filkins (MF) (Chang et al. 2002; Wendling and Kolody 1982); and, > Driver Risk Inventory (DRI) (Chang et al. 2002). In light of the strengths and weaknesses associated with many of the available instruments, many jurisdictions rely on the outcomes of several instruments during the assessment process in order to produce a more complete picture of impaired driving offenders. The full report briefly describes some of the instruments that are most commonly used across Canada and the United States. Each instrument is described in terms of type XXII IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

97 of administration, who it can be administered by, number of items, time required for administration, training required for administration, scoring, summary of psychometrics, limitations, cost, and source. In addition, a few key references are identified in relation to each instrument in order to provide additional information to practitioners seeking more knowledge about the risk assessment instrument. The following is a list of the instruments described in the full report: > ADS (Alcohol Dependence Scale); > ASUDS-R (Adult Substance Use and Driving Survey Revised); > ASI (Alcohol Severity Index); > AUDIT (Alcohol Use Disorders Identification Test); > IDTS (Inventory Drug-Taking Situations); > DAST (Drug Abuse Screening Test); > LSI-R (Level of Service Inventory-Revised); > MAST (Michigan Alcoholism Screening Test); > SASSI (Substance Abuse Subtle Screening Inventory); > RIASI (Research Institute on Addiction Self Inventory); and, > Biomarkers. There are no clear indications of the superiority of any one screening instrument or set of instruments and procedures. To summarize, there are many impaired driver assessment instruments that are available and utilized across North America. Yet not all of these instruments have been validated on an impaired driver population and few have undergone rigorous or independent evaluation efforts. It is for this reason that many jurisdictions rely upon a combination of these instruments to guide the assessment process. It is essential to underscore that problem substance use behaviour in and of itself is not the source or cause of persistent impaired driving behaviour, but instead merely a correlate of it. Therefore while assessment instruments designed to identify the likelihood of relapse among substance using and even impaired driving populations provide valuable information, these tools frequently overlook the role of criminogenic and socio-psychological factors that are important contributors to chronic offending. Of the available risk assessment instruments to date, both the LSI-R and ASUS 6 instruments appear to be the most well-grounded in theory and based upon a solid theoretical 6 The Adult Substance Use Survey (ASUS) is a self-report survey that consists of 64 items designed to assess an individual s perceived alcohol and drug use. The survey also provides a brief mental health screen. It can either be self-administered (paper-and-pencil) or administered orally by a practitioner. Unlike the ASUDS-R, this screening instrument is not specific to an impaired driving offender population although both tools were developed by the Center for Addiction Research and Evaluation (CARE). EXECUTIVE SUMMARY XXIII

98 foundation. These instruments incorporate a range of recognized concepts stemming from several relevant disciplines including criminology, psychology, sociology and addictions, and these concepts have been repeatedly tested and validated through extensive research. Such a comprehensive approach is essential in light of the well-documented complexity associated with impaired driving behaviour and the diversity of underlying processes that have been used to explain persistent offending by this population. It should be underscored that assessment approaches that are multi-trait and multi-method provide more accurate results (Campbell and Fiske 1959). Looking forward, there is some clear direction as to ways to strengthen research that can guide the development of empirically-based risk assessment instruments. First, with regard to the evaluation of risk assessment instruments, Brown and Ouimet (2013) underscore that Longer duration perspective evaluations of assessment protocols for prediction of recidivism are urgently needed (p.311). Second, the research undertaken by Dugosh et al. (2013) provides a basis to begin to integrate criminological theories and empirically-based risk factors to enhance risk assessment tools for impaired drivers. The inclusion of these factors in risk assessment tools can help to strengthen the internal validity of them. Treatment interventions Alcohol education programs for impaired drivers show an average reduction in recidivism of approximately 10% (NHTSA 1986; Wells-Parker et al. 1995). Among offenders who suffered from some degree of substance misuse problems, those programs that utilized a therapeutic approach are considered to have a greater effect, illustrating the value of treatment as an intervention to encourage rehabilitation and behaviour change (Wanberg et al. 2005). The results of a risk assessment in conjunction with resources that are available are two critical components of any intervention strategy. There is growing evidence to suggest that combining appropriate sanctions and supervision with treatment interventions can be more effective than either strategy alone. The partnering of these different strategies can expand opportunities to achieve long-term risk reduction and to reduce and/or prevent repeat offending. In order to maximize the effectiveness of this approach it must be assessment driven and combine appropriate levels of supervision with appropriate treatment interventions. The full report briefly describes a variety of common approaches to treatment including: > screening and brief interventions (SBI); > motivational interviewing (MI); > cognitive behavioural therapy (CBT); > pharmacological interventions; and, XXIV IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

99 > web-based interventions. Each intervention is described in terms of purpose and objectives, general effectiveness, staff training requirements, mechanism of delivery, and strengths and weaknesses. Note that some of these interventions have been specifically evaluated on an impaired driving population whereas others are merely a source of emerging interest and more research is needed to gauge effectiveness with impaired drivers. In addition, a few key references are identified in relation to each intervention in order to provide additional information to practitioners seeking more knowledge about specific strategies. In summary, there is a range of treatment interventions that have been shown to be promising or effective in reducing recidivism among impaired driving offenders. However, each of these strategies rely upon different levels of resources, staff with different backgrounds and qualifications, different amounts of time, and have varying levels of cost. In addition, some interventions are more easily implemented and delivered than others. Perhaps what is most important is that efforts are made to best match interventions to the individual risks and needs of each offender. Best practices for treatment and rehabilitation of impaired driving offenders Health Canada produced a Best Practices report (2004) that was based upon a thorough literature review, consultation with experts, and interviews with key informants. The aim of the report was to compile current knowledge on driving while impaired remedial programs across Canada. Specifically, the report addresses the planning and delivery of education programs and treatment and rehabilitation programs. The report in its entirety can be found online: treatment_rehab_driving_impaired_practices.pdf Research gaps and future needs Much has been learned about the profile and characteristics of impaired drivers over the course of the past three decades. To a lesser extent, knowledge has also grown with regard to the factors that put them at risk, the types of assessment instruments that are appropriate for this population, and the types of treatment interventions that can begin to address their risks and needs. Still, continued efforts are needed to increase understanding of these topics and to inform approaches that can best prevent impaired driving behaviour, as well as manage, supervise and treat those that are detected and processed through the criminal justice system. A EXECUTIVE SUMMARY XXV

100 number of topics that reflect gaps in offender research, gaps in intervention research, and gaps in implementation and practice warrant future attention. > Perhaps most pressing in the field of research is the need to integrate existing knowledge stemming from diverse disciplines as a basis to explore and develop more holistic, robust and complex models of impaired driving behaviour that acknowledge the heterogeneity of this population. A core feature of this initiative should be to increase understanding of the interactions and effects of different characteristics of offenders. > Greater knowledge and understanding of relevant risk factors that influence future offending is also a critical need. > The development of valid, reliable and practical screening and risk assessment instruments that can accurately distinguish between offenders not only with regard to risk related to substance use but also risk of re-offending and individual-specific trajectories to impaired driving behaviour are essential to inform decision-making and the allocation of resources. > Future efforts to investigate the effectiveness of interventions must account for not only the increasingly complex environment in which such interventions are delivered, but also the web of factors that play an important role. > A range of research questions remain that must be addressed. These include:» Is it possible to achieve an optimal balance between sanctions/supervision and rehabilitation/treatment for offenders with different levels of risk?» What interventions or combination of interventions provide the best outcomes for different subpopulations of offenders?» Are there commonalities and differences across interventions that can provide insight into the essential ingredients of effective interventions? This may include an examination of content, delivery mechanisms, training, duration, key features, and the emphasis that is placed on sanctioning, rehabilitation or both.» Is there an optimal duration for the various interventions that are available, including educational programs, treatment, probation, and alcohol monitoring technologies?» Is it possible to achieve the outcomes associated with longer-term and more intensive treatment interventions using well-designed programs that are more cost-effective and shorter in duration? XXVI IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

101 » What characteristics of offenders are most useful to appropriately match them to effective interventions? > With regard to the implementation of interventions, the following issues should be addressed:» Increases in female involvement in impaired driving arrests and crashes warrant close monitoring and may have important implications for the delivery of interventions in order to account for differences across sexes and ages.» There is growing awareness that additional and complementary services may be required for specific sub-populations of offenders such as those who possess deficits in executive cognitive functioning, those who suffer from cooccurring disorders, and those offenders identified with polysubstance (i.e., alcohol and drugs) use.» While much has been learned with regard to effective interventions, less work has been focused on the implementation of such programs to ensure that they are delivered in ways that demonstrate fidelity to the model. EXECUTIVE SUMMARY XXVII

102 1. BACKGROUND Drinking and driving has been widely recognized as a major social problem in Canada for more than three decades. Due to the significant number of fatalities and serious injuries caused by impaired drivers each year, and the growing concern associated with the problem (Simpson and Mayhew 1991), jurisdictions have worked to develop a comprehensive approach to address it. Since the 1980s, education and awareness programs have expanded, enhancements have strengthened criminal and administrative laws, and enforcement activities have become prominent and commonplace. Heightened attention along with a myriad of efforts to combat the problem have resulted in significant declines, with the proportion of fatally injured drivers with BACs in excess of the legal limit dropping 27% between 1981 and 1988 (Mayhew et al. 1996). In the 1990s, progress continued, but declines were less pronounced as the proportion of fatally injured drivers with a BAC over the legal limit dropped just 13.9% between 1990 and 1998 (Mayhew et al. 2011). These shrinking declines were attributed to the fact that the characteristics of the problem had changed (Simpson et al. 1996). It was suggested that the deterrent effect associated with available countermeasures was less pronounced among heavier drinkers who persisted in driving after drinking, often with high BACs (Simpson et al. 1996) and who were responsible for a very significant portion of the problem (Beirness et al. 1997; Simpson et al. 1996). Hence, major decreases in the magnitude of the problem have been more difficult to achieve. Although a general decreasing trend in the number of persons killed in a traffic crash involving a drinking driver 1 continued in Canada between 1995 and 2008, the progress achieved since the late 1990s has been nominal and the number of persons killed and injured in crashes involving drinking drivers remains high. In 2010, (the most recent year for which data are available), 33.6% of fatally injured drivers in Canada had a BAC in excess of the legal limit of.08 (Brown et al. 2013). In addition, in 2010, 744 people were killed in Canada 1 Reported Canadian national data on alcohol-related crashes resulting in fatalities and serious injuries include all drivers that test positive for any amount of alcohol. This means that drivers that are below the legal limit for impairment as well as those above the legal limit are included in these counts. Hence the term drinking driver is used as opposed to impaired driver. BACKGROUND 1

103 in road crashes that involved a driver who had been drinking and approximately 2,733 drivers (excluding Newfoundland and Labrador) were involved in alcohol-related serious injury crashes in Canada (Brown et al. 2013). Public opinion polls demonstrate the high level of concern associated with impaired driving among young drivers in particular, and data indicate that concern about this population is warranted. A national survey revealed that a majority (82.1%) of young Canadian drivers (aged 16-24) agreed that young drinking drivers are a very or extremely serious problem; almost the same percentage of adult drivers (aged 25+) (83%) equally agreed that young drinking drivers were a problem (Marcoux et al. 2011). In 2010, drivers aged 25 and under accounted for 22.7% of all fatally injured drivers (312 of 1,372). Among fatally injured drivers who had positive BACs, 28.4% (123 out of 433) were aged 25 and under. Among drivers aged 25 and under, 45.4% of those tested (123 out of 271) had positive BACs (Brown et al. 2013). This is a concern because, while young drivers make up a small proportion of the drinking and driving problem relative to other drivers, some research has shown that when young drivers do drink and drive, they are more likely than adults to experience an alcohol-related crash (Bingham et al. 2009). To illustrate, among drivers under the age of 21, positive BACs are associated with higher relative crash risks compared to drivers over the age of 21 (Peck et al. 2008). Further, among young drivers, the likelihood of being involved in a crash has been shown to be higher at all BAC levels compared to older drivers (Peck et al. 2008). To summarize, according to Vanlaar et al. (2012) who reviewed recent trends in drinking and driving in Canada, There has been a continued and consistent decrease in the number of fatalities involving a drinking driver in Canada. This remains true when looking at the number of fatalities involving a drinking driver per 100,000 population, and per 100,000 licensed drivers. This decreasing trend is also still apparent when considering the percentage of persons killed in a traffic crash in Canada involving a drinking driver, although less pronounced. Data from [TIRF s] Road Safety Monitor further show that the percentage of those who reported driving after they thought they were over the illegal limit has also declined. However, regardless of the apparent decreasing trend in drinking driving fatalities and behaviour, reductions have been relatively modest, and fatalities in crashes involving drivers who have consumed alcohol remain high at unacceptable levels (p.297). Of equal concern, from there were 40,144 criminal convictions for impaired driving 2 in Canada (Dauvergne 2012), placing a significant burden on the criminal justice system. In the United States, progress in reducing alcohol-impaired deaths and injuries has been very comparable to Canada in that progress was more pronounced in the late 1980s and early 1990s, with little headway being achieved between the mid-1990s and mid-2000s. Only 2 The term impaired driving is used throughout this report to refer to the drivers who are considered to be legally impaired or have a blood alcohol concentration in excess of the legal limit of IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

104 more recently has renewed progress emerged. After holding steady between 13,000-14,000 impaired driving fatalities annually for several years, new decreases have been recorded, and fatalities have since dropped to a new low of 10,136 in 2010 and 9,878 in 2011 (NHTSA 2012). In addition, convicted impaired driving offenders represent approximately 15% of the four million adults on probation. To better manage and intervene with this population, probation agencies are seeking guidance to help identify better ways to supervise these individuals (DeMichele and Payne 2013). Progress in the European Union (EU) has been more difficult to gauge in light of inconsistencies in reporting across member states. However, it has been estimated that 2% of all vehicle kilometers travelled (VKTs) driven in the EU are with an illegal BAC, keeping in mind that the average legal BAC limit is more likely to be.05 or lower. It is estimated that there were 31,000 road deaths in 2010 in the EU and the European Commission estimates 25% of all road deaths in the EU are alcohol-related. It is important to note that comparisons of drink driving crashes and fatalities across countries should be made with caution in light of significant differences in data collection and reporting (ETSC 2011). In light of these trends, increased knowledge and understanding of the profile and characteristics of impaired drivers, the factors that put them at risk for recidivism, available risk assessment instruments and relevant treatment options can inform the activities of transportation, criminal justice, and health practitioners to better identify, manage, and address this high-risk population in the future. BACKGROUND 3

105 2. INTRODUCTION In light of the shrinking declines in the drinking and driving problem in the past decade, renewed efforts are needed to better target those Canadians who continue to drive after drinking and place themselves and others at high risk for death and injury. To this end, there is growing awareness among researchers and practitioners of the limitations of a solely punitive approach to the problem. Although there is less awareness of these limitations at a political or public level the get tough philosophy still dominates much of the application of justice. In particular, persistent impaired drivers are more often viewed by researchers and practitioners as offenders who may suffer from a treatable problem with substance misuse 3 (and who may or may not have other issues as well), as opposed to being viewed merely as persistent offenders who are unlikely and/or not able to change their behaviour. The good news is that there is increasing recognition of the importance and benefits of tools such as risk assessment and treatment as alternatives to complement punitive measures. Research shows that properly-designed strategies and tools developed to match offenders risks and needs with appropriate programs and interventions have beneficial effects (Bonta 2002; NIDA 2006; Oglaff and Davis 2004), including reductions in recidivism as well as reductions in substance misuse that translate into long-term risk reduction and higher levels of public safety. At the same time, growing economic challenges mean that jurisdictions are seeking ways to use resources more effectively and efficiently to best manage drinking and impaired drivers and protect the public. The use of evidence-informed risk assessment tools and practices is one means to attain this goal and a linchpin to making the best use of available resources to achieve greater declines in the magnitude of the problem. The use of risk assessment instruments has become increasingly commonplace to help practitioners differentiate among distinct types of impaired drivers, especially those more prone to recidivism. However, to use these tools effectively, it is important that practitioners possess a clear understanding of the research pertaining to the characteristics and profile of 3 Substance misuse is defined by the World Health Organization as the use of a substance for a purpose not consistent with legal or medical guidelines, as in the non-medical use of prescription medications. The term is preferred by some to abuse in the belief that it is less judgmental. (WHO 2013) 4 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

106 impaired drivers about the appropriate ways to apply risk assessment instruments to better manage impaired drivers. As such, it is imperative that research findings which provide insight into the profile and characteristics of impaired drivers, the factors that place them at risk for re-offending, and the reliability and validity of instruments to assess risk are translated to practitioners in the criminal justice, transportation and health systems to strengthen practice in the field. Moreover, knowledge about treatment interventions and best practices must be shared with practitioners to inform decision-making and to ensure that offenders are appropriately managed and treated to reduce long-term risk. INTRODUCTION 5

107 3. PURPOSE AND OBJECTIVES Research to inform the risk assessment of impaired drivers has grown in the past decade. In conjunction with this, demand for knowledge has increased as agencies seek to better utilize limited resources to effectively manage this population. Thus, it is timely to take stock of available research and knowledge about impaired drivers and their offending behaviour in an effort to inform practices that are currently applied to characterize, assess, and manage these drivers post-conviction in a way that promotes long-term risk reduction. The purpose of this report is to summarize available knowledge about the profile and characteristics of impaired drivers, relevant risk factors, risk assessment instruments and effective treatment interventions to treat impaired drivers as well as best practices in this field 4. It provides an overview of available research regarding the profile of male and female first offenders in relation to repeat offenders and highlights the inability of existing theories of behaviour to adequately disentangle the heterogeneous nature of the impaired driving offender population. It also provides a summary of relevant risk factors that have been linked to repeat impaired driving offences, while acknowledging some of the key limitations of the research in this field, and briefly reviews some of the available tools used to assess risk. Finally, it offers an overview of available treatment interventions that are applicable to impaired drivers, the research relating to their effectiveness, and current best practices for the treatment and rehabilitation of impaired driving offenders in remedial driver licensing programs. This report provides answers to the following questions: > What characteristics are associated with impaired driving? > Are there important differences between first and repeat impaired drivers? > Are there important differences between male and female impaired drivers? > What is known about risk factors associated with impaired driving? 4 There are a broad range of other policies, programs, and interventions for impaired drivers that have been developed, implemented, and evaluated in the past three decades which are beyond the scope of this report. 6 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

108 > What instruments are available to assess risk? > What treatment interventions are available for impaired drivers and how effective are they? > What is known about factors that are related to better treatment outcomes? > What best practices exist for the treatment and rehabilitation of impaired driving offenders? In addition, the report also summarizes priority gaps in offender research, gaps in intervention research, and gaps associated with the implementation of interventions in order to provide guidance for the field and to encourage ongoing collaboration between researchers and practitioners to begin to fill these gaps. The intent of this report is to provide a high level review of available knowledge that can benefit frontline practitioners working both in the remedial driver licensing system and the criminal justice system. For this reason, additional resources are provided at the end of some sections in order to afford practitioners an opportunity to review relevant research in more depth. PURPOSE AND OBJECTIVES 7

109 4. PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS A wealth of research has been conducted in the past three decades that examines the profile and characteristics of impaired driving offenders. While much of this research focuses on males and attempts to identify differences in the profile and characteristics of first versus repeat impaired drivers, some research has also included female offenders, as well as focused exclusively on female impaired drivers. This section briefly summarizes what is known about the profiles and characteristics of adult impaired drivers and draws from the research in criminology, psychology, transportation, health, addiction medicine, and neuroscience. It first examines what is known about male offenders followed by what is known about female offenders. Key dimensions that are considered include: demographic factors, personality and psychosocial factors, substance misuse including engagement in treatment, mental health, cognitive impairment, and driver and criminal history. In all of these sub-sections, distinctions are drawn between first versus repeat offenders. While there is also a wealth of research specific to young impaired drivers, this is beyond the scope of this report. However, individuals interested in more information on this topic should refer to Driving with Care Education and Treatment of the Underage Impaired Driving Offender: An Adjunct Provider s Guide by Wanberg, K.W., Milkman, H.B. and Timken, D.S. (2010) published by Sage, Thousand Oaks, CA. To help place these findings in context, it is worthwhile to highlight some of the limitations of this research, aptly described in Bud Perrine s theory of the Quick, the Caught, and the Dead (1990) and noted by Wanberg et al. (2005) in their book entitled Driving With Care: Education and Treatment of the Impaired Driving Offender (The Provider s Guide). There are three main sources of information that can inform our understanding of impaired drivers. Much of what is currently known has been drawn from observations of samples of offenders who have been caught by the criminal justice system. These so-called convenience samples of offenders are more easily studied, but are not necessarily representative of the entire 8 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

110 offender population. Some of what is known has been learned from studies of the dead, that is, those impaired drivers who have been killed in road crashes. Fatal road crashes are relatively rare events and arise from a confluence of factors (e.g., poor road design), so these drivers also are not necessarily representative of the entire offender population either. In sharp contrast, little is known about the Quick, or those impaired drivers who drink and drive, sometimes repeatedly, but are not detected. At this time, it is not known if these impaired drivers share common characteristics with their counterparts who are arrested or who die in road crashes. This is a recognized gap in the field; more information about this latter group of offenders is needed to increase our understanding of impaired driver behaviour and ways it can be prevented or reduced. 4.1 Male Impaired Drivers Demographic factors Several key demographic factors of male impaired drivers have been studied by a broad crosssection of researchers from different disciplines. Factors that have been examined include age, ethnicity, education, employment, marital status, BAC, life history, and environmental factors. There are a number of comprehensive resources that provide summaries of these factors that practitioners are encouraged to review, including Simpson and Mayhew (1991), Jones and Lacey (2001), Wanberg et al. (2005), and White and Gasperin (2006). What is known about each of these factors is described briefly below. Age. Most impaired drivers are between the ages of 20 and 45 years old with almost half of them being between the ages of 20 and 30 years old (Simpson and Mayhew 1991; Jonah and Wilson 1986; Jones and Lacey 2001; Wanberg et al. 2005). Generally speaking, drinking and driving behaviour begins to decrease substantially after the age of 45 years (Hingson and Winter 2003), though this behaviour persists in some drivers into their 60s. This aging out phenomenon is very consistent with patterns of behaviour exhibited by other criminal offenders (Nagin et al. 2008; PEW 2012). Hence, similar to other types of offences, a significant portion of the impaired driving problem is perpetrated by a subgroup of the population. Sex. Research shows that between 70% and 80% of impaired drivers are male. Studies in Canada and the United States have used several approaches including studies of arrested and/or convicted impaired drivers, studies of those in remedial program or treatment settings, and studies of fatally injured drivers in alcohol-impaired crashes (Waller 1997; Simpson and Mayhew 1991; Jones and Lacey 2001). For comparison purposes, an examination of arrest rates for all types of offences revealed similar numbers. Females accounted for only 23% of arrests for all offences in the United PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 9

111 States in 2004 (Schwartz and Steffensmeier 2007). Furthermore, the female share of arrests for most offences is less than 20% and is smallest for serious offences. An examination of incarceration rates in Canada reveals a different picture. The rate of crime among females is about one-quarter the rate among males and women account for only 6% of offenders in provincial/territorial corrections and 4% of offenders in federal corrections (Kong and AuCoin 2008). Similarly, in the U.S., the male imprisonment rate is 14 times higher than that of females and males account for 93.2% of incarcerated offenders (BJS 2012). Ethnicity. Research spanning 30 years suggests that a majority of impaired drivers are Caucasian, although there has been less research on ethnicity relative to other demographic factors such as age and sex. For example, Weisheit and Klofas (1992) compared the characteristics of impaired drivers in jail with a representative national survey of more than 5,000 jail inmates. It revealed that traffic offenders and impaired drivers were more likely to be Caucasian compared to other jail inmates. However, while ethnicity is one of the factors that is linked to impaired driving (Ferguson et al. 2002; Jones and Lacey 2001), differences between populations studied and the ways in which questions have been posed have resulted in inconsistent evidence in relation to this factor (Caetano and McGrath 2005). There is some evidence that non-white and non- Asian subgroups are overrepresented compared to their presence in the general population (Jones and Lacey 1998; Wolf and Lund 1991). Most recently, an analysis in Minnesota prison populations revealed that approximately 15% of incarcerated felony 5 driving while intoxicated (DWI) offenders (those who have had at least four prior driving while intoxicated offences in the past ten years) are American Indian, however American Indians comprise about one percent of the state s population, indicating that this population is substantially overrepresented among felony DWI offenders (T. Roy, personal communication 2012). Education. Most impaired drivers have completed elementary school and at least some high school, but the majority have no college or post-secondary education. Some studies suggest that as many as one-third of convicted offenders have at least some post-secondary education (Nochajski et al. 1993; Wilson 1991). A Manitoba study by Ambtman (1990) of participants in a remedial impaired driver program indicated that a minority of participants (less than one-fifth) had attained some level of post-secondary education. A 1996 TIRF study (Simpson et al.) concluded that impaired drivers have varying degrees of education, with the large majority having completed high school and a handful of them having some postsecondary education. To put these findings in context, in comparison to the general population in the U.S., correctional inmates report lower levels of educational attainment. An estimated 40% of state prison inmates, 27% of Federal inmates, 47% of inmates in local jails, and 31% of 5 In the United States, criminal offences are categorized as misdemeanor and felony offences. In Canada, offences are categorized as summary conviction and indictable offences respectively. 10 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

112 those serving probation sentences had not completed high school or its equivalent while 18% of the general population failed to attain high school graduation (Harlow 2003). Employment and income. Contrary to popular belief, the majority of impaired drivers are employed, although they are more likely to be unemployed relative to the general population (Wanberg et al. 2005). However, it is important to note that these offenders are more often in the lower-to-middle income range (Ambtman 1990; Wilson and Jonah 1985; Nochajski et al. 1993), and they are more apt to experience occupational instability. To place these findings in context, the income of impaired drivers is infrequently reported in studies. Moreover, when such information is reported, different income categories and time periods are used. Collectively, these differences make comparison of the findings across studies challenging. Of interest, self-report studies involving non-convicted drinking drivers suggest there are differences in income relative to convicted impaired driving offenders in that more self-reported drinking drivers declare income in excess of $60,000. While some have hypothesized that this discrepancy indicates that drinking drivers with higher incomes are better able to avoid detection by driving newer vehicles and having more disposable income to afford a private attorney, others have argued that drinking drivers with higher incomes may drive at lower BACs (Beirness et al. 1997). More research is needed to increase understanding of this issue. Marital status. Research on the marital status of impaired drivers is fairly consistent with some variations. Some studies suggest that more than two-thirds (65-75%) of impaired drivers are single, separated or divorced (Simpson et al. 1996; Wilson 1991; Nochajski et al. 1993). Thus, while approximately half of impaired drivers are in fact married (but perhaps separated), the other half are comprised of those who are currently unmarried or who have never been married. To some degree, the extent to which impaired drivers are single may also be a function of their young age, although this hypothesis has not been tested. To summarize, drivers who are either divorced or separated are overrepresented in the offender population relative to the general population. Interestingly, while in male offenders being married or in a stable relationship represents a protective factor against future impaired driving offences, among female offenders the marital or relationship status is not a protective factor but rather an aggravating one. This may arise in part because women are more often in relationships with spouses who also have alcohol problems (Brown et al. 1995). It is important to underscore that many of these studies were conducted two decades ago at a time when being married was more often equated with having a stable relationship, whereas today this may be less often the case. As such, it may be more useful and practical to consider the level of stability of any co-habiting relationship as opposed to focusing on the specific marital status of this population. PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 11

113 Blood alcohol concentration (BAC). Many impaired drivers possess BACs that are quite high relative to the legal limit in Canada of.08 (Simpson et al. 2004; NHTSA 2003). In Canada, between 1993 and 1997, the mean BAC among fatally injured drinking drivers was.17 (Mayhew et al. 1995; 1996;1997; 1998; 1999); a more recent estimate derived from TIRF s National Fatality Database 6 for revealed a mean BAC of.174, so there has been little change in this measure. In the U.S., the average BAC among drivers in fatal crashes is.18 (NHTSA 2010). There is evidence to suggest that while BAC is a good measure of level of alcohol use, it is not a reliable indicator of alcohol-related problems, involvement in impaired driving or risk of recidivism (Wieczorek et al. 1992). Repeat and/or hard core 7 impaired drivers. This segment of the impaired driver population generally has many similar characteristics to first impaired drivers, however these characteristics are often more pronounced. Sex and age. Research shows that some 90% of recidivists are male. Repeat offenders are mostly male and between the ages of 23 and 45 years. More than half of them (70%) are under the age of 40. Similar to other criminal offenders, repeat impaired drivers appear to age out of this offending behaviour beginning at age 35 with sharp declines between 40 and 50 years of age (Simpson et al. 1996). Ethnicity. While a majority of repeat offenders can be classified as Anglo-white (Jones and Lacey 2001; Wanberg et al. 2005), it has also been suggested that ethnicity is related to repeat impaired driver status, however this varies according to region. For example, more repeat offenders in northern parts of the U.S. are Caucasian whereas in the southwest the majority of offenders are Hispanic or Native American (Nochajski and Stasiewicz 2006). More research into ethnicity is needed to further validate these results. Education. Studies show that repeat offenders generally have less education than nonoffenders as well as first offenders (Simpson et al. 1996; Jones and Lacey 2001; Nochajski and Stasiewicz 2006). Employment and income. Similarly, research reveals that repeat offenders represent a crosssection of income levels, however most have moderate family incomes and lower personal incomes than first offenders, and are more likely to be unemployed (Nochajski and Stasiewicz 2006; Beirness 1997). Marital status. Repeat offenders are also more likely than first offenders to be never married, divorced, separated, or widowed (Wieczorek and Nochajski 2005; Simpson and Mayhew 1991). Also of interest, it has been reported that almost two-thirds (60%) of repeat impaired drivers have children (White and Gasperin 2006). 6 Since its inception by TIRF, the following agencies have provided funding for the Fatality Database: Health Canada (1973 to 1982); Transport Canada (1983); Transport Canada and Canadian Council of Motor Transport Administrators (1984 to 2010; their funding for the Fatality Database has been in support of the Strategy to Reduce Impaired Driving - STRID - for several years). 7 Hard core impaired drivers, also known as hard core drunk drivers are defined as drivers who drink and drive repeatedly, often at high blood alcohol concentrations, and have a history of prior convictions for impaired driving and or substance abuse problems. 12 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

114 BAC. Finally, among repeat offenders, arrests at higher BACs of.18 or over.20 are more common compared to first-time offenders (Wanberg et al. 2005) as is test refusal at the roadside (Robertson and Simpson 2002). Summary. An examination of several demographic characteristics suggests that many of these characteristics are more pronounced among repeat offenders in comparison to first offenders (Wieczorek and Nochajski 2005). That is, repeat offenders are more often single, separated, or divorced, have less education, lower levels of income, and higher levels of unemployment in comparison to first offenders Personality and psychosocial factors A wide range of personality and psychosocial factors have been examined in relation to impaired drivers including sensation-seeking, hostility, aggression, psychopathic deviance, assertiveness, antisocial personality, impulse control, risk perception, narcissistic personality, intermittent explosive disorder, external locus of control (i.e., blaming others for problems), and emotional adjustment. To illustrate, since the early 1960s, numerous studies have sought to differentiate between impaired drivers and other drivers on the basis of social, psychological, attitudinal, and behavioural characteristics including Donovan et al. (1983), Jonah and Wilson (1986), MacDonald (1989), Selzer et al. (1963), Cosper and Mozersky (1968), Yoder and Moore (1973), Meck and Baither (1980), Fine and Scoles (1974), MacDonald and Pederson (1990), Perrine (1975), and Steer and Fine (1978). These studies were reviewed and summarized in a Health Canada study (1997). The results of this review suggested that impaired drivers demonstrate higher degrees of hostility, aggression, and sensation-seeking among other factors in comparison to other groups of drivers (Beirness et al. 1997). Similarly, since 2000, there have been a number of studies that serve to reinforce these findings (Vingilis 2000; Jones and Lacey 2001; Cavaiola et al. 2003). In particular, a comprehensive review by Wanberg et al. in 2005 reported that the most salient personality variables associated with [DWI] behaviour include: agitation, irritability, resentment, aggression, overt and covert hostility; thrill and sensation-seeking; low levels of assertiveness, low self-esteem, feelings of inadequacy, and sensitivity to criticism and rejection; helplessness, depression, and emotional stress; impulsiveness, external locus of control (blame others for problems); social deviance and non-conformity, anti-authoritarian attitudes (p.23). Repeat and/or hard core impaired drivers. Studies from the early 1990s suggest there are personality differences between first and repeat offenders. For example, in a study comparing first and repeat impaired driving offenders, McMillan et al. (1992a). reported that repeat offenders demonstrated higher levels of hostility, sensation-seeking, psychopathic deviance, mania and depression as well as lower levels of assertiveness and emotional adjustment. More recently in 2005, a study by Wieczorek and Nochajski reported that repeat offenders PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 13

115 have lower levels of self-esteem, locus of control, social desirability, and higher levels of psychiatric symptoms and antisocial tendencies. A 2007 study by Cavaiola et al. based upon data derived from more indirect questions about behaviour revealed that sensation-seeking, hostility, depression, and psychopathic deviance are correlated with repeat impaired driving offences. Conversely, a 2002 study by Cavaiola and Wuth (2002), cited in a comprehensive review of the literature by Wanberg et al. (2005), noted that a majority of studies have not identified significant personality differences between first and repeat impaired drivers. Cavaiola and Wuth (2002) further suggested that this may reflect the fact that many first offenders are, in fact, repeat offenders who just have not yet been brought to the attention of the criminal justice system. While plausible, these apparently contradictory conclusions also reveal the fact the research to date has failed to adequately disentangle the significant heterogeneity observed in the impaired driver offender population. Some research (Schell et al. 2006) suggests that socially desirable responding among impaired drivers in self-report studies biases our understanding of personality and behavioural factors. Individuals who are high in socially desirability are less likely to admit to relevant behaviours including driving after drinking, drinking alcohol, aggressive driving, hostility, impulsivity, and sensation-seeking. Schell et al. concluded that the fact that personality factors are very difficult to modify in conjunction with evidence that their effects on driving after drinking are small and indirect suggests that personality factors may not be a promising point of intervention (p.39). At the same time, if reliable trait-like markers are uncovered in the future, they could serve to assist in the prediction of impaired driving recidivism and to trigger targeted selective prevention procedures Alcohol misuse The role of alcohol misuse in relation to impaired driving behaviour has been studied more than almost any other factor. However, while older research has suggested that substancerelated problems were a critical factor in impaired driving offending, more recent research has determined that, although substance use is strongly correlated with impaired driving behaviour, it is not a causal factor. This section summarizes what is known of this multi-faceted issue including: age of onset of drinking, family history, drinking patterns, substancerelated diagnoses, treatment history, and failure to enroll in or complete treatment. Age of onset of drinking. It has been well-established over the past 35 years that early onset of alcohol and other drug use are predictive of substance use and abuse in adulthood 14 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

116 (Hingson et al. 2003; 2002; Grant and Dawson 1997; Wanberg et al. 2005). Generally speaking, those individuals who begin drinking at an early age (under the age of 14) often consume more alcohol as compared to those who begin drinking in their late teens or at the age of 21 (the U.S. legal drinking age). While these individuals may or may not eventually become dependent, they are at higher risk for impaired driving. Equally concerning, these same individuals are more likely to believe that driving after drinking is only risky if individuals are obviously impaired, and may be less likely to believe that driving after drinking increases the risk for injury or crashes. In addition, persons who begin drinking before age 14 are more than seven times more likely to be in an alcohol-related crash than those who begin at age 21 (Hingson et al. 2002). Of concern, impaired drivers often report heavier drinking behaviour and involvement in binge drinking at a young age (Wechsler et al. 2003; Hingson et al. 2002; 2003). Family history. Research shows that there are two characteristics related to family history that are the most strongly associated with number of impaired driving offences as an adult. These include: having a father with a drinking problem (Schuckit 1999; 2009); and having a relative who was arrested for impaired driving (McMillen et al. 1992a; Wieczorek and Nochajski 2005). The modeling of drinking and driving behaviour within the family appears common (Elliott et al. 2006; Gulliver and Begg 2004). What family history represents in the impaired driving literature is often vague. Both genetic predisposition to alcohol abuse (e.g., tolerance for heavy drinking, euphoria vs. sedation, externalizing personality) and the social genetics of being brought up in an alcoholic environment (e.g., greater likelihood of exposure to alcohol at an earlier, neurodevelopmentally vulnerable age) are likely involved. Drinking patterns. Research investigating the drinking patterns of impaired driving offenders reveals that these individuals generally consumed greater amounts of alcohol per occasion and also consumed alcohol more often than the general population of drinkers (Beirness et al. 1997). Their drinking behaviours are also more likely to result in more alcoholrelated problems and they may consume alcohol to cope with personal or emotional issues (Wanberg et al. 2005). There is also research to indicate that a majority of impaired drivers are, in fact, binge drinkers (Caetano and McGrath 2005; Chou et al. 2006). A study by Flowers et al. (2008) indicated that 84% of alcohol-impaired drivers were binge drinkers while 88% of impaired driving episodes involved binge drinkers. These findings challenge a popular belief that alcoholism is at the root of impaired driving behaviour. Alcohol-related diagnoses. There are considerable discrepancies with regard to estimates of problem drinking and substance abuse problems across studies (Simpson et al. 1996; Wanberg et al. 2005; Vingilis 1983; Beirness et al. 1997; Baker et al. 2002; Kramer 1986; Maruschak 1999; Brinkmann et al. 2002). This may be a result of different procedures that studies have used to reach such estimates including: PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 15

117 > definitions of alcohol problems; > data reporting practices; > the populations sampled and sampling methods; and, > instruments to diagnose problem drinking and substance abuse, some of which may not have items related to impaired driving behaviour. The DSM-IV-TR identifies the criteria for two alcohol use disorders: alcohol abuse and alcohol dependence. It defines alcohol abuse as a maladaptive pattern of alcohol use leading to clinically significant impairment or distress, as manifested by one or more criteria (including recurrent alcohol use resulting in a failure to fulfill major role obligations; recurrent alcohol use in situations in which it is physically hazardous; recurrent alcohol-related legal problems; and continued alcohol use despite having persistent or recurrent social or interpersonal problems caused or exacerbated by the effects of the alcohol) occurring within a 12-month period that does not meet the criteria for alcohol dependence. Alcohol abuse is most common among youth aged 15 to 24 while 17.8% of Americans meet the criteria for this disorder at some point in their lifetime (Hasin et al. 2007). Alcohol dependence, previously referred to as alcoholism, has different symptoms than alcohol abuse, most notably tolerance and withdrawal. According to the DSM-IV-TR, it is defined as a maladaptive pattern of alcohol use, leading to clinically significant impairment or distress as manifested by three or more of the following, occurring at any point in a 12-month period: tolerance 8 ; withdrawal 9 ; alcohol is often used in larger amounts or over a longer period than was intended; there is a persistent desire or unsuccessful efforts to cut down or control alcohol use; a great deal of time is spent in activities necessary to obtain alcohol, use alcohol, or recover from its effects; important social, occupational, or recreational activities are given up or reduced because of alcohol use; and, alcohol use is continued despite knowledge of having a persistent or recurrent physical or psychological problem that is likely to have been caused or exacerbated by alcohol. During their lifetime, 12.5% of American adults meet the criteria for alcohol dependence (Hasin et al. 2007). Nevertheless, some research suggests that a diagnosis of alcohol abuse is more common among first offenders than repeat offenders, suggesting that this group may generally have lower levels of problem severity relative to repeat offenders (Wieczorek and Nochajski 2005). Attitudes about change and treatment. Research indicates that approximately only one-third of first offenders have a history of varying degrees of involvement in treatment (Wieczorek and Nochajski 2005). Moreover, many offenders, regardless of their number of prior offences, are assessed as being in the pre-contemplative stage in relation to the stages 8 Tolerance is defined by either 1) a need for markedly increased amounts of alcohol to achieve intoxication or desired effect, or 2) markedly diminished effect with continued use of the same amount of alcohol. 9 Withdrawal is manifested by either 1) the characteristic withdrawal syndrome for alcohol, or 2) alcohol (or a closely related drug such as valium) is used to relieve or avoid withdrawal symptoms. 16 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

118 of change with regard to their drinking and driving behaviour 10. This suggests that, despite the consequences of prior impaired driving convictions, a majority of offenders still fail to recognize the need to change their behaviour, or do not begin to think about changing their behaviour, much less develop the motivation to do so (Nochajski and Stasiewicz 2006; Wieczorek and Nochajski 2005). There is also research that demonstrates that impaired driving offenders may be more defensive of their drinking behaviour, and more resistant to self-disclose the extent of their alcohol consumption (BHRCS 2007) than the average patient that engages in alcohol treatment. Generally speaking, higher levels of resistance are most often observed during an initial screening and assessment level (Vingilis 1983; Wanberg et al. 2005, p.25; Chang et al. 2002; Owens 2001). As a practical consequence of their repeated experiences with offenders who fail to be forthcoming about their drinking behaviour, practitioners who conduct screening and assessment of these offenders more often tend to doubt or distrust reports of alcohol consumption disclosed by impaired drivers. However, it has been argued that clinicians should not underestimate or overlook the value of self-reported drinking by offenders. In fact, selfreports provide a good indication of an offender s perceptions of their drinking behaviour. Therefore, this information may be interpreted as an indication of the offender s self-appraisal of their impaired driving behaviour as opposed to their presumed denial or resistance. In sharp contrast, impaired drivers appear to be much less defensive to disclosing use of drugs than they are to disclosing alcohol use and associated disruptive symptoms (Wanberg et al. 2005), although the reason for this is not known. Failure to enroll in or complete remedial programs. A study by Boudreault et al. (2002) that examined 126 hard core recidivists in prison revealed that 62% had never complied with mandated remedial programs. Similarly, a Montreal study by Brown et al. (2002) reported data from Quebec that showed that just 50% of convicted impaired driving offenders had pursued participation in intervention programs that were mandated within the first year of relicensing eligibility. A more detailed examination of those offenders who delay participation conducted in 2008 by Brown et al. further revealed that more than 50% of offenders identified key reasons for this delay as being due to having made other transportation arrangements, the cost of remedial program participation, no access to a vehicle, and no interest in driving. One-third identified no interest in and/or ability to change their alcohol usage as an obstacle. 10 The transtheoretical stages of change model posits that individuals with behaviour problems, such as substance dependence, experience several conditions and differ in their willingness to acknowledge that they have a problem and work towards change (Alexander 2000). Interventions or treatment strategies are most likely to be successful when geared toward the stage of change that the individual client is in. Adapted from Prochaska et al. s (1992) readiness for change process stages, the various stages include: 1) Pre-contemplation (lack of awareness of a problem; no contemplation of change); 2) Contemplation (recognition of a problem; contemplation of change); 3) Preparation (consideration of behaviour change); 4) Action (taking steps to change behaviour such as participation in treatment); and, 5) Maintenance (relapse prevention). PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 17

119 This study further revealed that some important characteristics of non-participants in mandated remedial programs include poorer socio-economic status and disrupted neurocognitive performance in terms of memory, behavioural inhibition, and delayed reward discounting (i.e., they have lower motivation for delayed gratification and heightened attraction to immediate gratification), possibly manifested in their propensity for unlicensed driving, and reluctance to pay fines and other costs associated with relicensing and to commit to long-term behavioural change. The study further reported that the reluctance of offenders to engage in remedial impaired driving programs, was due, in part, to the fact that their personal objectives for change were inconsistent with the goal of the intervention which was to significantly reduce drinking. A comparison between impaired driving offenders who completed mandated remedial programs versus those who were non-compliant indicated that the latter group possessed the following characteristics: older, lower income in last 30 days, less likely to be married or without a partner, unemployment, similar drinking patterns, more cocaine dependence, higher proportion of positives on axis 1 disorders (e.g., anxiety, depression), and higher proportion of antisocial personality features. Logistic regression further revealed that unemployment was the main predictor of non-compliance (Nadeau 2010), suggesting that cost may be a major obstacle to increased participation among poorer offenders. Voas et al. (2010) also examined the propensity of first and repeat impaired drivers to delay licence reinstatement and the implications of this delay on their driving behaviours in seven U.S. states. In particular, this study revealed that: > Fewer first offenders (42%) delayed licence reinstatement for at least one year (from the time they became eligible) in comparison to 55% of repeat offenders. At three years post-eligibility, the proportion of offenders who reinstated were 70% and 58% respectively. It was also reported that no record of reinstatement could be located for 25% of first offenders and more than one-third (35%) of repeat offenders. > In the first year of eligibility, those offenders with prior offences were significantly less likely to reinstate their licence than first offenders. However, in the subsequent years, the length of the delay did not appear to differ between first and repeat impaired driving offenders. > Higher recidivism rates were reported during the suspension period in relation to offenders who delayed licence reinstatement for more than one year as well as offenders who failed to reinstate at any time. Similarly, post-reinstatement, offenders who delayed reinstatement also have higher rates of recidivism. 18 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

120 > It appears that recidivism rates are somewhat lower post-reinstatement among those offenders who do reinstate. In addition, recidivism rates are higher among offenders who do not reinstate in comparison to those who do. Repeat and hard core impaired drivers. Similar to first offenders, age of onset, family history, and alcohol misuse issues plays an important role in relation to repeat impaired driving offenders. It has been noted that there is a strong direct linear relationship between the total number of dependence criteria that repeat offenders present and the number of prior offences (Weiczorek and Nochajski 2005). A comprehensive review of the literature by Wanberg et al. (2005) similarly reported that repeat offenders have higher levels of disruptive alcohol use symptoms. Other studies have concluded that the incidence of problem drinking increases with the number of prior convictions (Perrine 1990; Nochajski and Stasiewicz 2006; McMillen et al. 1992a). A study by MacDonald and Pederson (1990) that examined impaired driving arrests among male, hospitalized alcoholics showed that multiple offenders were more likely to report a higher number of most drinks ever consumed in a day but less frequent drinking, a pattern indicative of binge drinking. It is estimated that more than two-thirds of second offenders and almost 90% of multiple offenders have some history of alcohol treatment involvement (i.e., alcohol education, outpatient, inpatient, or other recognized forms of treatment) (Weiczorek and Nochajski 2005). Of interest, repeat offenders appear to have a higher level of motivation for change and treatment and may be less defensive and more self-disclosing than first offenders Mental health A broad range of mental health and psychiatric conditions have also been linked to impaired driving offenders including antisocial personality disorder, anxiety, conduct disorder, impulse control disorder, narcissism, depression, post-traumatic stress disorder (PTSD), and bipolar disorder. Recognition of and interest in these factors has grown in the past decade, and even more recently as a result of the large number of soldiers and veterans that are involved in impaired driving events either overseas or upon their return to North America. A number of research studies suggest that psychiatric disorders are higher among impaired drivers (Shaffer et al. 2007; Lapham et al. 2001; McMillen et al. 1992a; Wieczorek and Nochajski 2005). In a study of offenders in New Mexico by Lapham et al. (2001), among offenders with alcohol use disorders, 33% of men had a least one additional psychiatric disorder, other than drug abuse or dependence. The most common conditions were major depression and post-traumatic stress disorder. This study also reported that 13% of men had a lifetime major depressive disorder (7% of men in the 12 months prior to the interview). A higher proportion of men (relative to women) in this study met criteria for PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 19

121 antisocial personality disorder, whereas a smaller proportion of men (relative to women) had experienced lifetime and 12-month dysthymic disorder, generalized anxiety disorder, and PTSD. Stress is also considered an important factor in relation to impaired driving behaviour. It frequently occurs in situations that exceed an individual s ability to cope with events and/or the demands made on them. It should be underscored that stress and its resulting emotional conditions (i.e., guilt, anger, depression) can significantly influence substance use given that people often rely upon alcohol and other drugs to either cope with or relieve stress and associated unpleasant emotions. Of greater concern, these emotions are closely tied to relapse and connected to negative outcomes, including impaired driving (Wanberg et al. 2005). For example, impaired drivers may experience stress due to relationship problems, financial problems, job or employment-related problems that may contribute to their alcohol and drug use and arrests, which may further compound stress. Research examining the effects of anxiety disorder in relation to substance use has also produced significant findings that may have important implications for impaired drivers. A study by Kushner et al. (2011) reported that the presence of an anxiety disorder affects the brain such that the transition from regular drunkenness to alcohol dependence is accelerated. An important factor in this process is the age of onset of anxiety disorder in relation to important drinking milestones. Men may experience shorter transition times relative to women, as women may not experience such telescoping of the development of alcohol dependence. Moreover, the study reported that the age of onset of drinking in this sample of alcohol dependent patients in a chemical dependency program was earlier for women than for men; in sharp contrast to findings from other studies of alcoholism. Drug use (other than alcohol). Many impaired drivers have substantial histories of drug use (Beirness and Davis 2008). Rates of drug use among first and repeat offenders are not only important but also are not limited to soft drugs like marijuana. One study reported that more than 40% of all impaired driving offenders in the study sample had used cocaine, hallucinogens, and amphetamines (Weiczorek and Nochajski 2005). A comprehensive review of the literature by Wanberg et al. (2005) revealed that about 11-12% of impaired drivers are multiple drug users who report significant involvement in drugs other than alcohol and marijuana; close to 50% report a history of marijuana use. A substantial percentage of impaired driving offenders reports involvement with other drugs. In a New Mexico study of these offenders, 32% of women and 38% of men had a drug use disorder (Lapham et al. 2001). Other studies of impaired driving offenders in treatment in Texas by Maxwell (2011) and Freeman et al. (2011) similarly demonstrate that a history of drug use among impaired drivers in not uncommon. To illustrate, Maxwell (2011) found that cannabis was identified as a primary problem among the youngest arrested impaired drivers 20 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

122 whereas alcohol and crack cocaine were more prevalent among older drivers. In addition, Caucasian arrestees more often had problems with other opiates, methamphetamines, and sedatives in contrast to Hispanics who more often had problems with powder cocaine and cannabis. The most recent U.S. roadside survey results showed that 31.8% of drivers at or over the legal BAC limit of.08 were positive for a drug. This result was twice as high as sober drivers (Lacey et al. 2009). To place these numbers in context, the National Survey on Drug Use and Health for revealed that, nationally, 7.3% of the population aged 12 or older was classified with alcohol dependence or abuse nationwide in the past year while individuals between the ages of 18 and 25 had the highest rate of alcohol dependence or abuse (15.9%). It also revealed that, 2.8% of persons aged 12 or older had past year illicit drug dependence or abuse which was unchanged from Again, the highest rates for illicit drug dependence or abuse in the past year were among the 18 to 25 year age group (7.8% nationally) (SAMHSA 2012). Repeat and/or hard core impaired drivers. Repeat offenders have significantly higher levels of psychiatric symptoms (Wieczorek and Nochajski 2005; Wanberg et al. 2005; Jones and Lacey 2001; Simpson et al. 1996). To illustrate, in a sample of 729 patients in a twoweek inpatient treatment facility for court-sentenced repeat impaired driving offenders (i.e., offenders electing to participate in treatment in place of prison time), Shaffer et al. (2007) found that the offenders had higher lifetime and past-year co-morbidity rates than the general population with regards to alcohol use and drug use disorders, conduct disorder, post-traumatic stress disorder, generalized anxiety disorder, and bipolar disorder. Almost half qualified for lifetime diagnoses of both addiction (i.e., alcohol, drug, nicotine, and/or gambling) and a psychiatric disorder. A recent Harvard Medical School study (Nelson et al. 2012) compared lifetime prevalence of substance use disorder diagnosis and lifetime prevalence of mental health disorder diagnosis among a sample of drivers enrolled in a two-week inpatient facility for repeat impaired drivers in Middlesex to results from a replication of the National Comorbidity Survey. A majority of repeat offenders in this study were male (81%) and more than half (62%) had two prior impaired driving offences; 24% had three priors. More than 30% of the sample was between the ages of 40 and 50 years. This study revealed that repeat offenders had a higher lifetime prevalence of drug dependence (16%), drug abuse (26%), alcohol dependence (42%), and alcohol abuse (56%) in comparison to the results of the replication of the National Comorbidity Survey which were 3%, 8%, 5%, and 13% respectively. Similarly, the repeat impaired driver sample had a much higher prevalence of conduct disorder (19%), bipolar disorder (8%), PTSD (14%), and generalized anxiety disorder (9%) in comparison to the replication of the National Comorbidity Survey which were 10%, 4%, 7%, and 6% PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 21

123 respectively. However, the repeat impaired driver sample also had a lower prevalence of major depressive disorder (12%) than reported in the survey results (17%) (Nelson et al. 2012). To date, there have been no published studies identifying the prevalence or co-morbidity of psychiatric disorders among repeat impaired drivers (Labrie et al. 2007). It has been reported that there are significant differences in drug use by the number of prior offences and persistent offenders have higher levels of use than first offenders (Wieczorek and Nochajski 2005; Wanberg et al. 2005; White and Gasperin 2006). Summary. Mental health issues among impaired drivers are an important consideration given that treatment is more difficult when individuals possess emotional and psychiatric problems in conjunction with substance-related problems (Lapham et al. 2001). Hence, not only can co-occurring disorders decrease the effectiveness of treatment, but they are also considered a predictor of poorer treatment outcomes (Lapham et al. 2001; Laplante et al. 2008; Shaffer et al. 2007) Cognitive impairment Executive cognitive function involves the set of abilities that allows one to select behaviour appropriate to a situation, including the ability to inhibit inappropriate behaviours and to focus on a specific task in spite of distraction (Brown et al. 2008, p.115). Deficits are linked to impulse control and self-regulation, capacity to learn and retain intervention content, problem solving, abstracting, and the speed of information processing, among other abilities. Research shows that a continuum of alcohol users ranging from both social drinkers to alcoholics may possess neurocognitive deficits (Parsons 1998). It has further been shown that persons who possess varying degrees of cognitive deficits are overrepresented in substance abuse programs (Teichner et al. 2002). Preliminary studies of neurocognitive characteristics of first offenders indicate that they are more likely to suffer deficits related to executive cognitive function compared to normal drivers (Brown et al. 2010; Couture et al. 2010, August). Generally, it is estimated that it may take a minimum of six months for individuals to begin to recover from reversible deficits in executive function due to alcohol (Zinn et al. 2004). In cases involving much heavier drinking for extended periods, it may take two to four years, and some individuals may never fully recover from all deficits (Parsons 1998). Repeat and/or hard core impaired drivers. Research reveals that repeat offenders are more likely to possess cognitive impairments. In a study by Glass and Chan (2000) involving 134 volunteer participants that attended one of two residential alcohol education programs in Massachusetts following a repeat impaired driving conviction, 73% were reported to possess one or more clinically significant cognitive deficits; an additional 12% were identified as borderline. The tests utilized in this study measured abilities in relation to comprehending instructions, learning, sustaining attention, and completing tasks. Scores below the 50th 22 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

124 percentile in relation to tests of word fluency, vocabulary, sustained attention, memory, executive functioning, and impulse control were common among a majority of participants. The most prevalent impairments related to planning, foresight, and impulse control. A study of sober recidivists by Ouimet et al. (2007) revealed that more than half (66%) of participants showed impaired performance on at least one test of a battery of tests of executive control functioning. In particular, deficits were identified in visuospatial constructional abilities, visual memory, and inhibition capacities. The study further noted that the neurocognitive functioning of impaired driving recidivists was disproportionately lower to that of the general population and that the number of past convictions was related to severity of memory difficulties. More recent investigations have revealed more disadvantageous decision-making in sober recidivists compared to normal drivers. One study (Maldonado- Bouchard et al. 2012) indicated that intractable impaired driving behaviour, as measured by frequency of past offending, was strongly associated with the propensity to exercise more disadvantageous decision-making. This involved repeatedly choosing smaller but immediate gains despite the greater risk of suffering larger, long-term losses (e.g., the decision to choose the convenience of driving to a drinking venue despite increasing the probability of having an impaired driving-related crash or citation). This relationship was independent of the severity of alcohol misuse. Another study (Brown et al. 2008) revealed that poorer executive control functioning in impaired drivers was associated to a greater tendency to delay or avoid participation in remedial relicensing programs. Overall, these findings indicate that the most intervention-resistant offenders have a decreased ability for self-regulation, for learning and retaining intervention content, and for exercising good decision-making even when sober. Not all of these problems are attributable to alcohol abuse severity. This suggests that new strategies in the design of remedial programs and interventions directed at some offenders with the highest risk of recidivism may be needed. Effective interventions for these individuals may need to target not only a reduction in their substance misuse, but also alteration in the decisions they make prior to drinking and driving (e.g., the decision to take their car to a drinking venue) Driver and criminal history Research has demonstrated that a significant proportion of impaired driving offenders may also have a history of other driving violations as well as other criminal history. In particular, the propensity for other driving and criminal offences appears to be more pronounced among repeat offenders (Simpson et al. 1996; Jones and Lacey 2001; Syrcle and White 2006; Wieczorek and Nochajski 2005). The relevant research is described in more detail below. PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 23

125 Driver history. Impaired drivers are less likely to use seatbelts, have more traffic violations, more involvement in crashes, and have often been compared to high risk drivers 11 (Wilson and Jonah 1985; Wilson 1992; McMillen et al. 1992a; Gebers and Peck 2003). Impaired driving offenders demonstrate a range of poor driving behaviours and/or involvement in road crashes (Beirness et al. 1997; Jones and Lacey 2001; Labrie et al. 2007; Taxman and Piquero 1998; McMillen et al. 1992b), which may also bring them into contact more frequently with police and increase the potential for an impaired driving arrest (Nochajski and Stasiewicz 2006). To summarize, impaired driving is likely not an isolated high-risk driving behaviour in some offenders, meaning that some individuals who drive while impaired may also have a history of other unsafe and/or high-risk driving behaviours (Beirness et al. 1997). Nevertheless, reliance solely on driving records to identify these drivers is problematic. To illustrate, information contained in official criminal or driving records has not permitted accurate prediction of prior impaired driving offences (Simpson and Robertson 2001; Nochajski and Stasiewicz 2006). Criminal history. The higher prevalence of criminal arrests for other offences among impaired driving offenders was first identified by Waller (1967). Later, in a study by Gould (1989) it was suggested that impaired driving is not an isolated incident of bad behaviour, but instead part of a continuing pattern of criminal activity. The research was based on an archival review of the prior criminal histories for people arrested for impaired driving in Louisiana and revealed that there is a substantial difference in the number of arrests for the group of individuals with an impaired driving arrest as compared to a random sample of all male licensed drivers in Louisiana. A 1992 study by Weisheit and Klofas examined the criminal history of impaired drivers in comparison to a large sample of other jail inmates. It revealed that impaired drivers are as likely to have prior arrests as other jail inmates, and many have substantial histories of property and violent offences. More recently, a study was conducted to analyze the past criminal histories of first impaired driving offenders in California (CA), Florida (FL), and New York (NY) in order to gauge whether there were common prior offences. Analyses of these data consistently revealed that between 26% and 44% had been engaged in criminal activities prior to their impaired driving arrest. Two of these states (CA and FL) included traffic offences and reported that more than one in three people had a prior arrest history for other offences at the time of their first impaired driving arrest (Caldwell-Aden et al. 2009). Results of this study also revealed that drug offences, assault, and theft offences were the most common reasons for arrest prior to the impaired driving offence among those with criminal histories in these states; and between 45% and 85% of those with a prior arrest had also been arrested for at least one of these three offences (Caldwell-Aden et al. 2009). 11 High risk drivers refer to a small population of drivers who repeatedly engage in a variety of hazardous or dangerous driving behaviours such as speeding, red light running, drinking and driving and non-use of seatbelts, and who are more resistant to traditional interventions and sanctions. 24 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

126 Similarly, a study by Syrcle and White (2006) reported that there were also differences between first and repeat offenders in relation to previous charges for controlled substances however these differences were less pronounced (11% vs. 19%). These studies illustrate that at least a portion of convicted impaired drivers have a history of other criminal offences and suggest that strengthening linkages between the criminal justice system and impaired driver treatment programs may be beneficial. As evidence of this, the past decade has seen a clear emergence of interventions that are working to bridge this gap, including specialty problem-solving courts 12 in which the results of alcohol and/or drug screens are made available to justice practitioners. Similarly, alcohol monitoring technologies delivered through either the criminal justice system and/or the administrative driver licensing system are also increasingly linked (albeit to varying degrees) with alcohol education programs and treatment services. This permits data from alcohol monitoring devices to be shared with both criminal justice and treatment professionals. A comprehensive review of the literature by Wanberg et al reported that antisocial and criminal conduct was more prevalent among impaired driving offenders compared to normal drivers. In particular, they tend to have greater involvement in the justice system for other offences than impaired driving and engage in socially acceptable behaviours less often. They further report more psychosocial disturbed problems, reluctance to comply with court mandates and frequent under-reporting of criminal conduct, and higher rates of traffic violations than the general population (Wanberg et al. 2005, p. 25). Repeat and/or hard core impaired drivers. Repeat offenders are also more likely to have more traffic offences and to have been involved in crashes more frequently than drivers that are convicted of a first impaired driving offence (McMillen et al. 1992a; Nochajski and Wieczorek 2000; Wieczorek and Nochajski 2005) according to official records and/or selfreport. In addition to having more driving violations and problems, repeat offenders also have a more pronounced history of involvement in road crashes, injuries, and fatalities (Simpson et al. 1996; Wanberg et al. 2005; Nochajski and Stasiewicz 2006). With regard to criminal history, involvement of repeat impaired drivers in a wide range of other criminal offences (including both property and personal injury offences that are prosecuted by both summary conviction and indictment) is also more frequent (Argeriou et al. 1986; Nochajski and Stasiewicz 2006; Syrcle and White 2006; Wanberg et al. 2005). A study by McMillen et al. (1992a) reported that non-traffic arrests for both misdemeanours and felonies (i.e., summary conviction and indictable offences) were substantially higher among repeat offenders, with the frequency of non-traffic arrests being three times greater. Similarly, Applegate et al. (1997) found that repeat offenders are more likely to be re-arrested for other crimes, a new alcohol-related offence, or a new impaired driving offence compared 12 Specialty problem-solving courts such as DWI courts and drug courts are more widespread in the United States than in Canada. For more information about these courts please see the National Association of Drug Court Professionals and the National Center for DWI Courts at and PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 25

127 to those with only one prior impaired driving conviction. A review by White and Gasperin (2006) reported that approximately 20-25% of prior convictions among repeat impaired drivers involved crimes against persons. Syrcle and White (2006) reported that a larger percentage of repeat offenders had prior sentences that involved a period of incarceration relative to first offenders, suggesting the more serious nature of their offending history. Finally, a Massachusetts study by Labrie et al. (2007) examined 1,281 repeat offenders that opted to participate in a treatment program in lieu of a period of incarceration. It revealed that more than half (61%) of participants had criminal histories that involved substancerelated crimes only and more than one-third had a more extensive criminal history. Among this one-third, almost half (45%) had committed only property crimes, one-fifth (22%) had committed only crimes against persons, and one-third had histories that involved property and person-related crimes. It further noted that more severe criminality (moving from substance-related to property crimes to crimes against persons) was related to higher levels of recidivism. Results showed that participants involved in property crime were 1.4 times more likely to be re-arrested for impaired driving, and participants involved in crimes against persons were twice as likely to recidivate relative to those participants with a history of impaired driving only. The study also found that participants with less prior involvement in crime responded better to treatment whereas those with more criminal involvement did not respond as well. 4.2 Female Impaired Drivers For several decades, road safety research has demonstrated that fatalities and injuries related to road crashes (due to alcohol or other unsafe driving behaviours) have predominantly involved males (Mayhew et al. 1981; Beirness and Simpson 1988; Mayhew and Simpson 1990; Mayhew et al. 1990; Kelley-Baker and Romano 2010). Similarly, impaired driving has also predominantly been considered a male-based problem (Waller 1997; Simpson and Mayhew 1991; Jones and Lacey 2001). To illustrate, men and young adults are more likely than women or older age groups of drivers to self-report drinking and driving behaviour, to be arrested for impaired driving, or to be fatally injured or to fatally injure others while driving impaired (Mayhew et al. 2003; Zador et al. 2000). In Canada, females accounted for less than 10% of fatally injured impaired drivers prior to the 1990s. This increased slightly between 1991 and 2001 and ranged from 10-12%. Since 2002, females have accounted for 13-16% of fatally injured impaired drivers, reaching a high of 16.4% in 2006 (TIRF 2012). However, this percentage seems to have stabilized in the 26 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

128 past four years, and, overall, females continue to account for a minority of this population. By contrast, in Canada, the impaired driving rate for females generally declined up to 1997 and remained stable through to It has increased since 2005 and in 2011, females accounted for one in every six impaired drivers, compared to 1 in 13 in 1986 (Perreault 2013). A similar picture emerges using U.S. data. An examination of alcohol crash data from the U.S. Fatality Analysis Reporting System (FARS) indicates that the involvement of female drivers in alcohol-impaired road crashes has remained fairly stable with incremental increases. Females accounted for 12% of alcohol-impaired drivers in the 1980s, 13% in the 1990s, and 14% in the 2000s. Since 2006, the percentage of women drivers who tested positive for any amount of alcohol in fatal crashes has averaged 16% annually, while in ,837 fatalities in crashes involved an alcohol-impaired female driver (NHTSA 2009). By contrast, there is mounting evidence to suggest that impaired driving arrests for women in the U.S. have risen in the past three decades (NHTSA 2009; Schwartz and Steffensmeier 2007). To illustrate, in 1980, just 9% of those arrested for impaired driving were female; this percentage rose to nearly 15% by 1996 and 20% by The number of female impaired driving arrests in the U.S. rose nationally by 28.8% between 1998 and 2007 (Lapham et al. 2000; Schwartz and Rookey 2008). Thus, while in the 1990s it was estimated that about 10% of impaired drivers were female, as of the 2000s it has been estimated that women account for closer to 20% (Wanberg et al. 2005; Schwartz and Rookey 2008). Since the increase in female impaired driving behaviour first garnered attention in the late 1980s (Underhill 1986; Argeriou 1986), there have been three main hypotheses regarding contributing factors. These explanations have centred on changes in female roles in society (Popkin 1991; Bergdahl 1999; Mayhew et al. 2003; Robertson et al. 2011a; Tsai et al. 2008), changes in social norms (Gudrais 2011; Popkin 1991), and changes in social control mechanisms (Farrow and Brissing 1990; Robertson et al. 2011a; Schwartz and Rookey 2008; Schwartz and Steffensmeier 2007). One of the historical challenges associated with better understanding the characteristics of female impaired drivers has been the smaller number of them who are detected, arrested, and convicted for impaired driving, as well as the smaller number of female offenders who reoffend or recidivate (although their rate of recidivism following a first conviction is equivalent to males). Generally speaking, this has resulted in making it more difficult to conduct research on this sub-group of the impaired driver population (Moore 1994). While data on the characteristics of female impaired drivers has increased in recent years, much more research has been conducted on populations of females who consume alcohol generally, and not all of this research is specific to impaired driving offenders. Most recently, in 2013 a series of case studies were conducted with more than 150 convicted female impaired driving offenders who participated in interview focus groups in four U.S. PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 27

129 states (California, Michigan, Missouri and New York) (Robertson et al. 2013). In particular, three distinct profiles of female impaired drivers emerged, and it is estimated that more than three-quarters of the study participants matched one of these three profiles which are described in more detail below. Young women. It is estimated that at least one-quarter of the study participants were women under the age of 25, some of whom had accumulated multiple impaired driving offences in a rather short period of time. In fact, one participant had served one year in prison following her fourth offence at the age of 24. These young women reported that they did most of their drinking in bars or at house parties and that they had attempted to drive home from those locations when they were arrested. They often reported drinking to relax, to feel comfortable, or to fit in in social settings. Moreover, many of them reported that they felt pressure to keep up with male friends or boyfriends in terms of the amount of alcohol that they consumed. Young women who had grown up in a stable home environment also reported drinking in order to cope with the high expectations of family members and what they perceived as the pressure to succeed. Daily alcohol consumption and binge drinking was not uncommon among this subgroup and this is consistent with research findings identifying binge drinking among college-age women as a phenomenon of growing concern (CDC 2013). These women tended to be single or had a partner who also drank heavily and facilitated and/or encouraged their use of alcohol. Recently married women with children. This group of female impaired driving offenders reported that their drinking did not become a problem or take off until after their children were born. In some instances, these women suffered symptoms of postpartum depression and drank as a coping mechanism or as a result of feelings of isolation and loneliness. Much of the alcohol consumption occurred with family or friends at home (e.g., they would drink while they did household chores, while on the phone, or with friends or their partner). If a spouse was present, more often than not, they would also drink heavily which in some cases led to incidences of domestic violence. Of note, most of the women who fit into this profile stated that they did not have a drinking problem prior to entering into the relationship with the partner who abused alcohol and/or prior to the birth of their children. The circumstances that led up to the arrest of these women were often characterized by running errands close to the home such as picking up their children from school, buying groceries, or going to get gas. Many of these women were convicted of felony impaired driving offences on account of their children being passengers in the vehicle at the time of their arrest (this was especially common in New York due to the passage of Leandra s Law 13 ). While a majority of the women acknowledged that they were aware that they should not be driving after drinking with their 13 Leandra s Law was passed in This law made any DWI conviction where a child 16 years of age or under was present in the vehicle at the time of the arrest a felony. This law also provided for mandatory ignition interlocks for a minimum period of six months for all misdemeanor and felony DWI convictions. 28 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

130 children in the vehicle, it was often perceived as the only or the safest option (e.g., they were the more sober partner or childcare was not available). Divorced older women and/or empty nesters. Women who were not convicted of impaired driving until later in life typically reported that they developed a drinking problem in their late 30s or early 40s. Catalysts for their drinking included divorce or failed long-term relationships, shared custody arrangements or grown children leaving home, or parental illness/death. These women most often drank at home when they were alone and reported depression or feelings of isolation. Some of these women also reported drinking to feel comfortable in social settings, such as bars, because it had been a very long time since they had engaged in social activities of this nature. In particular, the women who fit this profile reported that they had more intense feelings of embarrassment and shame as their children were old enough to appreciate the stigma associated with their offending behavior, and in some cases, were also called to bail them out of jail following the arrest. While it is estimated that a small minority of participants did not fit into one of these three profiles, a majority of them possessed many of the characteristics frequently reported in the scientific literature including failed relationships, mental health problems, history of alcohol misuse within the family, multiple impaired driving arrests, history of trauma, and feelings of shame, guilt, and embarrassment. This section summarizes what is known about female impaired drivers. Key characteristics that are considered include: demographic factors, substance misuse, mental health, and driver and criminal history. Given that there has been much less research on females as compared to males, what is known about female offenders generally is summarized first, and what is known specifically about female repeat offenders is summarized at the end of this section Demographic factors Age. The average age of female first impaired driving offenders is 31 and the average age of recidivists is 30, although this fact is drawn from older research (Shore and McCoy 1987). Most recently, U.S. data from the Federal Bureau of Investigation (FBI) Uniform Crime Reports (UCR) in 2009 reveal that there were 860,689 men were arrested for impaired driving, compared to 251,695 women. Of the total impaired driving arrests for females, almost 28% were aged and almost 18% were aged In addition, women aged accounted for 12% of arrests; ages were 11%; and, ages and were 10% each. Finally, women aged 50 and older accounted for 11% of impaired driving arrests (FBI 2010). Robertson et al. (2013) also found that female impaired driving offenders ranged in age from late teens to mid-60s, suggesting that women of all ages drink and drive. However a majority PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 29

131 of participants were an estimated 20 to 40 years of age. The authors also noted that the number of college-aged women present in each of the interview focus groups was higher than expected and accounted for perhaps one-quarter of participants. Generally, rates of involvement in alcohol-impaired motor vehicle crashes decrease with age, and the population of greatest concern is often young females (Peck et al. 2008). In particular, the increasing involvement of young women with alcohol, in combination with their inexperience driving and their growing propensity for risky driving (Lynskey et al. 2007; Tsai et al. 2010) warrants our attention and further research. Education and employment. The literature regarding levels of education and employment among female impaired drivers is inconsistent. Some studies from the 1980s and 1990s indicate little difference in the levels of education between male and female impaired driving offenders (Chalmers et al. 1993). Conversely, a study of 274 women and 3,151 men convicted of impaired driving offences and ordered to attend a safety action program in Mississippi between 1976 and 1979 revealed that female offenders had higher levels of education and were older compared to men (Wells-Parker et al. 1991). Similarly, the study by Robertson et al. (2013) reported it was estimated that more than three-quarters of the study participants reported having completed high school or their General Equivalency Diploma (GED) and at least one-third of these women also reported having initiated and/or completed some type of post-secondary education to obtain a professional degree, licence, or certificate. Additionally, a comprehensive review of impaired driving studies focussing on females concluded educational underachievement is part of the pattern of risk for [impaired] driving for both young men and young women (McMurran et al. 2011, p.918). With regard to employment, a study of the Drinking Driver Program (DDP) in New York in 1992 reported that, of 800 female impaired driving offenders aged almost 70% were employed full-time and had at least some college education (Parks et al. 1996). Conversely, Chang et al. (1996) reported lower rates of employment for female impaired driving offenders compared to males. A 2008 study of 729 repeat impaired driving offenders participating in a residential education and treatment facility in Massachusetts reported that female offenders had more education than males though their level of income was lower (Laplante et al. 2008). In a study by Robertson et al. (2013), female impaired drivers reported a wide range of occupations such as nurses, dental assistants, paralegals, teachers, corporate employees, self-employed entrepreneurs, and bartenders. Of interest, approximately one-third of participants had worked in bars and restaurants at some point in their employment history. To summarize, female impaired drivers are generally older than men and have higher levels of education (Peck et al. 2008) but lower paying jobs (Chalmers et al. 1993; Shore and McCoy 1987). Low academic achievement in young females represents a risk factor for impaired driving comparable to that observed in males (McMurran et al. 2011). 30 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

132 Marital status. Research into the marital status of female impaired drivers has produced more consistency, showing that female impaired drivers, when compared to male impaired drivers, are even more likely to be divorced or single (McMurran et al. 2011; Chang et al. 1996; Shore and McCoy 1987; Argeriou et al. 1986). A study of the Drinking Driver Program in New York in 1992 that involved 800 female impaired driving offenders aged reported that 44% of females had never been married. It further noted that females who were not yet diagnosed (in relation to alcohol issues) were more often married (64%) than those in the abuse group (52%) or dependent group (55%) (Parks et al.1996). In 2000, a study by Lapham et al. reported that female impaired driving offenders, when compared to male offenders, were less likely to be married, to have prior impaired driving convictions, or to be referred for treatment. In 2013, Robertson et al. reported that the lack of stable and supportive relationships among women was a common characteristic of the women in their study. It was estimated that more than one-half of women were single, separated, or divorced at the time of the study, and approximately one-quarter of women were currently in a relationship. Of those involved in a relationship, the majority of women reported having a partner or spouse who drank frequently and/or had a drinking issue whereas a minority of women reported having a sober, healthy relationship. It has been suggested that the higher divorce rate among female impaired drivers compared to males may indicate that relationship failure has had a stronger impact on the drinking behaviour of females compared to males. This has been linked to the possibility that females have a more internalized response to stress than males, such as alcohol or drug use, which can increase their risk of other dangerous behaviour such as impaired driving. McMurran et al. (2011) concluded that females, distressed by their marital situation, may turn to alcohol as a coping mechanism. However, it should be noted that females in a relationship were most likely to be living with someone who had an alcohol problem (McMurran et al. 2011). To summarize, a significant proportion of female impaired drivers are single, divorced, or separated, or are more likely to be living with a partner with an alcohol problem compared to women with no past impaired driving offences (McMurran et al. 2011; Chang et al. 1996; Shore and McCoy 1987; Argeriou et al. 1986). Generally speaking, female impaired drivers are more likely to be the primary caretaker of children at the time of arrest, are more likely to have experienced abuse, and are more likely to have physical and mental health needs compared to their male counterparts (Bloom et al. 2003) Personality and psychosocial factors In contrast to the availability of research examining this issue among male impaired drivers, there have been fewer studies examining the prevalence of personality and psychosocial factors among female impaired drivers. However, a few studies shed some light on this issue. A U.S. study by Moore (1994) involving 180 young women (aged 16-20) with an impaired PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 31

133 driving conviction revealed that antisocial females represented just 19% of the sample. However this subset reported the majority of incidents of binge drinking, intoxication, impaired driving, crash involvement, and drug use (other than alcohol). This group of females also indicated higher levels of psychosocial stress (e.g., boredom, problems at home and school), however they viewed this as a part of daily life and failed to recognize the influence of these stressors. In sharp contrast, the women in the sample who were diagnosed as neurotic experienced similar stressors but were upset by them. The results of this study illustrated that, as is the case with a population of male impaired drivers, there are also subgroups among female impaired drivers. A smaller study by Lex et al. (1994) involving female impaired drivers in prison suggested that adult onset of antisocial personality disorder in conjunction with substance abuse may manifest differently in females as compared to males. A study of convicted female impaired drivers in New York by Parks et al. (1996) also examined this issue and reported females who were diagnosed as alcohol dependent possessed higher levels of sensation-seeking and hostility, were more anxious and depressed, had less external control, and were less interpersonally competent. In 2007, a study by Maxwell and Freeman examined differences between men and women convicted of impaired driving and who entered a public substance abuse treatment facility in Texas between 2000 and This study revealed that women were more likely to receive diagnoses of depression or bipolar disorder, and were also more likely to have prescription medication for a mood disorder in comparison to men. In summary, a review of these studies suggests that psychosocial problems among female impaired drivers may not be uncommon and that, at least a portion of these women may experience depression, boredom, and problems at home and school that are related to their drinking (McMurran et al. 2011) Alcohol misuse Alcohol use among women is a very important factor to consider in relation to impaired driving for several reasons. Research shows that women metabolize alcohol differently than men. Women initially metabolize much less (only about one-quarter as much) alcohol in the stomach and intestines as compared to men. This means that more alcohol is absorbed into the blood as ethanol, which is then available to pass through the blood-brain barrier (Gudrais 2011; Greenfield 2002). In addition, females generally have less water in the body and a lower body mass. These factors work to intensify the effects of alcohol for women relative to men and, as a result, women become intoxicated after drinking half as much alcohol (Greenfield 2002; Waller 1997; Lex et al. 1991). Of equal concern, these physiological differences also contribute in part to the more rapid progression of alcohol dependence such 32 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

134 that women often require medical intervention an average of four years earlier than males who are problem drinkers (Gudrais 2011). It is also important to note that a study by Elliott et al. (2006) found that women are more vulnerable to all types of traffic incidents following alcohol consumption. It reported that there were significantly stronger associations between women s alcohol use/misuse and crashes, and their marijuana use and offences, than among men (p. 259). Research has also demonstrated that women arrested for impaired driving and female drivers testing positive for alcohol in fatal crashes were less intoxicated than their male counterparts (Popkin et al. 1988; White and Hennessey 2006). Women who had a BAC over.05 were found to be twice as likely as men to be involved in a motor vehicle collision (Elliott et al. 2006). As such, it appears that at any given BAC, women have a higher crash risk than men, making alcohol use and driving an issue of particular concern in women. Age of onset. Studies show that youth who became drunk for the first time at a younger age (as compared to those who were drunk for the first time at 19), were more likely to drive after drinking, drive after five or more drinks, and ride with a driver who was high or drunk (Hingson et al. 2003, p.27). Even more worrisome was that those who were first drunk at a younger age also believed that they could drink more while still driving both safely and legally (Hingson et al. 2003). While research shows that females tend to consume less alcohol than males (Jones and Lacey 2001; Greenfield 2002), more recently the onset of drinking and heavier drinking among females is occurring at an earlier age, and the gender gap between young females and young males in relation to alcohol dependence is also shrinking (Greenfield 2002; Robertson et al. 2011a). Robertson et al. (2013) reported that the extent of substance use varied substantially across study participants. It is estimated that almost one-half of women reported early onset of drinking with many experimenting with alcohol and/or drugs in their early or mid-teen years; the lowest reported age of onset drinking was nine years old. In many cases, they indicated that their first exposure to alcohol and drugs was either in their own home, with relatives, or with friends. Conversely, it is estimated that between one-quarter and one-third of women did not begin to regularly use or develop a problem with alcohol or drugs, or begin to drive after using these substances, until they were in their 30s or 40s. Family history. A constellation of family history factors are associated with female impaired driving offending to varying extents, however the specific influence of each factor is unclear. Many female impaired driving offenders who were admitted to addiction treatment in Illinois possessed multiple characteristics that potentially contributed to their alcohol consumption. These included a history of alcoholism within the family, experience with abuse, anxiety and depression, and family and personal relationships that encouraged heavy drinking (White and Hennessey 2006). PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 33

135 Past and current literature has noted that there are a range of individual, family, environmental, and social factors that can contribute to the increased risk of drinking and driving among youth. Risk of general traffic offences and collisions has been correlated to substance abuse, poor school performance, lack of parental involvement, and other risky behaviours (Elliott et al. 2006). Similarly, close contact with family members who had problems with alcohol was associated with a higher risk for alcohol problems among female impaired driving offenders, as well as an increased recidivism risk (Lapham et al. 2000). Equally concerning, when offenders in treatment return to a family environment that lacks sources of support, they are more likely to repeat their pattern of alcohol and/or drug abuse (Maxwell and Freeman 2007). Most recently, Robertson et al. (2013) revealed that the reported family history of women who participated in their study varied considerably. It was estimated that slightly more than half of women reported a history of dysfunctional family relationships combined with prevalent alcohol and drug use and/or abuse to varying degrees. Alcohol-related diagnoses. Estimates of alcohol diagnoses among female impaired drivers vary but are significant and comparable to males. To illustrate, a five-year follow-up study of convicted impaired driving offenders revealed that 85% of female offenders (compared to 91% of male offenders) were diagnosed with either alcohol abuse or alcohol dependence (Lapham et al. 2000). In contrast, a study of admissions of impaired driving offenders (who were either on probation for impaired driving, were referred to treatment by probation, or had been arrested for impaired driving in the past year) to publicly funded treatment in Texas between 2000 and 2005 found that women were more impaired and experienced more problems than their male counterparts (Maxwell and Freeman 2007). In addition to the finding of dependence on sedatives and other opiates among women, this study also found that there was a shorter period of time between the first use of these drugs and admission to treatment among women compared to men, which speaks to the addictive potential of these drugs (Maxwell 2011). Of course, sampling may introduce an important bias in interpreting these findings since individuals in alcohol treatment are often alcoholic, and these findings are consistent with the alcohol literature in relation to clinical samples. However, while some of the treatment population may be alcoholic or dependent, not all of them can be assessed as such. Differences in the populations, instrumentation, related procedures, and interpolation, along with various jurisdictional policies may explain the disparate findings. In the study by Robertson et al. (2013) it was estimated that study participants equally reported patterns of daily drinking or binge drinking. Approximately one-quarter of participants reported drinking heavily for a brief period which was followed by an extended 34 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

136 period of sobriety that could last several months. A universal theme that emerged in all of the interview focus groups was that women reported that they drank for emotional reasons, or that alcohol consumption was a coping mechanism to help them manage their emotions and stress. These studies demonstrate that a substantial proportion of female impaired driving offenders are experiencing substance abuse problems, and that the gravity and complexity of those problems are significant (White and Hennessey 2006). There is also some evidence to suggest that female alcohol consumption in general may be a result of issues specific to women such as their tendency to act as caretakers, sometimes to the exclusion of their own needs (Gudrais 2011, p.10) Mental health Findings indicate that there is a need to treat some female impaired drivers not only for alcohol misuse problems but mental health problems as well (McMurran et al. 2011). A recent study by Freeman et al. (2011) of impaired driving and non-impaired driving patients in substance abuse treatment in Texas between 2005 and 2008 found that both were more likely to be diagnosed with mental health problems and more likely to be placed on medications upon admission to treatment compared to males. Female impaired driving offenders have significantly higher psychiatric co-morbidity relative to their male counterparts (Laplante et al. 2008). Diagnoses of anxiety, depression, and post-traumatic stress disorder (PTSD) are common among female impaired driving offenders. Wanberg & Milkman (2008) reported that, in a study of 10 large judicial samples including impaired driving offenders, non-impaired driving probation clients, and non-impaired driving offenders sentenced to prison (total N=15,910), in every sample, female offenders scored significantly higher (with moderate effect sizes) on the scales in the Adult Substance Use Survey-Revised (Wanberg 2006) and the Adult Substance Use and Driving Survey-Revised (Wanberg and Timken 2006) that measure mood and psychological adjustment problems. The Level of Supervision Inventory-Revised (LSI-R: Andrews and Bonta 2003) was available for eight of these samples. In all of the samples female offenders scored significantly higher than males on the LSI-R Mental Health Scale. Mental health issues were also frequently reported by female impaired drivers in a recent U.S. study by Robertson et al. (2013). It was estimated that three-quarters of the study participants reported using one or more prescription medications for disorders such as anxiety, depression, PTSD, bi-polar disorder, and schizophrenia. A small number of participants further acknowledged sexual assaults, domestic violence or abortions as influencing their mental state, and some also indicated prior suicide attempts, suggesting that histories of trauma are not uncommon. PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 35

137 Drug use. While drug use among female impaired drivers is prevalent, some research suggests that involvement in drug use may be more comparable among males and females. In a study of 812 female offenders in the New York Drinking Driver Program conducted by Parks et al. (1996) in which 43% of offenders were alcohol abusers and 25% were alcoholdependent, among these two latter groups, 19% and 50% respectively, also reported drug problems. Similarly, a study of 1,105 impaired driving offenders in New Mexico found that of those with alcohol use disorders, 32% of females (compared to 38% of males) also had a drug use disorder (Lapham et al. 2000). However, Maxwell and Freeman (2007) reported that the use of illicit drugs was higher among females as compared to males. More recently, a study examining the characteristics of convicted impaired drivers in treatment found that females were most likely to be diagnosed with a primary problem with sedatives or opiates, whereas males were most likely to be diagnosed with a primary problem with alcohol and cannabis (Maxwell 2011). Similarly, Freeman et al. (2011) found that females were more likely than males to have problems with methamphetamines, cocaine, and opiates. More recently, Robertson et al. (2013) reported that, although prescription drug use was common, less than one-third of female impaired drivers reported use of illicit substances. Among many of these women, marijuana and methamphetamines were the most common drugs of choice, although use of cocaine, hashish, and ecstasy was also reported. Often the drug use was connected to the presence of a partner or spouse who also used drugs. There was also a very small minority of focus group participants who reported that alcohol was not their drug of choice and that they did not have a problem with alcohol. In summary, given that the use of drugs appears to be somewhat common among female impaired drivers, it is important that female offenders are appropriately screened, identified, and treated for all drug use disorders Cognitive impairment While there has been limited research investigating cognitive deficits specifically among a female impaired driver population, a recent five-year longitudinal study related to predictive role of executive function in DWI recidivism was conducted by Brown, T. et al. (2013). This study involved a sample of 225 community-recruited first-time impaired driving offenders that included 136 males and 87 females, with a control group comprised of 79 individuals (37 males and 42 females). In particular, the study found that female offenders, not male offenders, showed more signs of poorer executive control compared to their non-offender controls, both functionally and psychometrically. Overall, it revealed a pattern of reduced executive control in female first-time impaired driving offenders with alcohol use disorder, and that, among alcoholics, females are more susceptible to alcohol s neurocognitive effects than 36 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

138 males. No differences were found between male and female offenders on relative measures of abuse even though males typically use and misuse alcohol more than females. A cross-sectional analysis suggested a greater role for executive control deficits in the transition from first-time impaired driving to repeat impaired driving status in male offenders than in female offenders. However, the MMPI-Mac Scale (a measure of cognitive impairment, reward-seeking, and externalizing personality features associated with alcoholism) distinguished female first-time impaired driving offenders from controls, but not in males. The authors further noted that executive control appears to be a feature of female first impaired driving offending and that their ability to identify goals, plan, execute, inhibit old behavior patterns, and learn from experience is reduced. These impairments worsened with alcohol intake. As such, alcohol appeared to contribute to female first impaired driving offending through acute and chronic disruption of executive control functioning (Brown et al. 2013) Driver and criminal history There are limited data to suggest that a smaller number of female impaired driving offenders relative to males have a history of other traffic offences or criminal offences, although more research into this topic is needed. Common criminal offences in females may include drug offences, theft offences, and assault (Caldwell-Aden et al. 2009). This finding was echoed in the study by Robertson et al. (2013) who estimated that less than 20% of participants reported prior involvement in other criminal activities in addition to their impaired driving arrest(s). Most often, their involvement in criminal activities was linked to an existing relationship with a male partner, or a group of friends engaged in criminal activity Repeat female impaired drivers Female repeat impaired driving offenders often share similar characteristics to their male counterparts. Older research suggests that repeat female offenders are approximately 30 years old but more current research on this issue is needed. Similar to males, there is also evidence that this population has lower levels of education, employment, and income, and is much more likely to be single, separated, or divorced than first offenders. Like their male counterparts, repeat female offenders are more likely to drink more frequently and exhibit higher levels of impairment, more often abuse drugs, and utilize treatment services (Argeriou et al. 1986). However, there are some differences between female and male repeat offenders. For example, repeat female impaired driving offenders have higher levels of psychiatric comorbidity than male repeat offenders and are more likely to also use drugs (Laplante et al. 2008; Maxwell 2011). PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 37

139 Recidivism rates among male and female impaired drivers show some consistent patterns, depending on the studies consulted. Available data suggest recidivism risk may be higher for young males than women (Argeriou et al. 1986; Jones and Lacey 2001; McMurran et al. 2011; Webster et al. 2009; Wells-Parker et al. 1991), but it appears that risk of recidivism may converge as adults of both genders age (Lapham et al. 2000). A study in 2000 involving a five-year follow-up of 2,615 convicted first-time impaired driving offenders in New Mexico revealed that overall 26% of offenders had been re-arrested (20% of females and 33% of males according to Lapham et al. 2000). The study further reported that, after controlling for a range of factors, young males had a recidivism rate 2.5 times that of women. However, a comparison of rates among older offenders revealed few differences between sexes (Laplante et al. 2008). A more recent population-based study (2010) in Maryland, reported that following their first conviction for impaired driving offences, the risk of recidivism is equivalent between female and male offenders. The study also noted that on average, drivers with repeat alcohol offences (as measured by violations on their driving record) were younger than drivers who did not have repeat alcohol offences on their driving record (Rauch et al. 2010). As relatively few studies have specifically examined this issue, more research is needed. 4.3 Summary of Characteristics In the past four decades, much has been learned about the profile and characteristics of impaired drivers. Nevertheless, there is far more research about males compared to females, with much of the female research being dated. This is an important research gap. However, there are some important similarities and differences between men and women that are worthy of our attention and consideration to inform efforts to better manage this offender population. What is perhaps most important however, is that impaired drivers of both sexes represent a heterogeneous population that is comprised of many different subgroups that require closer investigation and study Similarities between male and female impaired drivers On average, impaired drivers of both sexes are generally aged 20 to 40, with many offenders being in their 30s. Relative to the general population, impaired drivers of both sexes also are more likely to have less education and lower levels of employment and income; this finding is more pronounced among repeat offenders. Similarly, impaired drivers of both sexes are more likely to be single, separated, or divorced. Again, this finding is more pronounced among repeat offenders. Alcohol-related diagnoses are very common among impaired drivers of both sexes. In particular, the age of onset of drinking and family history warrant attention. To reiterate, while such diagnoses are highly correlated with impaired driving offending, they are not 38 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

140 necessarily a causal factor. Both male and female impaired drivers have higher levels of psychiatric symptoms relative to the general population so co-occurring disorders should not be overlooked during screening and assessment of this population. Moreover, recidivism rates for impaired driving among men and women of adult age appear similar following a first alcohol-related conviction Differences between male and female impaired drivers There are also some important differences between male and female impaired drivers. Men appear to exhibit a higher degree of antisocial attitudes and behaviours relative to women, although research comparing these populations on this dimension is sparse. Conversely, women experience more severe psychological and mental health symptoms as well as report greater involvement in drugs. Men may be more defensive about alcohol problems and, in particular, repeat male impaired drivers may demonstrate a greater readiness for change. In addition, younger males appear to have higher recidivism rates relative to females in this age category. Male impaired drivers also have more extensive histories of driving offences and other criminal offences as well as more prior experience with impaired driving interventions. To summarize, this research makes abundantly clear why the interventions based upon simple theoretical models that emphasize distinct aspects of behaviour, as opposed to a broader examination of the constellation of behaviours that are intimately linked to impaired driving, have failed to produce more dramatic results. This critical fact was succinctly captured in Wanberg et al. (2005) who stated there is no simple cause and effect model that can explain, let alone predict, impaired driving conduct. Many factors - early age drinking, environmental events, problem behaviour, personality characteristics, stress and the emotional syndromes of stress, cognitive and behavioural reinforcement and the impaired control - that interact with drinking and driving to result in impaired driving behaviour (p.20). Of clinical importance, this highlights the need for increased collaboration across disciplines to inform the development and delivery of interventions that are better suited to both match and target the diverse characteristics of this offender population. PROFILE AND CHARACTERISTICS OF IMPAIRED DRIVERS 39

141 5. IMPAIRED DRIVING RISK FACTORS Risk factors are characteristics that are identified (according to sufficient research evidence) to be indicators of the potential for a group of individuals with shared characteristics to engage in a specific behaviour in the future. It cannot be underscored enough that understanding the factors associated with recidivism is critical to our capacity for better detection of highrisk offenders and our ability to orchestrate effective countermeasures (Ouimet et al p. 743). Generally speaking, risk factors are organized in two distinct categories: 1) static factors (e.g., number of prior offences) that cannot be changed; and, 2) dynamic factors (e.g., substance abuse) which may change over time (Gendreau et al. 1996; DeMichele and Lowe 2011). Again, risk factors are relative to a group and not an individual and, subsequently, these measures are not very robust (Nadeau 2010). Risk assessment is a process that utilizes identified risk factors (usually in relation to multiple domains) to predict future behaviour. Risk assessment is not an exact science and risk factors only provide insight into the probability or likelihood of recidivism of offenders based upon existing research that is available. In this regard, much of the research around risk prediction has focused on criminal offenders and, in particular, those who have committed violent and/or sexual offences. Historically, risk assessment instruments were viewed as little more than educated guess work and, generally speaking, studies have demonstrated that the accuracy of risk assessment tools is questionable. As such, practitioners are cognizant of the potential for both false-positives and false-negatives (Miller and Brodsky 2011). False-positives are the application of a highrisk classification to offenders who do not recidivate. False-negatives, on the other hand, are the application of a low-risk classification to offenders who do recidivate. Strategies used to reduce the frequency of false-positives and negatives often utilize multiple factors and combine actuarial evaluation and clinical observation. 40 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

142 More recently, the quality of instruments 14 used with offenders generally has greatly improved (Andrews & Dowden 2006 ) as our understanding of risk factors has grown. To illustrate, a study conducted by Latessa et al. (2009) found that, among offenders who were three years post-release, 10% of offenders classified as low-risk were re-arrested compared to the rearrest of 70% of offenders classified as high-risk. Risk assessment instruments that possess a higher degree of accuracy in prediction generally account for multiple risk factors to reach a determination as to the probability of recidivism, and place a greater emphasis on objective measures as opposed to just the reliance on professional judgment which is more often subjective. A broad range of risk factors have been noted in the literature regarding impaired drivers including: sex, age, marital status, socio-economic status, history of prior treatment, impaired driving history, criminal history of violent aggression, prior traffic offences, test refusal or high-bac, and drinking patterns to name a few (Syrcle and White 2006). Yet, these studies vary dramatically in terms of the population studied, sample size, variables and measures utilized, data sources, analyses conducted, comparison groups employed, the time period used to measure recidivism, and the interpretation of results. Moreover, the number of studies that have examined the reliability of each individual risk factor is relatively small, which makes the drawing of conclusions a challenge. Hence, to date, there are no reliable predictors of risk among impaired drivers (Nadeau 2010). Moreover, what research there has been regarding the prediction of risk among impaired drivers has focused more on males than females (Lapham et al. 2006). In light of the limitations associated with research investigating risk factors associated with impaired driving, what is currently known about impaired driving risk factors should be interpreted cautiously. At best, no single impaired driving risk factor provides a clear indication regarding the potential for future impaired driving recidivism. Collectively, however, these risk factors may provide some insight that enable practitioners to gauge the need to further explore individual cases and the need for more intensive interventions. This recommendation is consistent with recommendations in the research literature (C de Baca et al. 2001; Nochajski and Stasiewicz 2006; Syrcle and White 2006). It has also been recommended that studies should assess relevant self-reported measures for response bias as this can influence outcomes in studies investigating predictors of recidivism (Schell et al. 2006). A brief overview of some of the key research studies that have been conducted on impaired driving risk factors is provided below. Inconsistent findings across studies are clearly evident in relation to some factors. An emphasis has been placed mainly on studies that have been conducted since 2000 with a few exceptions. Practitioners interested in more detailed 14 It is equally important that risk assessment instruments demonstrate proven reliability and are scientifically validated and standardized on an appropriate population. IMPAIRED DRIVING RISK FACTORS 41

143 information about risk factors are encouraged to consult the individual studies cited and to carefully consider the research designs that were used in the drawing of conclusions. 5.1 Male Risk Factors Demographic factors Age. A number of studies examine age as a factor. Lapham et al. (2000) determined that age group at screening was strongly associated with impaired driving recidivism for males but not for females. In particular, it was noted that younger age among males was an important factor in predicting recidivism. Other studies that have similarly reported that offenders that are of a younger age (under 30) are at greater risk to receive a subsequent impaired driving offence include C de Baca et al. (2001), Taxman and Piquero (1998), and Syrcle and White (2006). Most recently, this finding was again affirmed in a study by Rauch et al. (2010) which reported that younger males have a higher risk of recidivism than older males or females of any age group. Sex. A number of studies have suggested that males are more likely to be repeat offenders and this is perhaps one of the most common factors that is noted in the risk literature (Nochajski 1999; C de Baca et al. 2001; Syrcle and White 2006). However, more recently, a study by Rauch et al. (2010) reported that adult males and females are at equal risk for recidivism following their first alcohol-related violation. Although, the proportion of female drivers varied little between 1999 and 2004, their proportion decreased dramatically with increasing counts of prior violations. Women accounted for 51%, 18%, 13%, and 8% of the drivers with 0, 1, 2, and 3 or more prior violations, respectively. The male to female ratio of violation rates also decreased with increasing prior counts. The risk for men relative to women was 1.2 for drivers with 1 prior violation, 1.0 for drivers with 2 prior violations, and 1.0 for drivers with 3 or more prior violations (Rauch et al. 2010). Marital status. Lapham et al. (2000) reported that marital status is significant as a predictor but only when using a univariate model and not when included in a multivariate model. A study by Syrcle and White (2006) indicated that marital status, in particular, having never been married or being divorced, was a predictor for men and women. Most recently, in the development of the Impaired Driving Assessment (IDA) instrument for screening impaired driving for risk, needs, responsivity, and traffic safety (American Probation and Parole Association (2013)), never-married marital status was found to have a significant correlation with and contributed significant variance in predicting a twelve-month outcome scale comprised of variables measuring: any arrests, positive for drugs, missed judicial supervision appointments, revocation of probation, and re-arrest for impaired driving during the 12 month period of judicial supervision. 42 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

144 Ethnicity. The number of studies that have investigated this specific factor have been fairly limited, however a few studies have reported that ethnicity is a predictor of recidivism (Lapham et al. 2000; Christopherson et al. 2002). Of note, findings vary according to the nature and location of the study and are not consistent. C de Baca et al. (2001) noted that ethnicity was a predictor using a univariate model, but not in a multivariate model. Life history. Lapham et al. (2000) reports that the presence of family members or spouses with alcohol problems is predictive both of higher risk for alcohol problems among offenders, and also increased recidivism risk, and this finding was based upon self-reported characteristics. The study further notes that given the association of these indicators with recidivism, it appears prudent to elicit this information during a DWI evaluation (p.1653). Similarly, Wieczorek and Nochajski (2005) also reported that a father with a drinking problem and a relative arrested for impaired driving were the family factors most strongly associated with the number of prior impaired driving offences. Taking a slightly different approach, Begg et al. (2003) reported that aggressiveness at age 18 when combined with alcohol dependence at age 21 was a predictor for future involvement in an alcohol-related crash Personality and psychosocial factors According to a study by Mann et al. (2004), aggression is a risk factor for future impaired driving recidivism and also other public safety risks such as road rage. This is further substantiated by recent findings of alcohol problems among those involved in road rage incidents as Zuckerman (2000) found reckless driving was related to drinking. A number of studies have examined personality and psychosocial factors, including Wieczorek and Nochajski (2005) which suggest that psychiatric conditions could be useful for identifying potential recidivists. Conversely, Schell et al. (2006) conclude that there are no strong psychological predictors of recidivism (p. 34) Substance misuse Early onset of alcohol and drug use and abuse is predictive of adult impaired driving (Hingson et al. 2002; Hingson et al. 2003; NHTSA 2001). Specifically, early onset drinking is a predictor of several relevant behaviours including: future driving after any drinking, driving after five or more drinks, riding with an intoxicated driver, and involvement in alcohol-related crashes (Hingson et al. 2003). Frequency of drinking has been reported by Schell et al. (2006) as the single strongest predictor of driving after drinking. He further noted that persons who expect positive emotional outcomes as a result of drinking, and who drink frequently are more likely to continue to drive after drinking. IMPAIRED DRIVING RISK FACTORS 43

145 Lapham et al. (2000) reported that admission to lifetime use is a risk factor for recidivism. Similarly, Schell et al. (2006) noted that impaired driving offenders with the most severe alcoholism had the greatest risk for repeat impaired driving convictions. Finally, a 2006 study by Syrcle and White confirmed that drinking larger quantities of alcohol over extended periods of time prior to driving was also a predictor or recidivism risk. An examination of drug use as a factor by Wieczorek and Nochajski (2005) revealed that there were significant differences in drug use according to the number of prior offences, and indicated higher levels of drug use among repeat offenders BAC Although often cited as a reliable predictor of recidivism, research findings on this specific variable are mixed at best. C de Baca et al. (2001) reported that BAC was a significant predictor of recidivism whereas Wieczorek and Nochajski (2004) reported that offenders with lower BACs were more likely to recidivate. This is consistent with their earlier findings (Nochajski and Wieczorek 1997) which noted that a low BAC (under.16) is a better predictor of recidivism than a high BAC (.18 or greater). This is not to suggest that BAC is not an important variable for other purposes. In particular, BAC is a significant predictor of degree of involvement in and disruption from alcohol use and abuse and it should be used along with information about alcohol and drug use as a key factor in determining appropriate placement in treatment interventions. BAC at the time of arrest is generally recognized as an important factor to distinguish between different types of impaired drivers and their need for assessment and/or intervention (Wanberg et al. 2005; Syrcle and White 2006). In a study examining the characteristics of impaired driving recidivists, Caviola et al. (2007) concluded that BAC may have limited utility for the purposes of screening. In particular, the study reported that this should not be interpreted to mean that high blood alcohol levels at the time of arrest do not have clinical utility. Rather, it is recommended that BAC be interpreted cautiously or in conjunction with other predictors of potential DUI recidivism risk (p.859). Most recently, a study by Dugosh et al. (2013) provides evidence to indicate that a driver s BAC level at arrest, in the absence of other information, also may not be a reliable indicator of the degree of alcohol-related problems including diagnoses of abuse and dependence Instruments There are several risk assessment instruments that have reported some strength in predicting impaired driving recidivism risk. First, the MAST has been found to significantly predict recidivism status as reported in two key studies (Lapham et al. 2000; Cavaiola et al. 2003). However, the Lapham et al. (2000) study only determined that the MAST was associated with 44 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

146 recidivism when univariate analyses were conducted, and when a multivariate model was utilized the results were no longer significant. Second, the MAC scale of the MMPI, which measures general personality traits characterized by sociability, boldness, rebelliousness, and pleasure-seeking also has shown some positive results. In particular, high scores on the MAC (a raw score of 23 or higher) have been shown to be predictive of impaired driving recidivism (Lapham et al. 1997). Most recently, a review of the evaluation literature in relation to risk assessment instruments by Brown and Ouimet (2013) concluded that there is support for the MAC s scale s predictive validity for [DWI] risk assessment but more mitigated support for other MMPI scales (p.310). Third, a study by Nochajski and Wieczorek (1998) (cited in Cavaiola et al. 2007) reported that subtle items of alcoholism included in the RIASI were predictive of recidivism. Finally, a study by Syrcle and White (2006) reported that 14 of the 16 scales of the Adult Substance Use Survey Revised-Illinois (ASUDS-RI) (Wanberg and Timken 2006) uncovered significant differences between first and repeat impaired drivers. A current demonstration project being conducted by the American Probation and Parole Association and funded by the National Highway Traffic Safety Administration (DeMichele et al. 2013) has developed a preliminary Impaired Driving Assessment screening instrument. Preliminary findings are promising with respect to providing probation intake services guidelines for judicial supervision placement and referral to outside services (Wanberg and Lowe 2013). Of note, there is important research that illustrates how variations in base rates of failure and selection ratios affect conclusions concerning the efficacy of different instruments as a strategy to demonstrate the value of evaluation standards in order to make valid comparisons between risk prediction instruments (Anderson et al. 2000, p. 915). In layman s terms, this means that different jurisdictions or offender samples will have higher or lower rates of failing, and that agencies need to make decisions about how to balance the positive and negative predictions. That is, assessment is an exercise in prediction, and prediction has error. Hence, some offenders will be predicted to recidivate but do not (false-positive), whereas others will be predicted to recidivate and they do (true positive). Similarly, those predicted to be low risk may recidivate (false-negative) and others will not recidivate (true negative). It is a bit of an art to balance these issues, but also a matter of agency capacity. The bottom line is that due to decisions regarding instrument precision, practitioners should be careful about comparing different assessments and even the same assessment across different populations Biomarkers Impaired drivers, both first and repeat offenders, suffer from high rates of alcohol use disorders (AUDs) (Lapham et al. 2001). Biomarkers can detect the presence of these disorders IMPAIRED DRIVING RISK FACTORS 45

147 fairly accurately. A number of studies have investigated the extent to which biomarkers are predictive of impaired driving recidivism. Couture et al. (2010) showed biomarkers were not a good predictor of recidivism, individually or as a group. They failed to differentiate between first and repeat impaired driving offenders. The primary reason for this is that biomarkers may not capture the drinking patterns that are most common among impaired driving offenders e.g., binge drinking (Couture et al. 2010). Biomarkers more accurately identify severe and chronic patterns of alcohol use as opposed to the episodic heavy drinking that often precipitates impaired driving. Moreover, alcohol misuse alone is not enough to identify the propensity of an individual to recidivate as there are a combination of other factors (such as personality traits or cognitive impairments) that can interact with substance misuse to lead to high-risk behaviour such as impaired driving (Brown et al. 2009; Nochajski and Stasiewicz 2006; Ouimet et al. 2007). As such, biomarkers of chronic patterns of heavy drinking may not be adequate in and of themselves to capture the multiple processes that appear to promote recidivism such as binge drinking and other risky behavioural and personality features (Couture et al. 2010, p. 307) Driver and criminal history Driver history. In a study by Peck et al. (1994), driving records of first and repeat impaired drivers (using a four-year follow-up period) were analyzed using multivariate analyses to assess predictors of impaired driving recidivism. Prior involvement in crashes and traffic violations were the strongest predictor of membership in the repeat offender group. Similarly, NHTSA (1996) reported that the risk of future arrests rises in conjunction with the number of prior impaired driving arrests. A major study in Maryland by Rauch et al. (2002) involving several thousand driver records confirmed that any alcohol-related driving event is predictive of future impaired driving behaviour. In 2005(a), Wieczorek and Nochajski confirmed this finding as did Schell et al. (2006) who noted that high-risk driving style was a significant predictor with a moderate effect size. Most recently, Cavaiola et al. (2007) also concluded that a poor driving record that includes offences both prior to and following the initial impaired driving offence is predictive of recidivism. However, some have noted that prior impaired driving arrests may not be a good predictor as the presence of prior arrests is influenced to a large extent by the level of impaired driving enforcement as well as the length of the look-back period for counting prior arrests (Nochajski and Stasiewicz 2006). Criminal history. Some studies have reported that prior criminal history other than impaired driving offences is a predictor of impaired driving recidivism (Syrcle and White 2006). A 2007 study by Labrie et al. examined criminality and continued impaired driving offences and concluded that rates of recidivism increased with the severity of criminal behaviour (e.g., crimes progressing from substance-related crimes to property crimes to crimes against persons). Compared to the DUI only type, the property crime subjects were 1.4 times more 46 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

148 likely to be re-arrested for DUI and person crime subjects were twice as likely to recidivate (Labrie et al. 2007, p ). Moreover, research investigating risk factors associated with criminal re-offending has identified a number of objective and verifiable risk indicators that are useful to distinguish between first and repeat impaired drivers. These variables are associated with an offender s criminal history and include age at time of first arrest for any criminal act, age at time of first impaired driving conviction, having a prior summary of alcohol- or drug-related offence, having a prior misdemeanor offence, having a misdemeanor offence for a crime against persons, or having five or more prior moving violations (Dugosh et al. 2013, p.8). In addition, other risk variables that have been shown to differentiate between first and repeat impaired driving offenders include age of onset of substance abuse, having a prior treatment episode, or loss of employment or expulsion from school because of drug or alcohol use (Dugosh et al. 2013, p.8) Interlock fails Research suggests that a high rate or pattern of BAC fail readings from the alcohol interlock, particularly in excess of.02, is predictive of the likelihood of impaired driving recidivism (Marques et al. 2003; Beirness and Marques 2004). A major study conducted in Alberta analyzed 5.5 million BAC tests provided by 2,200 offenders (Marques et al. 2001). It was subsequently demonstrated that the likelihood of future impaired driving convictions in the first two years following the removal of the interlock can be strongly predicted based on the rate of elevated (greater than.02) interlock BAC tests (Beirness and Marques 2004). A subsequent study in Quebec involving 7,200 offenders who provided 18.8 million breath tests confirmed this finding (Marques et al. 2003). In fact, more interlock warnings and failures logged during the first five months of interlock usage predict greater than 60% of repeat impaired driving offence with a false-positive rate (which occurs when a clean breath sample is erroneously determined as containing alcohol) of less than 10% (Marques et al. 2001). Researchers have also determined that the presence of elevated BAC tests during early morning hours can also assist in predicting future impaired driving offences (Beirness and Marques 2004). Early morning high BAC tests are usually a result of drinking the prior evening and indicate the extent of drinking that occurred. The presence of two or more elevated BAC test results during the morning hours further bolsters the predictive model regarding the likelihood of future impaired driving offences (Marques et al. 2003). Prediction of repeat offences has been associated with a profile of drivers who are both multiple offenders and who have more than a few elevated interlock BAC tests (Marques et al. 2003). Marques and Voas (2008) found that the number of failed BAC tests logged is predictive of repeat impaired driving offenders. The higher the rate of failed tests, the more likely offenders will recidivate once the interlock is removed. Also, those offenders who are IMPAIRED DRIVING RISK FACTORS 47

149 in the top 20-30% of elevated interlock BAC tests have significantly higher levels of alcohol biomarkers associated with problem drinking (Marques and Voas 2008) Repeat and/or hard core impaired drivers According to a presentation by Nadeau (2010) at an international conference in Canada, a number of recent studies have identified risk factors among repeat offenders in comparison to first offenders. Low levels of participation or involvement in treatment and treatment interventions is considered predictive of recidivism (Aharonovich et al. 2003; Crews et al. 2005). This is further confirmed is a study by Syrcle and White (2006) and a review of the literature by Wanberg et al. (2005). Neurocognitive deficits have also been reported as predictive of recidivism among repeat offenders. More specifically, these deficits can contribute to variation in affect, impulsivity, problem solving, perception and memory (Glass et al.2000; Ouimet et al. 2007). Finally, a reduced ability to change is also predictive among repeat offenders of future impaired driving offences (Buntain-Ricklefs et al. 1995; Glass et al. 2000; Ouimet et al. 2007). 5.2 Female Risk Factors There is one key study that examined differences in risk factors among men and women. For the most part, few differences were found in terms of predictive variables with the exception that women were more likely to report a history of aggressive behaviour towards a partner than were males, and this indicator was associated with increased recidivism (Lapham et al. 2000). Of interest, the Lapham et al. (2000) study further noted that, while rates of physical and sexual abuse among men and women are high among those with substance abuse problems, this factor is not associated with recidivism for either sex. 5.3 Summary While it is clear that a wide range of risk factors have been examined in relation to the prediction of repeat impaired driving offences in the past two decades, the findings from this research are inconsistent in many cases and far from conclusive. There are only a small handful of common factors that have been investigated across several studies, however with regard to criminological research, more is known about risk factors among repeat drunk drivers. For these reasons, practitioners in the field are encouraged to take a broader view of and approach to the use of these factors, and focus on the presence of a number of risk factors collectively as a basis to inform decisions, as opposed to the presence or absence of individual factors. Much more research on this issue is needed before definitive conclusions can be reached. 48 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

150 5.4 Some Reflections on Estimating Impaired Driving Recidivism By: Dr. Ken Wanberg & Dr. David Timken There are a number of problems and questions that impact on this approach or any recidivism-risk prediction model. First, given the evidence thus far, the best set of predictor variables or scales that can be gleaned from multivariate studies will serve only as estimates of recidivism. Most models are linear: those with high scores are high risk or positive for recidivism; those with low scores are low risk or negative for recidivism. And, these models can do a fair job of estimating the percent of individuals who will or will not recidivate. Predictive models become complex when we look at the false-positive and false-negative issue because these models tend to focus on identifying those who are positive for reoffending. Those who do not offend in this positive group are false-positives. Thus, if the false-positive rate is 35%, then the predictive model is correct 65% of the time. However, what about the residual group, or those negative for re-offending? What percent of those do re-offend? If the linear model is reliable, then we can decrease the false-positives by choosing a higher cut off value and putting fewer clients in the potential to re-offend. However, this just increases the risk of false-negatives or a higher percent of those not positive for reoffending who do re-offend. Second, there are many unknown or un-measured idiosyncratic variables that can occur in the individual s life that will contribute variance to outcome. For example, a never-married male with high potential for recidivism based on the best predictor variables gets married, has a child, and engages in a life-path of responsibility. Or, someone identified as a low risk for recidivism experiences a traumatic life-event (e.g., divorce, losing a job), becomes depressed and doesn t care and drives impaired. In our clinical experience we have found these to be rather frequent occurrences. These new events usually cannot be predicted by retrospective measures and certainly, evaluators do not have a magician s ball to predict these occurrences. Third, any measured prediction of recidivism at either the group level or individual level will be affected by service interventions provided to impaired driving offenders. Most, if not all, sentenced impaired driving offenders receive judicial supervision, education and/or treatment services or a combination of all three. If these are effective, and the literature indicates this to be the case, then they will mitigate the estimated prediction. That is, effective intervention services will tend to increase the false-positives. Any risk-prediction model must consider the variance resulting from intervention services. Unfortunately, since intervention services vary considerably as to method, type and efficacy, one can only estimate the impact of these services (e.g., reducing recidivism rates by 10 to 20%). IMPAIRED DRIVING RISK FACTORS 49

151 Fourth, since it is common to provide fewer services for those evaluated as having a low recidivism risk, this can contribute to an increase of false-negatives. This is based on the common view held among judicial workers that those identified as low risk need minimal or no intervention services. What this does is generate a relatively large cadre of those identified as negative for re-offending which increases the probability of a false-negative. It is quite likely that the Wieczorek and Nochajski (2005) findings that those with lower BACs had a higher probability or recidivism had more to do with a significant percent of those with low BACs being placed in no or very minimal intervention services. Fifth, the difficulty of predicting impaired driving recidivism is increased by the low variance of outcome variables. Outcome measures such as revocations, re-arrests, missing judicial supervision appointments, and so forth, taken at six-month post-probation intake typically have low variance (Wanberg and Lowe 2013). This can limit the use of predictive statistical methods when samples are relatively small. For example, if only 5% of the clients are rearrested six months post-sentencing, even a sample of 500 will provide only 25 offenders in the re-arrest category. Sixth, re-offending statistics are usually based on the big-face valid variable of re-arrests. However, the literature indicates that the percent who drive impaired is much greater than those who are re-arrested (see Wanberg, Timken and Milkman 2010). The pool of re-arrests must come from that group; but, is it a random sample of that group? Most likely it is not. Thus, recidivism prediction models should take into account those who drive impaired but are not arrested which provides an outcome variable that has a higher percent of variance. Seventh, the value of impaired driver screening and assessment is significantly diminished when its main focus is only on risk assessment. More importantly, its value lies in providing guidelines for the type of intervention services including judicial supervision and alcohol and other drug and impaired driving intervention services. Finally, impaired driving assessment becomes more effective when it is based on a convergent validation model (Wanberg and Milkman 1998; 2008; 2010; Wanberg et al. 2005). Based on the classic study of Campbell and Fiske (1959), the convergent validation assessment model holds that both self-report and other report information and data are used to converge on the best estimate of the individual s conditions related to alcohol and other drug use and factors contributing to impaired driving conduct and the best estimate of the individual s service needs. Self-report is seen as essential in this model in that it is a valid representation of where the individual is at the time of assessment and their willingness to self-disclose. Comparing self-report with other report data provides a basis for not only estimating the individual s condition and service needs, but also their level of defensiveness at the time of assessment. If services are working and the individual s willingness to self-disclose increases, this increases the probability of favourable outcomes. 50 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

152 6. RISK ASSESSMENT INSTRUMENTS The effective management of the many different types of impaired drivers is based upon the identification and development of a range of supervision strategies and interventions specifically geared towards those offenders who are more or less amenable to behaviour change. This is a fundamental principle of evidence-based practices. Of considerable importance, the use of valid and reliable risk assessment instruments is essential to accurately differentiate between the different types of impaired drivers that exist and ensure that they are streamed into appropriate interventions designed to address their specific risks and needs. A number of standardized assessment instruments are available to help quickly identify current and potential alcohol problems 15. Generally speaking, these instruments are based upon a series of cutoff scores associated with the probabilities of re-offending to place offenders into specific groups or types (DeMichele and Payne 2013). Regarding their use with impaired driving offenders, it has been suggested these instruments place a disproportionate focus on alcohol use and, as such, suffer from tunnel vision. Importantly, research has shown that some offence types are more accurately classified when using instruments developed specifically for those offence categories (e.g., domestic violence offenders, sex offenders). These assessment tools are designed to identify as many potential cases as possible, while at the same time minimizing the number of false-positives (i.e., identifying someone as highrisk for re-offending when they are not). Some of these instruments are not as strong and have demonstrated limited validity and reliability in relation to the accurate prediction of future impaired driving events, including the following: > Mortimer Filkins (MF) (Chang et al. 2002; Wendling and Kolody 1982); and > Driver Risk Inventory (DRI) (Chang et al. 2002). 15 It is important to recognize that not all screening instruments commonly used among impaired drivers are designed to detect drug use. Instruments that will detect drug use include Alcohol, Smoking, Substance Involvement Screening Test (ASSIST), Cut-down, Annoyed, Guilty, Eye-opener Adapted to Include Drugs (CAGE AID), Addiction Severity Index (ASI), and Global Appraisal of Individual Needs (GAIN). RISK ASSESSMENT INSTRUMENTS 51

153 In light of the strengths and weaknesses associated with many of the available instruments, a majority of jurisdictions rely on the outcomes of several instruments during the assessment process in order to produce a more complete picture of impaired driving offenders. This section briefly describes some of the instruments that are most commonly used across Canada and the United States. Each instrument is described in terms of type of administration, who it can be administered by, number of items, time required for administration, training required for administration, scoring, summary of psychometrics, limitations, cost, and source. In addition, a few key references are identified in relation to each instrument in order to provide additional information to practitioners seeking more knowledge about the risk assessment instrument. The following is a list of the instruments described in this section: > ADS (Alcohol Dependence Scale); > ASUDS-R (Adult Substance Use and Driving Survey Revised); > ASI (Alcohol Severity Index); > AUDIT (Alcohol Use Disorders Identification Test); > IDTS (Inventory Drug-Taking Situations); > DAST (Drug Abuse Screening Test); > LSI-R (Level of Service Inventory-Revised); > MAST (Michigan Alcoholism Screening Test); > SASSI (Substance Abuse Subtle Screening Inventory); > RIASI (Research Institute on Addiction Self Inventory); and, > Biomarkers. There are no clear indications of the superiority of any one screening instrument or set of instruments and procedures. Research suggests that the selection of specific instruments and procedures should be guided by the needs and resources in particular jurisdictions (Beirness et al. 1997). It warrants mentioning that adapting instruments can jeopardize their validity and may require further research. More importantly, relevant laws pertaining to copyright should be reviewed. When reading through the summary of psychometrics for each instrument there are some important caveats to note with regard to the metrics pertaining to reliability and validity. Validity measures examine how well an instrument separates recidivists from non-recidivists. The area under the curve is a commonly used statistic to assess predictive discrimination. Reliability looks at how consistently the instrument can be scored across raters (e.g., if ten 52 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

154 raters each score the same case to what extent will there be agreement among raters). The intraclass correlation coefficient (ICC) is a commonly used measure of inter-rater reliability. In particular, Cronbach s alpha provides another look at reliability insofar as it examines internal consistency among items. So in these instances, Cronbach s alpha is, in fact, a somewhat narrow, albeit important, look at reliability as it examines internal consistency among items. With regard to the studies reviewed, a commonly accepted interpretation of these numbers is 0.80 and over is considered a very good; 0.70 to 0.80 is considered acceptable; 0.60 to 0.70 is considered questionable; 0.50 to 0.60 is poor; and less than 0.50 is unacceptable (George and Mallery 2003). 6.1 Alcohol Dependence Scale (ADS) Brief description The ADS provides a quantitative measure of the severity of alcohol dependence consistent with the concept of the alcohol dependence syndrome. The 25 items cover alcohol withdrawal symptoms, impaired control over drinking, awareness of a compulsion to drink, increased tolerance to alcohol, and salience of drink-seeking behaviour. The ADS yields a measure of the severity of dependence that is important for treatment planning, especially with respect to the intensity of treatment. The printed instructions for the ADS refer to the past 12-month period. However, instructions can be altered for use as an outcome measure at selected intervals (e.g., 6 months, 12 months, 24 months) following treatment. The ADS can be completed in approximately five minutes and as a result can be used for screening and case-finding in a variety of settings including health care, corrections, general population surveys, workplace, and education. Guidelines are given for using the ADS with respect to treatment planning, particularly with respect to the level of intervention. A French language translation is available. Type of administration Pencil-and-paper self-administered > Interview > Computer self-administered Administered by > Self Number of items > 25 RISK ASSESSMENT INSTRUMENTS 53

155 Time required for administration > 5 minutes Training required for administration > Yes, only basic training needed Scoring > Administrator or by computer Summary of psychometrics (reliability/validity) The ADS is widely used as a research and clinical tool, and studies have found the instrument to be reliable and valid. The ADS can be used for basic research studies where a quantitative index is required regarding the severity of alcohol dependence. For clinical research, the ADS is a useful screening and case-finding tool. It is also of value with respect to matching clients with the appropriate intensity of treatment and for treatment outcome evaluations. Items making up the ADS were found to have good internal consistency (measuring whether several items that propose to measure the same general construct produce similar scores; a=.92 Skinner and Horn 1984). The scale consists of three factors: the first major factor accounted for items reflecting withdrawal symptoms, the second and third smaller factors were made up of items reflecting obsessive compulsive drinking patterns and loss of behavioural control (Skinner and Horn 1984). The ADS has good concurrent validity (demonstrated that test correlates well with a measure that has previously been validated). Skinner and Horn (1984) reported that the ADS score was correlated with both daily consumption of alcohol and lifetime use of alcohol, social consequences from drinking, prior treatment for alcohol abuse, use of alcohol to change mood and feelings of guilt over drinking. The ADS was also significantly correlated with the MAST (Skinner and Horn 1984; Ross et al. 1990). The ADS has been successfully adapted for use with a variety of different cultures and ethnic groups (Fu et al. 2008; Rajendran and Cheridan 1990; Solís et al. 2007). The translated versions of the ADS were found to have high internal reliability (referring to the extent to which a measure is consistent within itself). The ADS was found to correlate well with alcohol-related problems and post-release drinking goals with incarcerated male offenders (Hodgins and Lightfoot 1989). The ADS appears to be an equally valid and reliable measure of alcohol dependence in women (Chantarujikapong et al. 1997; Drake and Mercer-McFadden 1995). Internal consistency (measuring whether several items that propose to measure the same general construct produce similar scores) is also high in this population (ranging from.85 [dependent participants] to.99 [total sample]; Chantarujikapong et al. 1997). The ADS has been used 54 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

156 successfully in several studies investigating alcohol dependence in homeless and incarcerated women, and studies of alcohol misuse in Australian female university students (e.g., Biron et al. 1995; Chantarujikapong et al. 1997; Williams et al. 1998). Limitations > Cost for use > Limited to alcohol dependence assessment only Cost User s guide > ISBN pages booklet/guide > $17.95 published 1984 product code PG010 > Available in English only Questionnaires (package of 25) > $11.00 published 1984 product code P143 > Available in English and French ADS kit (User s Guide and 25 Questionnaires) > $20.95 published 1984 product code PG011 Download a PDF version of the Order Form: Publications/camh_publications_orderform.pdf The Centre for Addiction and Mental Health 33 Russell Street Toronto, Ontario M5S 2S1 Phone: ext publications@camh.net Website: html RISK ASSESSMENT INSTRUMENTS 55

157 Source Harvey Skinner Multi-Health Systems regarding the Computerized Lifestyle Assessment: (Canada) References Biron, L.L., Brochu, S., & Desjardins, L. (1995). The issue of drugs and crime among a sample of incarcerated women. Deviant Behaviour, 16(1), Chantarujikapong, S.I., Smith, E. M., & Fox, L.W. (1997). Comparison of the Alcohol Dependence Scale and Diagnostic Interview Schedule in homeless women. Alcoholism: Clinical and Experimental Research, 21(4), Drake, R.E., & Mercer-McFadden, C. (1995). Assessment of substance use among persons with chronic mental illness. In A.F. Lehman & L.B. Dixon (Eds.), Double Jeopardy: Chronic Mental Illness and Substance Use Disorders. Chur, Switzerland: Harwood Academic Publishers. Fu, S.S., Kodl, M., Willenbring, M., Nelson, D.B., Nugent, S., Gravely, A.A., et al. (2008). Ethnic differences in alcohol treatment outcomes and the effect of concurrent smoking cessation treatment. Drug and Alcohol Dependence, 92(1-3), Hodgins, D.C., & Lightfoot, L.O. (1989). The use of the Alcohol Dependence Scale with incarcerated male offenders. International Journal of Offender Therapy and Comparative Criminology, 33(1), Rajendran, R., & Cheridan, R.R. (1990). Tamil adaption of the Alcohol Dependence Scale. Journal of the Indian Academy of Applied Psychology, 16, Ross, H.E., Gavin, D.R., & Skinner, H.A. (1990). Diagnostic validity of the MAST and the alcohol dependence scale in the assessment of DSM-III alcohol disorders. Journal of Studies on Alcohol and Drugs, 51, Saxon, A.J., Kivlahan, D.R., Doyle, S., et al. (2007). Further validation of the Alcohol Dependence Scale as an index of severity. Journal of Studies on Alcohol and Drugs, 68(1), Skinner, H.A., & Allen, B.A. (1982). Alcohol dependence syndrome: Measurement and validation. Journal of Abnormal Psychology, 91, Skinner, H.A., & Horn, J.L. (1984). Alcohol Dependence Scale: Users Guide. Toronto, Canada: Addiction Research Foundation. Solís, L., Cordero, M., Cordero, R., & Martínez, M. (2007). Caracterización del nivel de dependencia al alcohol entre habitantes de la Ciudad de México. [Characterization of level of alcohol dependence in Mexico City inhabitants.]. Salud Mental, 30(6), IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

158 Williams, R.J., Connor, J.P., & Ricciardelli, L.A. (1998). Self-efficacy for refusal mediated by outcome expectancies in the prediction of alcohol-dependence amongst young adults. Journal of Drug Education, 28(4), Adult Substance Use and Driving Survey Revised (ASUDS-R) Brief description The ASUDS-R is a 123 item psychometric-based, self-report screening instrument that also incorporates information gathered through collateral data and an individual interview. Its purpose, according to Wanberg and Timken (2006) is to provide a differential screening assessment of the driving while impaired (DWI) offender in the areas of substance use and abuse, alcohol involvement, and other areas of life-adjustment problems (p.8). The ASUDS-R assesses an individual s alcohol and other drug use involvement in ten categories of drugs, and measures the degree of disruption that might result from the use of these drugs. The ASUDS-R provides a specific measure of the degree of involvement in the use of alcohol, and a specific measure of driving-risk attitudes and behaviours. It also provides a screen for emotional or mood adjustment problems, a measure of social non-conformity, a measure of legal non-conformity, a measure of defensiveness or resistance to self-disclosure, and a scale to assess motivation and readiness for treatment. It provides measures of alcohol and other drugs (AOD) involvement and legal conformity for the most recent six month period the client has been in the community. The ASUDS-R can be used to provide guidelines for assessing levels of alcohol or other drug problems, abuse, and dependence. It can also be used to provide referral guidelines for various levels and types of services for impaired driving offenders. It can be used to assess during and post-treatment changes. Type of administration > Pencil-and-paper and automated version Administered by > Self or practitioner Number of items > 123 Time required for administration > minutes RISK ASSESSMENT INSTRUMENTS 57

159 Training required for administration > Comprehensive training is required to administer the instrument and interpret its scores. Scoring The ASUDS-R is scored using four weighted scales and can also include collateral data, including BAC at arrest, prior substance abuse treatment, and prior impaired driving arrests or convictions. Summary of psychometrics (reliability/validity) A study conducted by Wanberg and Timken (2007, unpublished) of the psychometric properties found the analytical results were largely favorable. Internal consistency (measuring whether several items that propose to measure the same general construct produce similar scores) reliabilities were within optimal range. Each scale was found to render a unique dimension, inter-correlations among scales were consistently positive, consistency of measurement among different samples was strongly supported, and robust correlations were found with external criterion tests and scales. Evidence was found to support the use of the ASUDS-R scales independently and in combination with collateral variables to provide service guidelines for impaired driving offenders. Wanberg and Timken (2007, unpublished) also found the ASUDS-R to be a valid selfreport differential screening instrument that provides sound guidelines for decision-making, particularly when integrating findings from other report data (e.g., BAC, prior offences), and when used in combination with placement criteria such as those developed by the American Society of Addiction Medicine. The User s Guides for the original ASUDS and ASUDS-R provide a detailed summary of scale construct validation studies. Limitations > Cost of use > Limited peer reviewed examination of the psychometric properties in published journals > Review of instrument has only been done by the authors of the instrument > Has not been included in any meta-analysis reviews of screening instruments Cost Costs are not listed on website. Please use the following links to contact Dr. Wanberg about costs. 58 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

160 The computerized version can be purchased from Dr. Wanberg s web site: com/software_applications/how_to_buy.htm. To inquire about use of the paper version, see Source Center for Addiction Research and Evaluation (CARE) P.O. Box Arvada, CO Tel: References Ferris, M. (2007). DWI Screening in Ramsey County: An Assessment of the ASUDS-R. St. Paul, MN, December. Wanberg, K.W., & Timken, D. (1998). The Adult Substance Use and Driving Survey (ASUDS). Arvada, CO: Center for Addictions Research and Evaluation. Wanberg, K.W., & Timken, D. (2006). The Adult Substance Use and Driving Survey (ASUDS-R). Arvana, CO: Center for Addictions Research and Evaluation. Wanberg, K.W., & Timken, D. (2007, unpublished manuscript). Adult Substance Use and Driving Survey-Revised (ASUDS-R): Psychometric properties and construct validity. White, W.L., & Syrcle, J. (2008). Evaluating the hard-core drinking driver: Illinois prioritizes the identification of offenders who need services the most. Addiction Professional, 6(1), Alcohol Severity Index (ASI) Brief description The ASI is an assessment instrument designed to be administered as a semi-structured interview. The instrument gathers information about seven areas of a patient s life: medical, employment/support, drug and alcohol use, legal, family history, family/social relationships, and psychiatric problems. In approximately one-hour a trained interviewer can gather information on recent (past 30 days) and lifetime problems in all of the aforementioned areas. The ASI provides an overview of problems related to substance, rather than focusing on any single area. The ASI can be used effectively to explore problems within any adult group of individuals who report substance abuse as their major problem. It has been used with psychiatrically ill, homeless, pregnant, and prisoner populations, but its major use has been with adults seeking RISK ASSESSMENT INSTRUMENTS 59

161 treatment for substance abuse problems. The ASI has been used extensively for treatment planning and outcome evaluation. Outcome evaluation packages for individual programs or for treatment systems are available. More recently, the Treatment Research Institute (TRI) that developed the ASI has released a Risk and Needs Triage (RANT) decision support tool for judges and other justice decision makers to assist in matching drug-involved offenders to the community corrections program best suited to their supervision and treatment needs. Efforts are underway to develop a similar tool that is designed for an impaired driving offender population (Marlowe 2008). Type of administration > Pencil-and-paper self-administered > Interview > Computer self-administered Administered by > Self Number of items > 161 Time required for administration > 45 to 75 minutes Training required for administration > Training is required. There is a self-training packet available, as well as onsite training by experienced trainers. Scoring > Takes about 5 minutes to score. > The ASI provides two scores: severity ratings are subjective ratings of the client s need for treatment, derived by the interviewer; composite scores are measures of problem severity during the prior 30 days and are calculated by a computerized scoring program. Summary of Psychometrics (reliability/validity) According to SAMHSA (2005) The ASI is highly correlated with objective indicators of addiction severity. The ASI is also one of the few instruments that measure several different functional aspects of psychosocial functioning related to substance abuse and provide a concise estimate of the history of substance abuse as well as recent use. The instrument 60 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

162 provides severity ratings in each functional area assessed, which are useful both clinically and for research purposes (p. 20). The instrument has demonstrated high reliability and concurrent validity (demonstrated that test correlates well with a measure that has previously been validated - Leonhard et al. 2000; McLellan et al. 1992a; Schottenfeld and Pantalon 1999). The items in each of the seven areas have been tested for understanding and test-retest reliability (measures stability of the scores over time) as well as concurrent, predictive, and discriminate validity (tests whether concepts or measurements that are supposed to be unrelated are, in fact, unrelated) among adults of both genders and most large ethnic groups (McLellan et al. 1985; Kosten et al. 1983; Hendriks et al. 1989). The ASI has become very widely used mainly due to extensive psychometric testing, a comprehensive training manual (plus other instructional materials), and its availability in the public domain. A self-report version of the ASI has been shown to be a reliable and accurate alternative to the counselor-administered instrument (SAMHSA 2005), however, the latter is the more preferred approach given the recognized limitations associated with self-report instruments. Numerous publications were found reporting the reliability and validity of the ASI for opioid users, crack and cocaine users, those with mental illness, the homeless, gamblers, and those in rehabilitation, detoxification, and various drug treatment programs (Drake et al. 1995; Fureman et al. 1994; Hendricks et al. 1989; Hodgins and El-Guebaly 1992; Joyner et al. 1996). Limitations > It should not be used in group testing or for fast screening > Limited research using an impaired driving offender population > Designed as an assessment tool and not a screening tool Cost There is no cost as a result of the instrument being public domain. A minimal charge for photocopying and mailing may apply. A free scoring disk is provided with the training materials, and there is a software program to provide written evaluations and treatment plans (there is a cost for this program). Source A.T. McLellan, Ph.D. Building #7 PVAMC University Avenue Philadelphia, PA Phone: Fax: tmclellan@tresearch.org RISK ASSESSMENT INSTRUMENTS 61

163 References Butler, S.F., Budman, S.H., Goldman, R.J., Newman, F.L., Beckley, K.E., Trottier, D., & Cacciola, J.S. (2001). Initial validation of a computer-administered Addiction Severity Index: The ASI-MV. Psychology of Addictive Behaviours, 15(1), Drake, R.E., Mchugo, G.J., Biesanz, J.C., & Validity and Reliability Committee of the NIAAA (1995). The test-retest reliability of standardized instruments among homeless persons with substance use disorders. Journal of Studies on Alcohol, 56(2), Fureman, I., McLellan, A.T., & Alterman, A.I. (1994). Training for and maintaining interviewer consistency with the ASI. Journal of Substance Abuse Treatment, 11(3), Hendricks, V.M., Kaplan, C.D., Van-Limbeek, J., & Geerlings, P. (1989). The Addiction Severity Index reliability and validity in a Dutch addict population. Journal of Substance Abuse Treatment, 6(2), Hodgins, D.C., & El-Guebaly, N. (1992). More data on the Addiction Severity Index: Reliability and validity with the mentally ill substance abuser. Journal of Nervous and Mental Disease, 180(3), Joyner, L.M., Wright, J.D., & Devine, J.A. (1996). Reliability and validity of the Addiction Severity Index among homeless substance misusers. Substance Use and Misuse, 31(6), Kosten, T.R., Rounsaville, B.J., & Kleber, H.D. (1983). Concurrent validity of the addiction severity index. Journal of Nervous and Mental Disease, 171(10), Leonhard, C., Mulvey, K., Gastfriend, D.R., & Shwartz, M. (2000). Addiction Severity Index: A field study of internal consistency and validity. Journal of Substance Abuse Treatment, 18(2), Marlowe, D.B. (October 9, 2008). Personal communication. McLellan, A.T., Luborsky, L., & Cacciola, J. (1985). New data from the Addiction Severity Index: Reliability and validity in three centers. Journal of Nervous and Mental Disease, 173(7), McLellan, A.T., Luborsky, L., O Brien, C.P., & Woody, G.E. (1980). An improved diagnostic instrument for substance abuse patients: The Addiction Severity Index. Journal of Nervous & Mental Diseases, 168, McLellan, A.T., Alterman, A.I., Cacciola, J., Metzger, D., & O Brien, C.P. (1992). A new measure of substance abuse treatment: Initial studies of the Treatment Services Review. Journal of Nervous and Mental Disease, 2, Moos, R.H., Finney, J.W., Ferderman, E.B., & Suchinsky, R. (2000). Specialty mental health care improves patients outcomes: Findings from nationwide program to monitor the quality of care for patients with substance use disorders. Journal of Studies on Alcohol, 61(5), Rosen, C.S., Henson, B.R., Finney, J.W., & Moos, R.H. (2000). Consistency of self-administered and interview-based Addiction Severity Index composite scores. Addiction, 95(3), IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

164 Substance Abuse and Mental Health Services Administration (SAMHSA). (2005). Substance Abuse Treatment for Adults in the Justice System: A treatment Improvement Protocol TIP 44. Washington, D.C.: U.S. Department of Health and Human Services, Center for Substance Abuse Treatment. Treatment Research Institute. The Addiction Severity Index Alcohol Use Disorder Identification Test (AUDIT) Brief description The AUDIT was developed by the World Health Organization (WHO) to identify persons whose alcohol consumption has become hazardous or harmful to their health. The AUDIT is a 10-item screening questionnaire with three questions on the amount or frequency of drinking, three questions on alcohol dependence, and four questions on problems caused by alcohol. The AUDIT screening procedure is linked to a decision process that includes brief intervention with heavy drinkers, or referral to specialized treatment for patients who show evidence of more serious alcohol involvement. A French language translation is available. Type of administration > Pencil-and-paper self-administered > Interview > Computer self-administered Administered by > Health professional Number of items > 10 Time required for administration > 2 minutes Training required for administration Scoring > Yes, there is a user s manual and a videotape training module that explains proper administration procedures, scoring interpretation, and clinician management. > An easy-to-use brochure has been designed to guide the interviewer and to assist with scoring and interpretation. RISK ASSESSMENT INSTRUMENTS 63

165 Summary of psychometrics (reliability/validity) AUDIT s test-retest reliability has shown good temporal stability (r =.88) (Daeppen et al. 2000). Internal reliability (referring to the extent to which a measure is consistent within itself) has been consistently strong, with Cronbach s alpha scores in the range of (Allen et al. 1997; Bohn et al. 1995; Shields and Caruso 2003). AUDIT scores have been used to predict alcohol-related physical disorders and social problems (Conigrave et al. 1995a; Conigrave et al. 1995b). Similarly, the AUDIT score was also shown to be a better predictor of subsequent alcohol-related medical and social problems than standard biochemical markers (Conigrave et al. 1995b). The psychometric properties of the AUDIT have been explored in a number of populations, including inpatient care, rural and urban communities, emergency room patients, the unemployed, and college students (Reinert and Allen 2002). Research shows that the AUDIT may be especially useful when screening women and minorities (Reinert and Allen 2002). A 2007 meta-analysis of 19 relevant studies (Berner et al.) reported that sensitivity ranged from.31 to.89 and specificity ranged from.83 to.96 across the eight studies conducted in primary care. A single trial in general hospital inpatients found a sensitivity of.93 and a specificity of.94; another trial in emergency department patients found a sensitivity of.72 and a specificity of.88. A study involving university students found a sensitivity of.82 and a specificity of.78. Three studies in elderly patients found sensitivities between.55 and.83 at a pooled specificity of.96. The authors concluded the large heterogeneity between results could only partly be explained by setting diversity (Berner et al. 2007). Limitations > Limited to alcohol screening > Not enough research has been completed to determine precise cut-off points > Designed for early detection of alcohol problems in the general population Cost > Test and manual are free > Training module costs $75.00 Source World Health Organization Division of Mental Health & Prevention of Substance Abuse CH-1211, Geneva 27, Switzerland Website: 64 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

166 References Allen, J.P., Litten, R.Z., Fertig, J.B., & Babor, T. (1997). A review of research on the Alcohol Use Disorders Identification Test (AUDIT). Alcoholism: Clinical and Experimental Research, 21(4), Babor, T.F., Biddle-Higgins, J.C., Saunders, J.B., & Monteiro, M.G. (2001). AUDIT: The Alcohol Use Disorders Identification Test: Guidelines for Use in Primary Health Care. Geneva, Switzerland: World Health Organization. Berner, M.M., Kriston, L., Bentele, M., & Härter, M. (2007). The Alcohol use disorders identification test for detecting at-risk drinking: A systematic review and meta-analysis. Journal of Studies on Alcohol and Drugs, 68, Bohn, M.J., Babor, T.F., & Kranzler, H.R. (1995). The Alcohol Use Disorders Identification Test (AUDIT): Validation of a screening instrument for use in medical settings. Journal of Studies on Alcohol, 56, Conigrave, K.M., Hall, W.D., & Saunders, J.B. (1995). The AUDIT questionnaire: Choosing a cut-off score. Addiction, 90, Conigrave, K.M., Saunders, J.B., & Reznik, R.B. (1995). Predictive capacity of the AUDIT questionnaire for alcohol-related harm. Addiction, 90, Daeppen, J.B., Yersin, B., Landry, U., Pecoud, A., & Decrey, H. (2000). Reliability and validity of the Alcohol Use Disorders Identification Test (AUDIT) embedded within a general health risk screening questionnaire: Results of a survey in 332 primary care patients. Alcoholism: Clinical and Experimental Research, 24, Reinert, D.F., & Allen, J.P. (2002). Alcohol Use Disorders Identification Test (AUDIT): A review of recent research. Alcoholism: Clinical and Experimental Research, 26(2), Saunders, J.B., Aasland, O.G., Babor, T.F., de la Fuente, J.R., & Grant, M. (1993). Development of the Alcohol Use Disorders Screening Test (AUDIT). WHO collaborative project on early detection of persons with harmful alcohol consumption. II. Addiction, 88, Shields, A.S., & Caruso, J.C. (2003). Reliability generalization of the Alcohol Use Disorders Identification Test. Educational and Psychological Measurement, 63(3), Inventory of Drug-Taking Situations (IDTS) Brief description The IDTS, developed by Annis and Martin (1985), is a 50-item self-report questionnaire that provides a profile of the situations in which a client has used alcohol or another drug over the past year. The IDTS is a treatment-planning tool that provides a profile of a client s highrisk situations for drinking (or other drug use) that can be used in the development of an individual treatment plan. It is a parallel instrument to the Inventory of Drinking Situations (IDS). Clients are asked to indicate their frequency of heavy drinking or drug use in each of 50 situations on a four-point scale ranging from never to almost always. The questionnaire RISK ASSESSMENT INSTRUMENTS 65

167 may be administered in either pencil-and-paper or computerized version; the latter allows a client to name up to three substances that are currently causing a problem; the 50 IDTS items are presented for each substance in turn, and a computer-generated report is produced for each substance. Eight subscales are used, providing a profile of the client s use across eight types of high-risk situations: unpleasant emotions, physical discomfort, pleasant emotions, testing personal control, urges and temptations, conflict with others, social pressure to use, and pleasant times with others (Marlatt and Gordon 1980; 1985). A French language translation is available. Type of administration > Pencil-and-paper self-administered > Computer self-administered Administered by > Self Number of items > 50 Time required for administration > 10 minutes Training required for administration > No, detailed instruction for administration and scoring are given in the User s Guide. The software version presents instructions for administration on-screen and provides instantaneous scoring and presentation of the client s profile. Scoring > Can be done by hand or computer Summary of psychometrics (reliability/validity) The IDTS is a well-validated assessment tool that has been used in a wide array of clinical and treatment contexts. This instrument is appropriate for use in both individual and group programs, and with clients whose substance problems range from mild to severe. Within some programs, the IDTS is used to provide an individualized profile of a client s drug and/or alcohol use. This information is used to plan treatment, identify and address high-risk triggers for relapse, and assist in planning for aftercare. Turner et al. (2007) established validity evidence for the IDTS by demonstrating correlations with measures of drug consumption, problem severity, and dependence. Clients who 66 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

168 reported drinking heavily or using drugs across situations on the IDTS also reported higher levels of consumption and greater problem severity (i.e., years of problematic use, perceived seriousness of the problem, and perceived difficulty quitting). There has been a strong pattern of correlations observed between IDTS scores and measures of dependence, such as DAST and ADS (Skinner 1982; Skinner and Horn 1984), which provides further external validity evidence for the IDTS. In addition to a strong association with IDTS total score, high levels of drug dependence were most strongly associated with elevations on the IDTS negative subscales (i.e., Unpleasant Emotions, Physical Discomfort, and Conflict with Others) and with the Urges and Temptations to Use subscale. These results are in agreement with previous findings involving cluster analysis of scores on the IDS; high negative profile clients were found to have higher alcohol dependence scores (Annis and Graham 1995). Internal consistency (measuring whether several items that propose to measure the same general construct produce similar scores) values for each subscale range from , and most were over.80 for a sample of incarcerated offenders (Addictions Research Foundation 1998). Limitations > Limited in scope (because it focuses on drug use) but useful in examining specific aspects of substance use. Cost IDTS user s guide > ISBN pages paperback > $34.95 published 1997 product code PG082 IDTS alcohol questionnaires (package of 30) > $16.45 published 1997 product code P162 IDTS drug questionnaires (package of 30) > $16.45 published 1997 product code P163 Source The Centre for Addiction and Mental Health 33 Russell Street Toronto, Ontario, M5S 2S1 Phone: ext publications@camh.net RISK ASSESSMENT INSTRUMENTS 67

169 Website: situations.html References Annis, H.M., & Graham, J.M. (1988). Situational Confidence Questionnaire (SCQ-39): User s Guide. Toronto: Addiction Research Foundation. Annis, H.M., & Graham, J.M. (1995). Profile types on the Inventory of Drinking Situations: Implications for relapse prevention counseling. Psychology of Addictive Behaviour, 9(3), Annis, H., Sklar, S.M., & Turner, N.E. (1997). IDTS, Inventory of Drug Taking Situations: User s Guide. Toronto: Addiction Research Foundation. Annis, H., Turner, N.E., & Sklar, S.M. (1997). DTCQ, Drug-Taking Confidence Questionnaire: User s Guide. Toronto: Addiction Research Foundation. Annis, H.M., & Martin, G. (1985). Inventory of Drug-Taking Situations. Toronto: Addiction Research Foundation. Annis, H.M., Turner, N.E., & Sklar, S.M. (1997). Inventory of Drug-Taking Situations: User s Guide. Toronto: Addiction Research Foundation, Centre for Addiction and Mental Health. Marlatt, G.A., & Gordon, J.R. (1980). Determinants of relapse: Implications for the maintenance of behaviour change. In P.O. Davidson and S.M. Davidson (Eds.), Behavioural Medicine: Changing Health Lifestyles, (pp ). New York: Brunner-Mazel. Marlatt, G.A., & Gordon, J.R. (1985). Relapse Prevention: Maintenance Strategies in the Treatment of Addictive Behaviours. New York: Guilford Press. Skinner, H.A. (1982). The Drug Abuse Screening Test. Addictive Behaviours, 7, Skinner, H.A., & Horn, J. (1984). Alcohol Dependence Scale: User s Guide. Toronto: Addiction Research Foundation. Turner, N.E., Annis, H.M., & Sklar, S.M. (1997). Measurement of antecedents to drug and alcohol use: Psychometric properties of the Inventory of Drug-Taking Situations (IDTS). Behaviour Research and Therapy, 35(5), Drug Abuse Screening Test (DAST) Brief description The DAST was developed in It is constructed similarly to the earlier MAST, and the DAST items tend to be parallel with those of the MAST. The purpose of the DAST is to provide a brief, simple, practical, but valid method for identifying individuals who are abusing psychoactive drugs and to yield a quantitative index score of the degree of problems related to drug use and misuse. Respondents are instructed that drug abuse refers to the use of prescribed or over-the-counter drugs in excess of the directions and any non-medical use of drugs. 68 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

170 Since the DAST is one of the few instruments for assessment of drug use and related problems that has reported the relationship of the scores obtained to diagnosis of abuse, it may be of interest to those programs that are more diagnostically or psychiatrically oriented. The DAST provides a score that should be sensitive to changes in substance using experiences over a 6 and 12-month follow-up period. The questions do not refer to the use of alcoholic beverages. Type of administration > Pencil-and-paper self-administered > Interview > Computer self-administered Administered by > Self Number of items > There is a 10-item version, a 20-item version, and a 28-item version Time required for administration > 5 to 20 minutes depending on the version Training required for administration > For a qualified drug counsellor, only a careful reading and adherence to the instructions in the DAST Guidelines for Administration and Scoring, which is provided, is required. No other training is required. Scoring > Administrator or by computer Summary of psychometrics (reliability/validity) The DAST has been evaluated and demonstrated excellent reliability and diagnostic validity in a variety of populations and settings. The DAST has been shown to have good internal consistency reliability (measuring whether several items that propose to measure the same general construct produce similar scores; 28-item DAST; Cronbach s alpha =.92; 20-item DAST, Cronbach s alpha =.95) and criterion validity (assessing the correspondence between the score on the instrument and the scores on selected outcome variables; Skinner 1982). It was found to correlate highly with the ASI (Skinner and Goldberg 1986). Scores have also been found to correlate highly with the RISK ASSESSMENT INSTRUMENTS 69

171 frequency of use for a range of drugs including cannabis, barbiturates, amphetamines, and opiates. DAST scores also discriminated accurately between alcohol and drug problems (Appleby et al., 1997). The authors suggest a cut-off score of 5/6 for optimum sensitivity and specificity on the 28-item DAST. Similarly, a cut-off score of 3 on the 10-item DAST correctly classified 93% of patients (Bohn et al. 1991). In a recent meta-analysis, the DAST was found to be an easy to administer, reliable, and valid tool with good sensitivity and specificity. In general, all versions of the DAST yielded satisfactory levels of reliability and validity for use as clinical or research tools (Yudko et al. 2007). Internal reliability (referring to the extent to which a measure is consistent within itself) was consistently high ( ) for each version of the DAST. A test-retest (measures stability of the scores over time) correlation coefficient of 0.85 was reported for DAST-28, 0.78 for DAST-20, 0.71 for DAST-10, and 0.89 for an adolescent version (DAST-A). A review also found evidence supporting the construct, criterion, and discriminant validity of the DAST (Yudko et al. 2007). Research has evaluated the DAST with various populations and settings including psychiatric patients (Cocco and Carey 1998; Maisto et al. 2000; Staley and El Guebaly 1990), prison inmates (Peters et al. 2000), substance abuse patients (Gavin et al. 1989), primary care (Maly 1993), in the workplace (El-Bassel et al. 1997), and been adapted for use with adolescents (Martino et al. 2000). Overall, these studies support the reliability and diagnostic validity of the DAST in diverse contexts. Limitations > Does not screen for alcohol use/abuse > Since the content of the items is obvious, clients may fake results > Scores may be misinterpreted > Should NOT be administered to persons actively under the influence of drugs or who are undergoing drug withdrawal reaction Cost > $18.95 booklet/guide + pad of 100 questionnaires» published: 1992 product code: PZ077 > $14.95 pad of 100 questionnaires only» published: 1992 product code: PZ IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

172 Source and copyright Copies of the DAST may be obtained from H. Skinner, Centre for Addiction and Mental Health, 33 Russell Street, Toronto, Ontario, Canada M5S 1A8, Telephone: A computerized version of the DAST is included in the Computerized Lifestyle Assessment (Skinner 1994) published by Multi-Health Systems, Toronto ( Telephone: in Canada References Appleby, L., Dyson, V., Luchins, D.J., et al. (1997) The impact of substance use screening on a public psychiatric inpatient population. Psychiatric Services, 48, Bohn, M.J., Babor, T.F., & Kranzler, H.R. (1991). Validity of the Drug Abuse Screening Test (DAST-10) in inpatient substance abusers: Problems of drug dependence. Proceedings of the 53rd Annual Scientific Meeting, The Committee on Problems of Drug Dependence, Inc., DHHS Publication No. (ADM) NIDA Research Monograph, vol (pp. 233). Rockville, MD: Department of Health and Human Services. Cocco, K.M., & Carey, K.B. (1998). Psychometric properties of the Drug Abuse Screening Test in psychiatric outpatients. Psychological Assessment, 10, El-bassel, N., Schilling, R.F., Schinke, S., et al. (1997). Assessing the utility of the Drug Abuse Screening Test in the workplace. Research on Social Work Practice, 7, Gavin, D.R., Ross, H.E., & Skinner, H.A. (1989). Diagnostic validity of the DAST in the assessment of DSM-III drug disorders. British Journal of Addiction, 84, Maisto, S.A., Carey, M.P., & Carey, K.B., et al. (2000). Use of the Audit and the DAST-10 to identify alcohol and drug use disorders among adults with a severe and persistent mental illness. Psychological Assessment, 12, Maly, R. (1993). Early recognition of chemical dependence. Primary Care, 20, Martino, S., Grilo, C.M., & Fehon, D.C. (2000). Development of the Drug Abuse Screening Test for adolescents (DAST-A). Addictive Behaviours, 25, Peters, R.H., Greenbaum, P.E., Steinberg, M.L., et al. (2000). Effectiveness of screening instruments in detecting substance use disorders among prisoners. Journal of Substance Abuse Treatment, 18, RISK ASSESSMENT INSTRUMENTS 71

173 Skinner, H.A. (1982). The Drug Abuse Screening Test. Addictive Behaviours, 7, Skinner, H.A. (1994). Computerized Lifestyle Assessment Manual. Toronto: Multi-Health Systems. Skinner, H., & Goldberg, A. (1986). Evidence for a drug dependence syndrome among narcotic users. British Journal of Addiction, 81, Staley, D., & El Guebaly, N. (1990). Psychometric properties of the Drug Abuse Screening Test in a psychiatric patient population. Addictive Behaviours, 15, Yudko, E., Lozhkina, O., & Fouts, A. (2007). A comprehensive review of the psychometric properties of the Drug Abuse Screening Test. Journal of Substance Abuse Treatment, 32(2), Level of Service Inventory-Revised (LSI-R) Brief description The foundation of the LSI-R instrument is entrenched in social and psychological theories that explain the propensity towards criminal behaviour. It is a quantitative survey of attributes of offenders and their situations relevant to level of supervision and treatment decisions. Designed for ages 16 and older, the LSI-R helps predict parole outcomes, success in correctional halfway houses, institutional misconducts, and recidivism. The 54 items are based on legal requirements and include relevant factors needed for making decisions about risk and treatment. The LSI-R has ten domains including criminal history, education/employment, financial, family/marital, accommodation, leisure/recreation, companions, alcohol/drug problem, emotional/personal, and attitudes/orientation. The LSI-R Manual explains the use of the LSI-R and summarizes research studies on its reliability and validity. The LSI-R can be used by probation and parole officers and correctional workers at jails, detention facilities, and correctional halfway houses to assist in the allocation of resources, help make decisions about probation and placement, make appropriate security level classifications, and assess treatment progress. Type of administration > Interview Administered by > Health professional Number of items > 54 Time required for administration > minutes 72 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

174 Training required for administration > A professional with advanced training in psychological assessment or a related discipline must assume responsibility for the use, interpretation, and communication of the results. Scoring > Administrator or by computer Summary of psychometrics (reliability/validity) According to the manual the LSI-R has strong reliability and validity, which has been demonstrated in the many studies presented in the Technical Manual. The following psychometrics come from the LSI-R User s Manual including details on the reliability and validity of the LSI-R assessment (Andrews and Bonta 2001). The test-retest reliability (measures stability of the scores over time), which is consistent over the short term, can be seen because many items are dynamic and the LSI-R is changeable over the long term. Internal consistency reliability (measuring whether several items that propose to measure the same general construct produce similar scores) shows mild to moderate statistically significant positive correlations. Face validity (a property of a test intended to measure something) is evident because the LSI-R items were based on practitioner input. Construct validity (the extent to which what was meant to be measured was actually measured) is shown through LSI-R scores relationship to rule violations. The LSI-R has a low false-negative rate which demonstrates discriminant validity (tests whether concepts or measurements that are supposed to be unrelated are, in fact, unrelated; Andrews and Bonta 2001). Limitations > The LSI-R is somewhat effective in predicting the risk for offenders; it was developed for the purpose of correctional management, not for correctional counseling. > Does not include items that assess how offenders perceive themselves and interactions with others and how they explain their conflicts. Cost > LSI-R Complete Kit - $484.00» Includes Manual; 25 Interview Guides; 25 Forms; 25 Profile Forms > Hand Scoring Materials > LSI-R Manual - $ > LSI-R Interview Guides (25) - $ RISK ASSESSMENT INSTRUMENTS 73

175 > LSI-R Forms (25) - $ > LSI-R ColorPlot Profile Forms (25) - $55.00 > LSI-R Training DVD Series - $660.00» A training DVD series presenting a non-technical approach to using the LSI-R. > LSI-R Trainer Workbook - $93.50 > Computer Scoring Materials» LSI-R Manual - $143.00» LSI-R Data Entry Sheets (50) - $66.00» Pack of 50. Optional for use when computer scoring. > Computer Generated Reports» LSI-R Profile Report (V5) - Min. purchase of 10 reports. Price per report. - $19.80 Source and copyright» Minimum purchase of 10 reports. Price per report. Don Andrews, Ph.D. & James Bonta, Ph.D. MHS Inc Victoria Park Ave. Toronto, Ontario M2H 3M6 Phone: or References Andrews, D.A., & Bonta, J. (2001). LSI-R User s Manual. Toronto: Multi-Health Systems. Andrews, D.A., & Bonta, J. (1994). The Psychology of Criminal Conduct. Cincinnati, OH: Anderson. Bonta, J., & Motiuk, L.L. (1985). Utilization of an interview-based classification instrument: A study of correctional halfway houses. Criminal Justice and Behaviour, 12, Andrews, D.A., Bonta, J., & Wormith, J.S. (2004). Level of Service/Case Management Inventory (LSI/CMI): An Offender Assessment System. User s Manual. Toronto: Multi-Health Systems. Andrews, D.A., Bonta, J., & Wormith, J.S. (2006). The recent past and near future of risk and/ or need assessment. Crime & Delinquency, 52(1), IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

176 6.8 Michigan Alcoholism Screening Test (MAST) Brief description The MAST is one of the most widely used measures for assessing alcohol abuse. The instrument is a 25-item questionnaire designed to provide a rapid and effective screening for lifetime alcohol-related problems and alcoholism. The MAST has been productively used in a variety of settings with varied populations. Type of administration > Pencil-and-paper self-administered > Computer self-administered Administered by > Practitioner or self Number of items > 25 Time required for administration > minutes Training required for administration > No training required Scoring > Scoring completed by staff Summary of psychometrics (reliability/validity) The original MAST validation sample of 526 included hospitalized alcoholics, drivers convicted of driving under the influence or who had amassed numerous driving penalty points, persons convicted of drunk and disorderly behaviour, and a control sample (Selzer 1971). Psychometric work includes internal consistency, predictive and concurrent validity, and factor analysis for confirmation of the purported domains. Early studies showed strong internal consistency (measuring whether several items that propose to measure the same general construct produce similar scores; Cronbach s alpha =.95) but more recent studies suggest a number of items are not highly correlated and that the instrument itself might not be measuring one factor but rather several factors related to problem-drinking (Selzer et al. 1975; Crook et al. 1994; Parsons et al. 1994; Saltstone et al. 1994). Selzer (1971) suggested a cut-off point of 5 to identify harmful or hazardous drinking. RISK ASSESSMENT INSTRUMENTS 75

177 However, a cut-off score of 13 (at which the test has sensitivity of.91 and specificity of.76) is suggested for detecting the presence of alcohol abuse and dependence (Ross et al. 1990). Reliability and validity data are available across a number of populations; internal consistency ranges from.83 to.95, while test-retest reliability (measures stability of the scores over time) values range from.84 to.97 (Kitchens 1994); lower values are associated with longer delays between administration. The original normative male-only sample covered a wide age range and assessed both clinical and non-clinical populations, and the popularity of the MAST has resulted in data available across numerous special populations, including offender populations (Millson et al. 1995; Swett 1984). Some factor analyses of the MAST have revealed four and six factors (Parsons et al. 1994); the four-factor structure has held across a number of samples, including a female offender population (Saltstone et al. 1994). However, the MAST is generally considered to be a uni-dimensional instrument. Modifications of the MAST include the 10-item Brief MAST (bmast), the 13-item Short MAST (SMAST), and the 9-item Malmo modification (Mm-MAST); these briefer instruments would seem perhaps more appropriate for screening purposes than the original 25-item scale. Connor and colleagues (2007) found the bmast to have good construct validity (the extent to which what was meant to be measured was actually measured) and both single-factor and two-factor scoring were equally effective as the AUDIT in assessing dependence severity. In a recent meta-analysis of the MAST and the SMAST, Shields et al. (2007) found that both the MAST and the SMAST observe moderate to good internal consistency reliability (measuring whether several items that propose to measure the same general construct produce similar scores) estimates. However, in individual assessment and outcome measurement where personal and social costs are considered significant, the MAST and SMAST should be used with caution. Limitations > Does not discriminate between past and present drinking (Dawe et al. 2002) > The MAST has been criticized for its obvious face validity > The MAST has little sensitivity to change, as most items are prefaced with Have you ever... Cost > $40.00 for copy, no fee for use. Source and copyright Melvin L. Selzer, M.D., 6967 Paseo Laredo, LaJolla, CA, IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

178 Phone: References Conley, T.B. (2001). Construct validity of the MAST and AUDIT with multiple offender drunk drivers. Journal of Substance Abuse Treatment, 20(4), Connor, J., Grier, M., Feeney, G., & Young, R. (2007). The validity of the Brief Michigan Alcohol Screening Test (bmast) as a problem drinking severity measure. Journal of Studies on Alcohol, 68(5), Crook, G.M., Oei, T.P.S., & Young, R.M. (1994). Structure of the MAST with an Australian sample of alcoholics. Drug and Alcohol Review, 13, Millson, W.A., Weekes, J.R., & Lightfoot, L.O. (1995, August). The effectiveness of the Offender Substance Abuse Pre-Release Program: Analysis of intermediate and post-release outcomes. Ottawa, Ontario, Canada: Correctional Research and Development, Correctional Service Canada. Parsons, K.J., Wallbrwon, F.H., & Myers, R.W. (1994). Michigan Alcoholism Screening Test: Evidence supporting general as well as specific factors. Educational and Psychological Measurement, 54, Ross, H.E., Gavin, D.R., & Skinner, H.A. (1990). Diagnostic validity of the MAST and the Alcohol Dependence Scale in the assessment of DSM-III alcohol disorders. Journal of Studies on Alcohol, 51, Saltstone, R., Halliwell, S., & Hayslip, M.A. (1994). A multivariate evaluation of the Michigan Alcoholism Screening Test and the Drug Abuse Screening Test in a female offender population. Addictive Behaviours, 19(5), Selzer, M.L. (1971). The Michigan Alcoholism Screening Test (MAST): The quest for a new diagnostic instrument. American Journal of Psychiatry, 127, Selzer, M.L., Vinokur, A., & VanRooijen, L. (1975). A self-administered short version of the Michigan Alcoholism Screening Test (SMAST). Journal of Studies on Alcohol, 36, Shields, A.L., Howell, R.T., Potter, J.S., & Weiss, R.D. (2007). The Michigan Alcoholism Screening Test and its shortened form: A meta-analytic inquiry into score reliability. Substance Use & Misuse, 42, Swett, C. Jr. (1984). Use of the Michigan Alcoholism Screening test in a prison hospital. American Journal of Drug & Alcohol Abuse, 10, Substance Abuse Subtle Screening Inventory (SASSI) Brief description The SASSI is a brief self-report, easily administered psychological screening measure that is available in separate versions for adults and adolescents. The SASSI was developed out RISK ASSESSMENT INSTRUMENTS 77

179 of concern about the potential for distortion of responses on substance abuse measures; the authors of the SASSI claim its resistance to efforts at faking. The SASSI includes both face valid and subtle items that have no apparent relationship to substance use. The subtle items are included to identify some individuals with alcohol and other drug problems who are unwilling or unable to acknowledge substance misuse or symptoms associated with it. Support materials for the SASSI include user s guides containing easy-to-understand instructions for administering, scoring, and interpretation, and manuals providing comprehensive information on development, reliability, and validity. Interpretations of the SASSI profiles suggest possible explanations that the clinician may find useful in understanding clients and providing effective feedback. Examples of clinical inferences that may be drawn on the basis of certain scale scores include indication of defensive responding, clients level of insight and awareness of the effects of their substance misuse, evidence of emotional pain, and relative risk of involvement with the legal/judicial system. In combination with other available assessment information, the clinical inferences suggested by examining SASSI profiles provide ideas for further evaluation and treatment considerations. Type of administration > Pencil-and-paper self-administered > Computer self-administered Administered by > Support staff Number of items > 93 Time required for administration > minutes Training required for administration > No Scoring > Administrator Summary of psychometrics (reliability/validity) Allen and Columbus (1995) note the effectiveness of the SASSI in identifying early stage substance abuse in those who have not yet acknowledged their patterns to themselves. 78 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

180 A recent meta-analysis by Feldstein and Miller (2007) found internal consistency (measuring whether several items that propose to measure the same general construct produce similar scores) is high for the overall SASSI and for its direct but not its indirect (subtle) subscales, suggesting that the instrument taps a single face-valid construct (a property of a test intended to measure something). SASSI classifications converged with those from other direct screening instruments, and were also correlated with ethnicity, general distress, and social deviance. Studies found test retest reliability (measures stability of the scores over time) lower than that reported in the test manuals. Sensitivity was found to be similar to that for public domain screening instruments, but on specificity the SASSI appears to yield a high rate of false-positives. Results from several studies support high internal consistency (measuring whether several items that propose to measure the same general construct produce similar scores) for the direct scales (Myerholtz and Rosenberg 1997, 1998; Clements 2002; Laux et al. 2005; Gray 2001). Additional data from these studies revealed generally lower internal consistency for the SASSI subtle scales, with high variability across samples. For the direct scales, no study reported alpha coefficients as high as those reported in the test manual (Miller and Lazowski 1999). Limitations > SASSI is quite lengthy > Limited literature that includes the SASSI > More research is needed to examine the instrument s psychometrics, since it has not been validated for an impaired driving offender population > It may be vulnerable to intentional faking Cost > Starting costs around $ and up > Call for product catalogue or visit the SASSI website Source Source: The SASSI Institute Phone: Website: RISK ASSESSMENT INSTRUMENTS 79

181 References Allen, J., & Columbus, M. (1995). Assessing Alcohol Problems: A Guide for Clinicians and Researchers. National Institute on Alcohol Abuse and Alcoholism Treatment Handbook Series 4. NIH Pub. No Bethesda, MD: NIAAA. Clements, R. (2002). Psychometric properties of the Substance Abuse Subtle Screening Inventory-3. Journal of Substance Abuse Treatment, 23, Feldstein, S.W., & Miller, W.R. (2007). Does subtle screening for substance abuse work? A review of the Substance Abuse Subtle Screening Inventory (SASSI). Addiction, 102(1), Gray, B.T. (2001). A factor analytic study of the Substance Abuse Screening Inventory (SASSI). Educational and Psychological Measurement, 61, Laux, J.M., Salyers, K.M., & Kotova, E. (2005). A psychometric evaluation of the SASSI-3 in a college sample. Journal of College Counseling, 8, Miller, F.G., & Lazowski L.E. (1999). The Adult SASSI-3 Manual. Springville, IN: The SASSI Institute. Miller, G.A. (1985). The Substance Abuse Subtle Screening Inventory (SASSI) Manual. Springville, IN: The SASSI Institute. Myerholtz, L.E., & Rosenberg, H. (1997). Screening DUI offenders for alcohol problems: Psychometric assessment of the Substance Abuse Subtle Screening Inventory. Psychology of Addictive Behaviours, 11(3), Myerholtz, L.E., & Rosenberg, H. (1998). Screening college students for alcohol problems: Psychometric assessment of the SASSI-2. Journal of Studies on Alcohol and Drugs, 59, Other SASSI References: Research Institute on Addiction Self Inventory (RIASI) Brief description The RIASI was developed for the New York State Drinking Driver Programs. It is a simple screening instrument. The RIASI covers specific risk factors as well as family history associated with alcohol and drugs. The RIASI is designed to screen for alcoholism using covert content items, i.e., items which do not directly mention drinking. A training manual is available. The inventory has three scales, one for detection of individuals with alcohol or other drug problems, a second scale for predicting impaired driving recidivism, and a three-question lie scale. Included are distal measures items, meaning the person cannot readily determine how to fake desirable versus undesirable responses to the question. Also these questions address 80 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

182 issues of hostility, sensation-seeking, depression, and other personality characteristics linked to impaired driving. The RIASI represents a careful and empirical development of a screening device for use with the impaired driving population. Developed specifically for the New York State Drinking Driver Programs, it is now being used in several states. Type of administration > Pencil-and-paper self-administered > Interview Administered by > Self or practitioner Number of items > 52 Time required for administration > minutes Training required for administration > Yes, only basic training needed Scoring > Administered by using a simple transparent overlay Summary of psychometrics (reliability/validity) The validity of RIASI has also been confirmed in the convicted drinking driver population in Ontario (Nochajski et al. 1997). In addition to a total score based on all the items on the instrument, Nochajski and colleagues developed a recidivism subscale of 15 items on the instrument that was able to correctly identify over 80% of individuals who were rearrested for drinking driving over a two-year period (Nochajski et al. 1993; Shuggi et al. 2006). Recommended cut-offs for referral of participants to more extensive follow-up were nine on the total score and three on the recidivism scale (Shuggi et al. 2006). Limitations > The authors have been engaged in research that has demonstrated some degree of validity, but more independent research is still needed > Does not have the computer automation and summary printout with treatment recommendations RISK ASSESSMENT INSTRUMENTS 81

183 Cost Source > Information on cost and material can be obtained from Thomas Nochajski (see Source) Thomas Nochajski, Ph.D. Research Society on Addiction 1021 Main Street Buffalo, NY Phone: References Nochajski, T.H. (2002). Training Manual for the Research Institute on Addictions Self Inventory (RIASI) Revised. Buffaly, N.Y.: Research Institute on Addictions. Nochajski, T.H., Miller, B.A., Wieczorek, W.F., & Parks, K.A. (1993). The Utility of Non- Obvious Indicators for Screening of DWI Offenders. Paper presented at the annual meeting of the Research Society on Alcoholism, San Antonio, Texas, June 19-24th,1993. Nochajski, T.H., Walter, J.M., & Wieczorek, W.F. (1997). Identification of drinker-driver recidivists. In C. Mercier-Guyon (Ed.), Alcohol, Drugs and Traffic Safety T 97 (pp ). Annecy, France: Centre d Etudes et de Reserches en Medecine du Traffic. Shuggi, R., Mann, R.E., Flam Zalcman, R., Chipperfield, B.G.A., & Nochajski, T. (2006). Predictive validity of the RIASI: Alcohol and drug use and problems six months following remedial program participation. American Journal of Drug and Alcohol Abuse, 32, Biomarkers Brief description Alcohol biomarkers are physiological indicators of alcohol exposure or ingestion and may reflect the presence of chronic and/or high level of use of alcohol (SAMHSA 2006). Alcohol biomarkers can be used in several ways. The major uses of biomarkers are screening for alcohol problems; motivating change in drinking behaviour; identifying relapse to drinking; evaluating interventions for alcohol problems; and documenting abstinence (SAMHSA 2006; 2012). Alcohol biomarkers are not a substitute for self-report measures found using risk assessments instruments. However, when used in combination with risk assessment instruments, biomarkers can serve as objective measures. Several biomarkers are considered useful including gamma glutamyl transferase (GGT), carbohydrate deficient transferrin (CDT), phosphatidyl ethanol (PEth), and fatty acid ethyl esthers (FAEE). These biomarkers have been investigated and found to have moderate to 82 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

184 high diagnostic sensitivity and specificity. There has been increased use of two specific biomarkers (ethyl Glucuronide (EtG) and Ethylsulfate (EtS)), particularly in the United States, which are detectable in urine. These biomarkers are direct metabolites of ethanol alcohol and have varying levels of sensitivity depending on which biomarker is used. However, it has been emphasized that urine EtG should not be utilized as a quantitative measure of alcohol use, mainly because it is impossible to predict the level of alcohol consumption using urine EtG value. The variable production of EtG can occur as a result of enzyme system variations, urine concentration variations, the amount of time since the last drink, the rate of alcohol consumption, and chronic drinking. For this reason, in 2012 the Center for Substance Abuse Treatment (CSAT) of the Substance Abuse and Mental Health Services Administration (SAMHSA) in the U.S. issued an updated advisory ( Revision.pdf) that cautions against the interpretation and use of EtG results alone to assess alcohol use. While it is recognized that the higher the EtG level, the more likely it is that drinking occurred; no clear cut-off values have been identified. Type of administration > Alcohol biomarkers used to indicate impaired driving risk include samples of blood, urine, hair, and saliva. Administered by > Technicians obtain and analyze specific biomarkers using empirically determined cut points. Number of items > N/A Time required for administration > Varies depending on which sample (blood, urine, hair, or saliva) is used. Training required for administration > Varies depending on which sample (blood, urine, hair, or saliva) is used. Scoring > Testing of the samples is analyzed using a clinical chemistry instrument within a laboratory. Summary of psychometrics (reliability/validity) The findings of Couture et al. (2010) suggest that biomarkers of chronic patterns of heavy drinking may not be adequate to capture the multiple processes that appear to promote RISK ASSESSMENT INSTRUMENTS 83

185 recidivism (e.g., binge drinking, other risky behavioural and personality features). Despite their objectivity, caution is warranted in the interpretation of a positive score on these biomarkers in an impaired driving assessment. This study found that alcohol biomarkers failed to differentiate groups (first vs. recidivists), which is inconsistent with earlier findings by Caviola et al. (2003) and McMillen et al. (1992). However, the current study used a communitybased sample whereas other studies used offender populations. Consideration of multiple biomarkers simultaneously did not significantly enhance prediction of recidivism status. A recent population-based study demonstrated that 88% of self-reported alcohol-impaired driving episodes involved binge drinking (e.g., for men an episode of five or more drinks) while 84% of the alcohol-impaired drivers were binge drinkers (Flowers et al. 2008). The results of the Couture study converge with other evidence that questions the emphasis on addiction approaches to impaired driving and its prevention for all offenders. Limitations Cost > Biomarkers provide an important indication of drinking status when used appropriately, but they must be used with a clear understanding of their strengths and potential weaknesses (SAMHSA 2006; 2012). Specific issues to be cognizant of are:» Understanding the difference between a test s sensitivity and positive predictive value;» Potential sources of false-positives;» High costs associated with testing and analyses;» High state of flux with new markers being discovered each year; and,» Many biomarkers are only detectible for relative short windows of time, meaning that the recovery time to normal levels is limited (i.e., 3 to 5 days up to 4 to 6 weeks). As such, the usefulness of biomarkers to detect alcohol consumption requires frequent testing following drinking events. > Varies depending on which sample (blood, urine, hair, or saliva) is used. References Caviola, A.A., Strohmetz, D.B., Wolf, J.M., & Lavender, N.J. (2003). Comparison of DWI offenders with non-dwi individuals on the MMPI-2 and the Michigan Alcoholism Screening Test. Addiction Behaviours, 28(5), Couture S., Brown T.G., Tremblay J., Ng Ying Kin N.M., Ouimet M.C., & Nadeau L. (2010). Are biomarkers of chronic alcohol misuse useful in the assessment of DWI recidivism status? Accident Analysis and Prevention, 42(1), IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

186 Flowers, N.T., Naimi, T.S., Brewer, R.D., Elder, R.W., Shults, R.A., & Jiles, R. (2008). Partterns of alcohol consumption and alcohol-impaired driving in the United States. Alcoholism: Clinical and Experimental Research, 32(4), McMillen, D.L., Adams, M.S., Wells-Parker, E., Pang, M.G., & Anderson, B.J. (1992). Personality trains and behaviours of alcohol-impaired drivers: A comparison of first and multiple offenders. Addiction Behaviours, 17(5), Substance Abuse and Mental Health Services Administration (SAMHSA). (2006). The Role of Biomarkers in the Treatment of Alcohol Use Disorders, Advisory, 5(4). HHS Publication No. (SMA) biomarkers.htm Substance Abuse and Mental Health Services Administration (SAMHSA). (2012). The Role of Biomarkers in the Treatment of Alcohol Use Disorders, 2012 Revision. Advisory, 11(2). HHS Publication No. (SMA) Advisory_Biomarkers_Revision.pdf 6.12 Summary There are many impaired driver assessment instruments that are available and utilized across North America. Yet not all of these instruments have been validated on an impaired driver population and few have undergone rigorous or independent evaluation efforts. It is for this reason that many jurisdictions rely upon a combination of these instruments to guide the assessment process. It is essential to underscore that problem substance use behaviour in and of itself is not the source or cause of persistent impaired driving behaviour, but instead merely a correlate of it. So while assessment instruments designed to identify the likelihood of relapse among substance using and even impaired driving populations provide valuable information, these tools frequently overlook the role of criminogenic and socio-psychological factors that are important contributors to chronic offending. Of the available risk assessment instruments to date, both the LSI-R and ASUS 16 instruments appear to be the most well-grounded in theory and based upon a solid theoretical foundation. These instruments incorporate a range of recognized concepts stemming from several relevant disciplines including criminology, psychology, sociology and addictions, and these concepts have been repeatedly tested and validated through extensive research. Such a comprehensive approach is essential in light of the well-documented complexity associated with impaired driving behaviour and the diversity of underlying processes that have been used to explain persistent offending by this population. It should be underscored that assessment approaches that are multi-trait and multi-method provide more accurate results (Campbell and Fiske 1959). 16 The Adult Substance Use Survey (ASUS) is a self-report survey that consists of 64 items designed to assess an individual s perceived alcohol and drug use. The survey also provides a brief mental health screen. It can either be self-administered (paper-and-pencil) or administered orally by a practitioner. Unlike the ASUDS-R, this screening instrument is not specific to an impaired driving offender population although both tools were developed by the Center for Addiction Research and Evaluation (CARE). RISK ASSESSMENT INSTRUMENTS 85

187 Looking forward, there is some clear direction as to ways to strengthen research that can guide the development of empirically-based risk assessment instruments. First, with regard to the evaluation of risk assessment instruments, Brown and Ouimet (2013) underscore that Longer duration perspective evaluations of assessment protocols for prediction of recidivism are urgently needed (p.311). Second, the research undertaken by Dugosh et al. (2013) provides a basis to begin to integrate criminological theories and empirically-based risk factors to enhance risk assessment tools for impaired drivers. The inclusion of these factors in risk assessment tools can help to strengthen the internal validity of such tools. 86 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

188 7. TREATMENT INTERVENTIONS Educational approaches to impaired driver programs have been utilized to target impaired drivers for more than four decades. A number of these programs have been evaluated and several comprehensive reviews have been produced, including a meta-analysis that reveals that these programs have limited effects in terms of reducing recidivism. Generally, these studies show an average reduction in recidivism of approximately 10% (NHTSA 1986; Wells- Parker et al. 1995). Among offenders who suffered from some degree of substance misuse problems, those programs that utilized a therapeutic approach are considered to have a greater effect, illustrating the value of treatment as an intervention to encourage rehabilitation and behavioural change (Wanberg et al. 2005). As a general caveat, available interventions generally require offenders to be committed to addressing their substance use problems, and this often requires hard work on their part. This is perhaps most clearly illustrated by the fact that, according to probation officers across the United States, many impaired drivers elect to serve time in prison rather than enrol in treatment. It is underscored that offenders may be reluctant (to varying degrees) to participate in treatment because they are often challenged and taken out of their comfort zones in order to tackle substance misuse problems. In the criminal justice literature, a strong emphasis is placed on the Risk Needs Responsivity model of offender rehabilitation (Ward et al. 2007). This model guides the selection of appropriate rehabilitation decisions for individual offenders according to three key principles: 1) the risk principle acknowledges that intensive services should be reserved for higher risk offenders; 2) the need principle recognizes that in order to reduce re-offending interventions must specifically target offender needs (Bonta et al. 2000; Ogloff and Davis 2004; Andrews and Dowden 2006); and, 3) the responsivity principle emphasizes the importance of designing and delivering treatment using strategies that accommodate offenders learning style, ability, ethnicity, and sex. The key feature of this model is matching an offender to an intervention based on their propensity or risk to re-offend (Ogloff and Davis 2004). This TREATMENT INTERVENTIONS 87

189 has been a dominant model that has influenced the development of offender treatment programs, for more than two decades, with research that has shown that programs that incorporate these principles are more effective than programs that do not (Dowden and Andrews 1999). The results of a risk assessment in conjunction with resources that are available are two critical components of any intervention strategy. It is these two factors that ultimately determine what types and to what degree treatment interventions are made available to offenders, how services are delivered and managed, and the skills and experience of staff that deliver these interventions. A majority of treatment agencies are equipped to provide a range of interventions that incorporate diverse techniques and approaches. Assessment results are used to match the most appropriate services (of those available) to individual offenders. There is growing evidence to suggest that combining appropriate sanctions and supervision with treatment interventions can be more effective than either strategy alone. The partnering of these different strategies can expand opportunities to achieve long-term risk reduction and to reduce and/or prevent repeat offending. In order to maximize the effectiveness of this approach it must be assessment driven and combine appropriate levels of supervision with appropriate treatment interventions. This section briefly describes a variety of common approaches to treatment including screening and brief interventions (SBI), motivational interviewing (MI), cognitive behavioural therapy (CBT), pharmacological interventions, and web-based interventions. Each intervention is described in terms of purpose and objectives, general effectiveness, staff training requirements, mechanism of delivery, and strengths and weaknesses. Note that some of these interventions have been specifically evaluated on an impaired driving population whereas others are merely a source of emerging interest and more research is needed to gauge effectiveness with impaired drivers. In addition, a few key references are identified in relation to each intervention in order to provide additional information to practitioners seeking more knowledge about specific strategies. 7.1 Screening and Brief Interventions (SBI) Purpose and objectives. SBI is a structured set of questions designed to identify individuals at risk for alcohol use problems, followed by a brief discussion between an individual and the treatment clinician or provider, with referral to specialized treatment as needed. A brief intervention consists of one or more time-limited conversations between a client and a clinician. Screening asks several questions to determine whether clients are misusing alcohol; that is, are they drinking too much, too often, or experiencing harm from their drinking? The provider evaluates the answers and then shares the results and their significance with the individual. The goals are to (1) help the drinker increase awareness of his or her alcohol use 88 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

190 and its consequences, and (2) encourage the person to create a plan to change his or her drinking behaviour to stay within safe limits. The conversations are typically 5-15 minutes, although they can last up to minutes for as many as four sessions (NHTSA 2005). Brief interventions, as the name implies, are much smaller in number and shorter in duration than traditional treatment approaches. Numerous types of brief interventions have been developed, ranging from providing advice to individuals to cut down on or quit drinking, to agreement on goals and objectives, to brief screening and feedback, motivational interventions, and contingency contracting. One of the more simple forms of brief intervention is screening itself. Given that screening often involves contact with the client in the context of questions and issues related to drinking behaviours, it can have some impact on the offender s behaviour. This makes screening not only a valuable tool for determining the nature and extent of alcohol problems but also a part of the therapeutic process itself. SBIs are increasingly being applied in a variety of settings and are recommended for offenders who misuse alcohol and are at risk for dependence but who are not yet alcohol dependent (Lapham 2004; 2005). General effectiveness. Brief interventions have been increasingly utilized as part of remedial programs for impaired drivers with alcohol-related problems. Studies conducted in the United States, Australia, Bulgaria, Mexico, the United Kingdom, Norway, and Sweden show that there is clear evidence that well-designed brief intervention strategies are effective, costefficient, and easy to administer (WHO 2010; Davis et al. 2012). The effectiveness of brief interventions has been demonstrated in various setting (e.g., Moyer et al. 2002; Poikolainen 1999), but few studies have examined the benefits with criminal justice populations. The exception seems to be a brief motivational intervention (usually in the form of feedback regarding test results and diagnosis), which has been shown to produce significant benefits in a criminal justice population (Moyer et al. 2002; Poikolainen 1999; McMurran et al. 2011). There is some evidence that recidivists who are younger, male, and exhibit more negative consequences and ambivalence towards their problem drinking show the most improvement as a result of SBI as compared to other groups (Brown et al. 2012). Staff training requirements. SBI does not require investment in extensive training or expensive instruments, and does not require lengthy amounts of time to conduct (APHA 2008). Screening can be done with a minimal amount of training depending on the screening tool(s) utilized. The process can be included in routine training and ongoing staff development. Mechanism of delivery. SBI can be offered within the criminal justice system or more commonly in remedial programs. TREATMENT INTERVENTIONS 89

191 Strengths and weaknesses. Strengths > Brief interventions are low in cost. > They can serve as treatment for hazardous and harmful drinkers, and as a way to facilitate referral of more serious cases of alcohol dependence to specialized treatment. > They require minimal clinician and client time. Weaknesses > Brief interventions (excluding motivational interviewing) are not designed to treat persons with alcohol dependence. > They can provoke clinician apprehension in primary care settings. Common concerns are that screening and brief intervention will require too much time and can antagonize clients over a sensitive personal issue. References Davis, H.T., Beaton, S.J., Von Worley, A., Parsons, W., & Gunter, M.J. (2012). The effectiveness of screening and brief intervention on reducing driving while intoxicated citations. Population Health Management, 15(1), Moyer, A., Finney, J.W., Swearingen, C.E., & Vergun, P. (2002). Brief interventions for alcohol problems: A meta-analytic review of controlled populations. Addiction, 97, Brown, T. G., Dongier, M., Ouimet, M. C., Tremblay, J., Chanut, F., Legault, L., & Kin, N. M. (2012). The role of demographic characteristics and readiness to change in 12-month outcome from two distinct brief interventions for impaired drivers. Journal of Substance Abuse Treatment, 42(4), Guides American Public Health Association and Education Development Center, Inc. (2008). Alcohol Screening and Brief Intervention: A Guide for Public Health Practitioners. Washington D.C.: National Highway Traffic Safety Administration, U.S. Department of Transportation. Available online: Higgins-Biddle, J., Hungerford, D., & Cates-Wessel, K. (2009). Screening and Brief Interventions (SBI) for Unhealthy Alcohol Use: A Step-By-Step Implementation Guide for Trauma Centers. Atlanta: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Available online: Lapham, S. (2004/2005). Screening and brief intervention in the criminal justice system. Alcohol Research and Health, 28(2), Retrieved from: publications/arh28-2/85-93.pdf 90 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

192 National Highway Traffic Safety Administration (NHTSA). (2007). Screening and Brief Intervention Tool Kit for College and University Campuses. DTNH22-02-H Washington, D.C.: U.S. Department of Transportation. Available online: friendsdrivesober.org/documents/sbi_college.pdf World Health Organization (WHO). (2010). Screening and Brief Intervention for Alcohol Problems in Primary Health Care. Retrieved from: MSD_MSB_01.6b.pdf 7.2 Motivational Interviewing (MI) Purpose and objectives. MI is one form of brief intervention. MI is a collaborative, goaloriented method of communication with particular attention to the language of change. It is designed to strengthen an individual s motivation for and movement toward a specific goal by eliciting and exploring the person s own arguments for change. The practice of MI involves the skilful use of certain techniques for bringing to life the MI spirit, demonstrating the MI principles, and guiding the process toward eliciting client change talk and commitment to change. Change talk involves statements or non-verbal communications indicating the client may be considering the possibility of change (Miller and Rollnick 2010). This tool can be very important in keeping reluctant clients in treatment. More recently, Miller and Rollnick expanded upon their work in the 3 rd Edition of their manual to further elaborate on the four key processes of MI (engaging, focusing, evoking and planning). It also contains work related to the use of MI in group settings as well as efforts to combine MI, and CBT to increase effectiveness. General effectiveness. There is 17 years of research on MI, beginning when the method was developed by Rollnick and Miller (1995) as a client-centred style of counselling that helps clients to explore and resolve their ambivalence about changing their behaviour. A metaanalysis conducted by Dunn et al. (2001) examined 29 randomized trials of MI and concluded that on average it took 15 hours to learn and deliver MI. Sixty percent of the 29 studies yielded at least one significant behavioural change effect size. There was substantial evidence that MI is an effective substance abuse intervention method when used by clinicians who are non-specialists in substance abuse treatment, particularly when enhancing entry to and engagement in more intensive substance abuse treatment (Dunn et al. 2001; Palmer et al. 2011). Additional studies have found that although MI may not be more effective than other addiction treatment approaches, it does work faster in remedying the client s addiction (Chanut et al. 2005). When provided as the sole treatment, MI can lead to improvements in outcomes that compare with those seen in a 12-step Alcoholics Anonymous (AA) program and in longer, more intensive cognitive-behavioural treatment interventions (Project MATCH Research Group 1997). TREATMENT INTERVENTIONS 91

193 A recent study by Brown et al. (2010) suggests that MI may be more appropriate for impaired driving recidivists who were unmotivated, reluctant or resistant to participate in treatment, and who failed to acknowledge or recognize their problem(s) with alcohol. In particular, the brief nature of this strategy makes it easier to utilize with those hard-to-reach individuals who do not readily participate in impaired driver re-licensing programs. More recently, Brown and Ouimet (2013) also noted that, although initial studies investigating the use of MI with impaired driving populations have shown promising results, methodological differences across studies have made it difficult to generalize findings and to gauge which features or content associated with MI applications result in positive outcomes. Staff training requirements. Training for MI varies. The most common method clinicians explore is to study print materials and view training videotapes. Although this can provide some understanding of the basic approach, self-training was not found to be effective in improving clinical skillfulness in MI (Miller and Rollnick 2010). Training of up to one day can acquaint the audience with basic concepts and methods of MI, but is unlikely to increase the clinical skilfulness of participants in the practice of MI. With the hours of training time, participants are provided with more in-depth understanding of the method of MI, and offered practical experience in trying this approach. Continuing education is also available. Mechanism of delivery. MI is most often offered in remedial programs. However, it can be offered within the criminal justice system with proper training. These one-on-one patientcentred, non-confrontational counselling sessions are brief, and may be used in at least three different stages of an offender s processing. First, if an offender screens positively for alcohol use problems, a health care professional can share the screening results and their significance with the offender in a short, minute interview. These are patient-centred and encourage the offender to create a plan of action which ranges from reducing their drinking to seeking substance abuse treatment (NHTSA 2005). Second, offenders who have been assessed as being unready to receive treatment may also be engaged in motivational interviewing, where the focus is on facilitating an offender s readiness for self-change or motivation to treatment (Marques and Voas 2005). The idea is to encourage the offender through engagement so they can accept their problem(s), understand the benefits of being treated for the problem, and then access the necessary services that are designed to help them overcome the problem. The premise of this technique is for professional staff to build a rapport with offenders and empower them to change on their own (Taxman et al. 2004). Third, MI is also useful throughout the supervision process for providing critical feedback to reinforce progress by helping offenders learn to analyze their own attitudes and behaviour and determine how they can advance their behavioural change (Taxman et al. 2004). Such aftercare programs may involve weekly counsellor-led sessions, offered at treatment sites (Harrison and Asche 2001). 92 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

194 Strengths and weaknesses. Strengths > Useful with clients in early stages of change. > Draws out the client s own ideas; based on the belief that the motivation to change comes from the client, not the clinician. Weaknesses > Change may not happen immediately. > Outside influences may be stronger, if a client returns home with peers and daily life pressures, motivation to change may cease. > Not all clinicians are willing to change their intervention approach in line with the practice of MI. > Maintenance of MI fidelity requires constant surveillance and quality assurance efforts. References Brown, T.G., Dongier, M., Ouimet, M.C., Tremblay, J., Chanut, F., Legault, L., & Ng Yin Kin, N.M.K. (2010). Brief motivational interviewing for DWI recidivists who abuse alcohol and are not participating in DWI intervention: A randomized controlled trial. Alcoholism: Experimental and Clinical Research, 34(2), Chanut, F., Brown, T., & Dongier, M. (2005). Motivational interviewing and clinical psychiatry. Canadian Journal of Psychiatry, 50(9), Dunn, C., Deroo, L., & Rivara, F.P. (2001). The use of brief interventions adapted from motivational interviewing across behavioural domains: A systematic review. Addiction, 96(12), Miller, W.R., & Rollnick, S. (2010). What s new since MI-2? Presentation at the International Conference on Motivational Interviewing (ICMI). Stockholm, June 6th, Miller, W., & Rollnick, S. (2012). Motivational Interviewing: Helping People Change (3rd ed.). New York: Guilford Press. National Highway Traffic Safety Administration (NHTSA). (2005). Toward a Comprehensive Strategy to Stop Impaired Driving: Alcohol Screening and Brief Intervention Overview. DOT HS Washington, D.C.: U.S. Department of Transportation. Available online: stopimpaireddriving.org/alcscreenweb2005/pages/alcoholscreen.pdf Palmer, E.J., Hatcher, R.M., McGuire, J., Bilby, C.A.L., & Hollin, C.R. (2011). The effect on reconviction of an intervention for drink-driving offenders in the community. International Journal of Offender Therapy and Comparative Criminology. Retrieved from: com.prox.lib.ncsu.edu/content/early/2011/04/20/ x full.pdf+html TREATMENT INTERVENTIONS 93

195 7.3 Cognitive Behavioural Therapy (CBT) Purpose and objectives. CBT is a form of psychosocial therapy with an action-oriented perspective. CBT encompasses a wide range of cost-effective psychotherapeutic approaches that deal with cognitions and beliefs as a means to reducing problematic behaviours (Beck 1993). The objective of this approach is to identify thoughts, assumptions, beliefs, and behaviours that are related to negative emotions and underlying dysfunctional problems (e.g., drinking problems) and to replace these with more realistic and functional ones. Ultimately, the goal is to change an individual s thoughts in order to change their behaviour. General effectiveness. A number of studies support the effectiveness of CBT in treating alcohol abuse: > Longbauch et al. (1999) found that alcohol abusers who received CBT had better drinking-related outcomes than those who did not receive therapy. > More than 24 randomized control trials found CBT to be comparable to or more effective than other treatments for alcohol abuse (Carroll 1996). > Carroll (1998) also found that CBT was particularly effective in reducing the severity of relapse. > Offering offenders with a high level of alcohol dependence extensive treatment such as CBT has been shown to be highly cost-effective (Holder et al. 2000; Berglund et al. 2003). There has also been some more recent research that demonstrates the effectiveness of combining CBT with MI (see Timken et al. 2012). However, it should be noted that, although CBT is one of the most studied substance abuse treatment interventions, research investigating the effectiveness of CBT in reducing impaired driving recidivism is limited and only a small number of studies have specifically and rigorously tested the effectiveness of CBT, or variations of it, in reducing either alcohol misuse or impaired driving behaviour among this offender population (Brown and Ouimet 2013). Staff training requirements. With appropriate training and supervision, a diverse range of therapists can implement CBT effectively. However, most training manuals focus on specific cognitive-behavioural techniques and do not cover basic clinical skills. Certain minimal requirements are recommended: > A master s degree or equivalent in psychology, counseling, social work, or a closely related field. > At least 3 years of experience working with a substance-abusing population. > Some familiarity with and commitment to a cognitive-behavioural approach. 94 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

196 Mechanism of delivery. CBT approaches are used with individual patients or with groups. Some of these approaches rely on more traditional client-therapist interactions; others rely on computer-based software. Strengths and weaknesses. Strengths > Behaviour change is often a central part of the process. > It is structured, which includes setting agendas and working toward clear goals. > It is usually relatively short-term. Weaknesses > CBT does not suit everyone. > It assumes the client has access to thoughts and emotions. > Being committed and persistent in improving substance abuse problems can be hard work. > Clients are challenged and often taken out of their comfort zones when tackling substance abuse problems. > It has not been directly evaluated with an impaired driving population. References Berglund, M., Thelander, S., Salaspuro, M., Franck, J., Andreasson, S., & Ojehagen, A. (2003). Treatment of alcohol abuse: An evidence-based review. Alcoholism: Clinical and Experimental Research, 27(10), Carroll, K. (1996). Relapse prevention as a psychosocial approach: A review of controlled clinical trials. Experimental Clinical Psychopharmacology, 4, Longabauch, R., & Morgenstern, J. (1999). Cognitive-behavioural coping-skills therapy for alcohol dependence. Alcohol Research and Health, 23, Guides Morgenstern, J., Morgan, T.J.,McCrady, B.S., Keller, D.S., & Carroll, K.M. (2001). Manualguided cognitive-behavioural therapy training: A promising method for disseminating empirically supported substance abuse treatments to the practice community. Psychology of Addictive Behaviours, 15(2), Wanberg, K., Milkman, H., & Timken, D. (2005). Driving With Care: Education and Treatment of the Impaired Driving Offender. Strategies for Responsible Living and Change. New York: Sage Publishing. TREATMENT INTERVENTIONS 95

197 Timken, D.S., Nandi, A., & Marques, P. (2012). Interlock Enhancement Counseling: Enhancing Motivation for Responsible Driving: A Provider s Guide. Centre for Impaired Driving Research and Evaluation. 7.4 Pharmacotherapies Purpose and objectives. It is generally agreed that greater use of pharmacological interventions could enhance treatment progress since it stabilizes the patient and creates a facilitating environment. According to National Institute on Drug Abuse s (NIDA) Principles of Drug Abuse Treatment for Criminal Justice Populations, medications are an important part of treatment for many drug abuse offenders (NIDA 2006, p. 5). Indeed, it has been argued that there is a need for greater receptiveness of the fact that medications may be an integral part of treatment (Robertson 2007), and despite immense progress in pharmacotherapy research, medications that have been approved to treat alcohol dependence are still underutilized (Arias et al. 2008). Programs and services that include a medicinal component may be referred to as pharmacotherapy, medication, drug therapy, and so forth. There are many medications that can be used for alcohol treatment purposes. However, pharmacotherapies are not frequently used to treat impaired driving offenders and their availability/use among this population is not known. Three products that are currently approved for treating alcohol dependence are naltrexone, acamprosate, and disulfiram (NIAAA 2008). They have been shown to help patients reduce drinking, avoid relapse to heavy drinking, achieve and maintain abstinence, or gain a combination of these effects. > Naltrexone (ReVia ; Vivitrol ) is an opioid antagonist that has a short half-life so it has limited clinical utility. Side effects include nausea, dizziness and fatigue. Usual adult dosage is 50 mg daily. > Acamprosate (Campral ) is a synthetic compound that is a putative glutamate modulator. Usual adult dosage is 2-3 grams. Common side effects include mild diarrhea. > Disulfiram (Antabuse ) interferes with the metabolism of alcohol by the liver, permitting a toxic breakdown product of alcohol to accumulate in the bloodstream. Usual adult dosage is 250 mg daily (ranging from 125 mg to 500 mg). Common side effects include metallic after-taste, dermatitis, and transient mild drowsiness. These medications are often used in combination with brief psychosocial interventions. These medications have been shown to help patients reduce drinking, avoid relapse to heavy drinking, achieve and maintain abstinence, or gain a combination of these effects (NIAAA 2005). For a comprehensive review of available pharmacotherapies for treating alcohol use, 96 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

198 please refer to Arias et al. (2008) in Alcohol Research and Health, the Journal of the National Institute on Alcohol Abuse and Alcoholism (NIAAA). General effectiveness. A 2004 meta-analysis of 17 randomized controlled trials (Mann et al.) found that acamprosate was associated with a significantly higher number of abstinent days and continuous abstinence rates at six months were significantly higher. Acamprosate appears to be especially useful in a therapeutic approach targeted towards achieving abstinence in recently detoxified, motivated alcohol-dependent patients (Bouza et al. 2004). In a 2005 meta-analysis of 24 randomized controlled trials, Srisurapanont and Jarusuraisin reported that naltrexone significantly decreased relapses but not a complete return to drinking (i.e., relapse decreased, but eventually some subjects did start drinking again). Naltrexone seems more indicated in or appropriate for programs geared towards controlled consumption. Treatment compliance is a significant issue in these and other studies and needs to be addressed adequately to assure their usefulness in clinical practice. Staff training requirements. Prescriptions are needed in order for a client to receive any pharmacological intervention; as a result, a clinician can only make a recommendation to a client to seek pharmacological treatment and management of alcohol misuse. Whether a medication should be prescribed and in what amount is a matter between clients and their health care providers. Mechanism of delivery. Prior to suggesting any pharmacological intervention it is recommended that the physician conduct a screening using a clinical interview and a screening instrument to determine the client s level of alcohol dependence. Most studies recommended that pharmacological interventions for alcohol dependence include some type of counseling, and it is recommended that all clients taking these medications receive at least brief medical counseling. Offering the full range of effective treatments will maximize patient choice and outcomes, as no single approach is universally successful or appealing to patients. The different approaches - medications for alcohol dependence, professional counseling, and mutual help groups - are complementary. Strengths and weaknesses. Strengths > Naltrexone is especially helpful for curbing consumption in patients who have drinking slips. > Acamprosate is thought to reduce symptoms of protracted abstinence such as insomnia, anxiety, restlessness, and dysphoria. Weaknesses > Generally speaking, compliance with the use of pharmacotherapies may be low. Long-acting injectable drugs such as Vivitrol may have greater compliance. TREATMENT INTERVENTIONS 97

199 > Naltrexone is less effective in maintaining abstinence. > Disulfiram can produce a very unpleasant reaction including flushing, nausea, and palpitations if the patient drinks alcohol. > The utility and effectiveness of disulfiram are considered limited because compliance is generally poor when patients are given it to take at their own discretion. References National Institute on Alcohol Abuse and Alcoholism (NIAAA). (2005). Helping Patients Who Drink Too Much: A Clinician s Guide Updated Available online: gov/publications/practitioner/cliniciansguide2005/guide.pdf National Institute on Alcohol Abuse and Alcoholism (NIAAA). (2008). Excerpt from Helping Patients Who Drink Too Much: A Clinician s Guide. Publication Washington, D.C.: U.S. Department of Health and Human Services, National Institutes of Health. niaaa.nih.gov/publications/practitioner/cliniciansguide2005/prescribingmeds.pdf Guides National Institute on Drug Abuse (NIDA). (2006). Principles of Drug Abuse Treatment for Justice Populations: A Research-based Guide. NIH Publication No Washington, D.C.: U.S. Department of Health and Human Services, National Institutes of Health. Available online: Internet-based Brief Interventions Purpose and objectives. Computerized and web-based interventions for persons with substance misuse problems have increased in popularity with the growing availability and use of computers. The use of computerized interventions presents an opportunity for broad dissemination and improved access to services (Copeland and Martin 2004; Cunningham et al. 2005). These interventions are highly automated, non-resource intensive, and have the potential to reach large audiences in a convenient and timely fashion (Kypri et al. 2003; Riper et al. 2009; Bingham et al. 2010; Webb et al. 2010). In addition, with the exception of costs for program development, the typical cost associated with staffing and the requisite training to deliver the intervention is nominal and/or non-existent. Internet therapies are accessible to a large segment of the population and can be a convenient option for those who may have difficulty accessing programming due to geographic location, time, or childcare constraints (Gainsbury and Blaszcynski 2010; Khadjesari et al. 2010). Another benefit associated with computerized or web-based applications relates to the level of openness and disclosure among subjects regarding their alcohol use patterns. Online programs can also help ensure the consistent delivery of interventions as the material and content is identical for all of those who access the program (Noell and Glasgow 1999; 98 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

200 Carroll and Rounsaville 2007; Newton et al. 2010). Web-based interventions have the ability to be efficient when changes to content are needed they can be implemented uniformly and immediately (Vernon 2010). General effectiveness. There has been a growing body of research that examines the effectiveness of computerized or internet-based brief interventions designed to reduce alcohol use among a variety of populations. It is important to recognize that the effectiveness of these interventions with criminal justice populations generally, or impaired driving offenders specifically, has not been investigated and research on this application could not be located. Web-based delivery may enhance the implementation of brief alcohol interventions. Studies have noted that there is great potential for web-based interventions to encourage changes in behaviours such as alcohol use (Pemberton et al. 2010). For example, Bendtsen et al. (2011) conducted a review of 85 studies (with a total sample of 43,236 subjects) and found that a variety of electronic screening and brief alcohol interventions showed small but significant effects on risky drinking behaviours for various age groups. Also, Rooke et al. (2010) conducted a meta-analysis and found that web-based interventions can be a cost-effective means of addressing uncomplicated substance use and related problems among adolescents, young adults, and adults (30+). Vernon (2010) notes that computer-based interventions for alcohol use designed for the general public are relatively new, rare, and scarcely studied. This would include programs targeted at an impaired driver population. Staff training requirements. Staff training will depend on the computerized and webbased interventions utilized. Mechanism of delivery. The computerized and web-based interventions can be delivered at the same time other interventions would take place. The only difference would be that the delivery will be by computer and not by a clinician. Strengths and weaknesses. Strengths: > Highly automated. > Convenient and can reach large audiences. > Opportunity for broad dissemination and improved access to services. > Consistently delivered and uniformly implemented. Weaknesses: > Could limit interaction with clinicians. > Initial costs of purchasing multiple computers and training staff can be expensive. TREATMENT INTERVENTIONS 99

201 References Gainsbury, S., & Blaszczynski, A. (2010). A systematic review of internet-based therapy for the treatment of addictions. Clinical Psychology Review, 31(3), Khadjesari, Z., Murray, E., Hewitt, C., Hartley, S., & Godfrey, C. (2010). Can stand-alone computer-based interventions reduce alcohol consumption? A systematic review. Addiction, 106, Riper, H., Kramer, J., Conijn, B., Smit, F., Schippers, G., & Cuijpers, P. (2009). Translating effective web-based self-help for problem drinking into the real world. Alcoholism: Clinical and Experimental Research, 33(8), Webb, T., Joseph, J., Yardley, L., & Michie, S. (2010). Using the internet to promote health behaviour change: A systematic review and meta-analysis of the impact of theoretical basis, use of behaviour change techniques, and mode of delivery on efficacy. Journal of Medical Internet Research, 12(1), e Summary There is a range of treatment interventions that have been shown to be promising or effective in reducing recidivism among impaired driving offenders. However, each of these strategies rely upon different levels of resources, staff with different backgrounds and qualifications, different amounts of time, and have varying levels of cost. In addition, some interventions are more easily implemented and delivered than others. Perhaps what is most important is that efforts are made to best match interventions to the individual risks and needs of each offender. 100 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

202 8. BEST PRACTICES FOR TREATMENT AND REHABILITATION OF IMPAIRED DRIVING OFFENDERS Health Canada produced a Best Practices report (2004) that was based upon a thorough literature review, consultation with experts, and interviews with key informants. The aim of the report was to compile current knowledge on driving while impaired remedial programs across Canada. Specifically, the report addresses the planning and delivery of education programs and treatment and rehabilitation programs. The report in its entirety can be found at the link: hc-sc.gc.ca/hc-ps/alt_formats/hecs-sesc/pdf/pubs/ adp-apd/bp_treatment-mp_traitement/treatment_ rehab_driving_impaired_practices.pdf For the convenience of practitioners in the field, the best practices identified and described in the Health Canada report are re-produced here. > Remedial impaired driver programs delivered on behalf of driver licensing authorities should be part of a comprehensive impaired driving countermeasures program and participation should be a mandatory condition of licence reinstatement for all convicted impaired drivers. This should be the same for driving while impaired by drugs other than alcohol. > Different types of remedial interventions for different types of impaired driving offenders and should incorporate at least two levels for people with differing levels of substance use and related problems. All programs for impaired drivers should incorporate both educational and therapeutic activities, regardless of program length. Mandatory clinical follow up post-reinstatement should be required for all participants in remedial programs. > All convicted impaired drivers should complete a screening/assessment process to inform decisions about interventions. Proven instruments should be included in screening procedure and their performance should be monitored on an ongoing basis. > Remedial programs should supplement and not replace licensing actions. BEST PRACTICES FOR TREATMENT AND REHABILITATION OF IMPAIRED DRIVING OFFENDERS 101

203 > Individuals who receive roadside suspensions should be considered for referral to assessment and participation in remedial programs. > Remedial programs should be located in an environment in which a behavioural health perspective and treatment orientation are well established and can be maintained. > Those providing remedial services to impaired drivers should be trained in substance use issues and in adult education (particularly those delivering educational interventions) and group facilitation (particularly those delivering therapeutic interventions). > Those providing remedial measures programs to convicted impaired drivers should be supported in accessing provincial or national training opportunities on an annual or bi-annual basis. > Remedial programs should be operated using an administrative model, where program completion is a requirement of relicensing. > Remedial programs should be operated by an agency other than the licensing authority. > There is a need for formal and clear mechanisms of coordination and collaboration between licensing authorities and remedial programs to ensure reciprocal exchange of information to serve the best interests of the clients and the public. > Measures should be taken to reduce the financial burden for offenders, particularly those assigned to more expensive program options. This could include applying a single blended fee for all clients or providing some form of financial assistance for low-income clients. > Program evaluation should be part of any remedial measures program. > Program evaluation and research costs should be built into program budgets. > More emphasis should be placed on quality assurance (to ensure the program is delivered as intended with regard to all aspects of delivery), and studies of costeffectiveness of programs and their component parts. References Health Canada. (2004). Best Practices: Treatment and Rehabilitation for Driving While Impaired Offenders. Ottawa: Author. 102 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

204 9. RESEARCH GAPS AND FUTURE NEEDS Much has been learned about the profile and characteristics of impaired drivers over the course of the past three decades. To a lesser extent, knowledge has also grown with regard to the factors that put them at risk, the types of assessment instruments that are appropriate for this population, and the types of treatment interventions that can begin to address their risks and needs. Still, continued efforts are needed to increase understanding of these topics and to inform approaches that can best prevent impaired driving behaviour, as well as manage, supervise and treat those that are detected and processed through the criminal justice system. A number of topics that reflect gaps in offender research, gaps in intervention research, and gaps in implementation and practice warrant future attention. These are briefly highlighted below. 9.1 Gaps in Offender Research Perhaps most pressing in the field of research is the need to integrate existing knowledge stemming from diverse disciplines as a basis to explore and develop more holistic, robust and complex models of impaired driving behaviour that acknowledge the heterogeneity of this population. In particular, this model must recognize the different developmental pathways of offenders who do not re-offend as well as those who persist in their behaviour. A core feature of this initiative should be to increase understanding of the interactions and effects of different characteristics of offenders. Such efforts can be useful to help identify clinically relevant subgroups and guide the development of appropriate interventions that specifically target them. Greater knowledge and understanding of relevant risk factors that influence future offending is also a critical need. At the same time, the development of valid, reliable and practical screening and risk assessment instruments that can accurately distinguish between offenders not only with regard to risk related to substance use but also risk of re-offending and RESEARCH GAPS AND FUTURE NEEDS 103

205 individual-specific trajectories to impaired driving behaviour are essential to inform decisionmaking and the allocation of resources. This is a pressing concern in light of shrinking budgets and resources. 9.2 Gaps in Intervention Research While knowledge of effective interventions has grown substantially since the 1990s, there has been a rather exclusive emphasis on research that has investigated individual interventions that are more punitive than rehabilitative in nature. Also of importance, effectiveness has largely been limited to measurement of alcohol use reduction and to lesser degree recidivism. However, the reality is that most interventions are delivered in complex systems of justice, licensing and health, and a majority of offenders are subject to a multitude of interventions. Moreover, there is a much broader range of outcome measures, beyond recidivism (e.g., employment, family stability, engagement in pro-social activities, health benefits), that are worthy of attention. Hence three important trends have emerged that will significantly influence the direction of intervention research moving forward. First, since 2005 there has been growing recognition among researchers, policymakers and practitioners of the value of treatment and rehabilitation as essential goals of the justice system for long-term risk reduction. Second, sanctions that are increasingly applied to impaired driving offenders are imposed with the intention of achieving a better balance between supervision and treatment. This means that offenders are more often subject to a combination of interventions that are delivered in different systems with different goals and objectives. And, third, a variety of factors or outcomes in addition to substance use, are relevant to reductions in recidivism and should be considered part of research designs. Hence future efforts to investigate the effectiveness of interventions must account for not only the increasingly complex environment in which such interventions are delivered, but also the web of factors that play an important role. And while much has been learned about effective interventions, a range of research questions remain that must be addressed. These include: > Is it possible to achieve an optimal balance between sanction/supervision and rehabilitation/treatment for offenders with different levels of risk? > What interventions or combination of interventions provide the best outcomes for different subpopulations of offenders. > Are there commonalities and differences across interventions that can provide insight into the essential ingredients of effective interventions? This may include an examination of content, delivery mechanisms, training, duration, key features, and the emphasis that is placed on sanctioning, rehabilitation or both. 104 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

206 > Is there an optimal duration for the various interventions that are available, including educational programs, treatment, probation, and alcohol monitoring technologies? > Is it possible to achieve the outcomes associated with longer-term and more intensive treatment interventions using well-designed programs that are more costeffective and shorter in duration? > What characteristics of offenders are most useful to appropriately match them to effective interventions? It must be underscored that answers to these questions may only be possible once our understanding of offenders has grown. 9.3 Gaps in the Implementation of Interventions In an environment that is heavily influenced and compromised by a growing number of practical and economic constraints, policymakers, agency administrators and practitioners will be forced to consider a range of implementation issues in the coming years that can have significant implications for the delivery of interventions. Some of these issues are briefly discussed below. First, recent increases in impaired driving behaviour among women (Perreault 2013), and research indicating that female offenders may possess clinically significantly differences relative to males (Robertson et al. 2011b) provide important food for thought. The same is true in relation to anecdotal evidence from frontline practitioners that perhaps more young drivers are participating in remedial impaired driver programs. These situations warrant close monitoring and may have important implications for the delivery of interventions in order to account for differences across sexes and ages. Second, there is growing awareness that additional and complementary services may be required for specific sub-populations of offenders such as those who possess deficits in executive cognitive functioning, those who suffer from co-occurring disorders, and those offenders identified with polysubstance (i.e., alcohol and drugs) use. Additionally, service delivery in rural jurisdictions continues to be a source of concern as does the delivery of culturally appropriate services for the ethnically diverse population in Canada. In this regard, strategic partnerships will play a pivotal role in filling these gaps and efforts are needed to encourage and facilitate these collaborations. Third, while much has been learned with regard to effective interventions, less work has been focused on the implementation of such programs to ensure that they are delivered in ways that demonstrate fidelity to the model. In some respects, this issue is intimately linked to efforts to promote high standards of effective and efficient programming across relevant systems. The achievement of this goal will require the prioritization of consistent training and RESEARCH GAPS AND FUTURE NEEDS 105

207 education for practitioners, the use of quality control procedures and, most importantly, an emphasis on both process and outcome evaluations of these interventions in the future. References Health Canada. (2004). Best Practices: Treatment and Rehabilitation for Driving While Impaired Offenders. Ottawa: Author. Timken, D.S. (2002). What Works: Effective DWI Interventions. In Harry E. Allen (ed.) What Works: Risk Reduction: Interventions for Special Needs Offenders. MD: American Correctional Association. 106 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

208 REFERENCES Aharonovich, E., Nunes, E., & Hasin, D. (2003). Cognitive impairment, retention and abstinence among cocaine abusers in cognitive-behavioral treatment. Drug and Alcohol Dependence, 71(2), Alexander, R. (2000). Counseling, Treatment, and Intervention: Methods with Juvenile and Adult Offenders. Toronto: Wadsworth. Ambtman, R. (1990). Impaired Drivers Program Evaluation Report. Alcoholism Foundation of Manitoba, The Awareness and Information Directorate. American Public Health Association and Education Development Center, Inc. (2008). Alcohol Screening and Brief Intervention: A Guide for Public Health Practitioners. Washington, D.C.: National Highway Traffic Safety Administration, U.S. Department of Transportation. Available online: Anderson, B.J., Snow, R.W., & Wells-Parker, E. (2000). Comparing the predictive validity of DUI risk screening instruments: Development of validation standards. Addiction, 95(6), Andrews, D., & Dowden, C. (2006). Risk principle of case classification in correctional treatment: A meta-analytic investigation. International Journal of Offender Therapy and Comparative Criminology, 50, Andrews, D., Bonta, J., & Wormith, J. (2006). The recent past and near future of risk and/or need assessment. Crime & Delinquency, 52(1), Argeriou, M., McCarty, D., Potter, D., & Holt, L. (1986). Characteristics of men and women arrested for driving under the influence of liquor. Alcoholism Treatment Quarterly, 3, Applegate, B.K., Langworthy, R.H., & Latessa, E.J. (1997). Factors associated with success in treating chronic drunk drivers. Journal of Offender Rehabilitation, 24(3), Babor, T.F., McRee, B.G., Kassebaum, P.A., Grimaldi, P.L., Ahmed, K., & Bray, J. (2007). Screening, brief intervention, and referral to treatment (SBIRT): Toward a public health approach to the management of substance abuse. Substance Abuse, 28(3), Baker, S. P., Braver, E.R., Chen, L-H., Li, G., & Williams, A.F. (2002). Drinking histories of fatally injured drivers. Injury Prevention, 8(3), Beadnella, B., Nasona, M., Stafforda, P.A., Rosengrena, D.B., & Daughertya, R. (2012). Shortterm outcomes of a motivation-enhancing approach to DUI intervention. Accident Analysis and Prevention, 45, REFERENCES 107

209 Behavioral Health Research Centre of the Southwest (BHRCS). (2007). Follow-up of Female DWI Offenders Five Years After Screening. Research Projects funded by the National Institute on Alcohol Abuse and Alcoholism. Retrieved from: htm Begg, D.J., Langley, J.D., & Stephenson, S. (2003). Identifying factors that predict persistent driving after drinking, unsafe driving after drinking, and driving after using cannabis among young adults. Accident Analysis and Prevention, 35(5), Beirness, D., & Davis, C. (2008). Driving after Drinking: Analysis drawn from the 2004 Canadian Addiction Survey. Ottawa: Canadian Centre on Substance Abuse. Available online: Beirness, D.J., & Marques, P.M. (2004). Alcohol ignition interlock programs. Traffic Injury Prevention, 5(3), Beirness, D.J., Mayhew, D.R., & Simpson, H.M. (1997). DWI Repeat Offenders: A Review and Synthesis of the Literature. Ottawa: Traffic Injury Research Foundation. Beirness, D., & Simpson, H. (1988). Lifestyle correlates of risky driving and accident involvement among youth. Alcohol, Drugs and Driving, 4, Bendtsen, P., Ekman, D., Johansson, A., Carlfjord, S., Andersson, A., Leijon, M., Johansson, K., & Nilsen, P. (2011). Referral to an electronic screening and brief alcohol intervention in primary health care in Sweden: Impact of staff referral to the computer. International Journal of Telemedicine and Applications, 2011, Bergdahl, J. (1999). An application of convergence theory to women s drinking and driving. Women and Criminal Justice, 10(4), Bingham, C.R., Shope, J.T., Parow, J.E., & Raghunathan, T.E. (2009). Crash types: Markers of increased risk of alcohol-involved crashes among teen drivers. Journal of Studies on Alcohol and Drugs, 70, Bingham, C., Barretto, A., Walton, M., Bryant, C., Shope, J., & Raghunathan, T. (2010). Efficacy of a web-based, tailored, alcohol prevention/intervention program for college students: Initial findings. Journal of American College Health, 58(4), Bloom, B., Owen, B., & Covington, S. (2003). Gender-Responsive Strategies: Research, Practice and Guiding Principles for Women Offenders. Washington, D.C.: National Institute of Corrections. Bonta, J. (2002). Offender risk assessment: Guidelines for selection and use. Criminal Justice and Behavior, 29, Bonta, J., Wallace-Capretta, S., & Rooney, J. (2000). A quasi-experimental evaluation of an intensive rehabilitation supervision program. Criminal Justice and Behavior, 27(3), Boudreault, J., Brassard, A., & Gagnon, J.P. (2002). Behavioural assessment of Quebec drivers convicted of a second or a subsequent impaired driving offence under the Criminal Code. In D.R. Mayhew and C. Dussault (Eds.), Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety, (pp ). Quebec City: Societé de l assurance automobile du Québec. 108 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

210 Brinkmann, B., Beike, J., Kohler, H., Heinecke, A., & Bajanowski, T. (2002). Incidence of alcohol dependence among drunken drivers. Drug and Alcohol Dependence, 66(1), Brown, S.W., Vanlaar, W., & Mayhew, D.R, (2013). The Alcohol-Crash Problem in Canada: Ottawa: Transport Canada, Road Safety and Motor Vehicle Regulation. Brown, T.G., Dongier, M., Ouimet, M.C., Tremblay, J., Chanut, F., Legault, L., & Ng Yin Kin, N.M.K. (2010b). Brief motivational interviewing for DWI recidivists who abuse alcohol and are not participating in DWI intervention: A randomized controlled trial. Alcoholism: Experimental & Clinical Research, 34(2), Brown, T.G., Dongier, M., Ouimet, M. C., Tremblay, J., Chanut, F., Legault, L., & Kin, N. M. (2012). The role of demographic characteristics and readiness to change in 12-month outcome from two distinct brief interventions for impaired drivers. Journal of Substance Abuse Treatment, 42(4), Brown, T.G., Kokin, M., & Seraganian, P. (1995). The role of spouses of substance abusers in treatment: Gender differences. Journal of Psychoactive Drugs, 27(3), Brown, T., Nadeau, L., Legeix, P., Lepage, M., Tremblay, J., & Seraganian, P. (2002). Nonadherents in mandatory substance abuse evaluation following a DUI offense: Their characteristics and reasons for non-compliance. In D.R. Mayhew and C. Dussault (Eds.), Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety, (pp ). Quebec City: Societé de l assurance automobile du Québec. Brown, T.G., & Ouimet, M. (2013). Treatments for alcohol-related impaired driving. In M. McMurran (Ed.), Alcohol-Related Violence: Prevention and Treatment (pp ). Mississauga: John Wiley & Sons, Ltd. Brown, T., Ouimet, M., Nadeau, L., Lepage, M., & Pruessener, J. (2013, January). Neurocognitive bases of sex differences in DWI behavior. Presented at Transportation Research Board 92nd Annual Meeting, Washington, D.C. Brown, T.G., Ouimet, M.C., Nadeau, L., Lepage, M., Tremblay, J., Dongier, M., & Ng Ying Kin, N.M.K. (2008). DUI offenders who delay relicensing: A qualitative and quantitative investigation. Traffic Injury Prevention, 9, Brown, T., Ouimet, M., Nadeau, L., Gianoulakis, C., Lepage, M., Tremblay, J., & Dongier, M. (2009). From the brain to bad behaviour and back again: Neurocognitive and psychobiological mechanisms of driving while impaired by alcohol. Drug and Alcohol Review, 28(4), Brown, T.G., Ouimet, M.C., Nadeau, L., Lepage, M., Pruessner, J. (2010a, August). Sex and gender effects in DWI first time offenders: Neurocognitive differences. Presented at the 19th International Conference on Alcohol, Drugs and Traffic Safety, Oslo, Norway. Buntain-Ricklefs, J., Rivara, F., Donovan, D., Salzberg, P., & Polissar, N. (1995). Differentiating bad drivers with and without a DWI. Journal of Studies on Alcohol, 56(3), Caetano, R., & McGrath, C. (2005). Driving under the influence (DUI) among U.S. ethnic groups. Accident Analysis and Prevention, 37, REFERENCES 109

211 Caldwell-Aden, L., Kaczowka, M., & Balis, N. (2009). Preventing First-Time DWI Offenses. First Time DWI Offenders in California, New York and Florida: An Analysis of Past Criminality and Associated Criminal Justice Interventions. DOT HS Washington, D.C.: U.S. Department of Transportation. Campbell, D.T., & Fiske, D.W. (1959). Convergent and discriminant validation by the bultitraitmultimethod matric. Psychological Bulletin, 56, Carroll, K., & Rounsaville, B. (2007). A vision of the next generation of behavioural therapies research in the addictions. Addiction, 102, Cavaiola, A., & Wuth, C. (2002). Assessment and Treatment of the DUI Offender. New York: Haworth. Cavaiola, A., Strohmetz, D., Wolf, J.M., & Lavender, N.J. (2003). Comparison of DWI offenders with non-dwi offenders on the MMPI-2 and the Michigan Alcoholism Screening Test. Addictive Behaviors, 28, Cavaiola, A., Strohmetz, D., & Abreo, S. (2007). Characteristics of DUI recidivists: A 12-year follow-up study of first time DUI offenders. Addictive Behaviors, 32, Center for Disease Control and Prevention (2013). Binge drinking: A serious, underrecognized problem among women and girls. CDC Vitalsigns, January Chalmers, D., Olenick, N.L., & Stein, W. (1993). Dispositional traits as risk in problem drinking. Journal of Substance Abuse, 5, Chang, I., Lapham, S., & Barton, K.J. (1996). Drinking environment and socio-demographic factors among DWI offenders. Journal of Studies on Alcohol, 57, Chang, I., Gregory, C., & Lapham, S.C. (2002). Review of Screening Instruments and Procedures for Evaluating DWI (Driving While Intoxicated) Offenders. Washington, D.C.: AAA Foundation for Traffic Safety. Chanut, F., Brown, T., & Dongier, M. (2005). Motivational interviewing and clinical psychiatry. Canadian Journal of Psychiatry, 50(9), Chou, S.P., Dawson, D.A., Stinson, F.S., Huang, B., Pickering, R.P., Zhou, Y. & Grant B.F. (2006). The prevalence of drinking and driving in the United States, : Results from the national epidemiological survey on alcohol and related conditions. Drug and Alcohol Dependence, 83, Copeland, J., & Martin, G. (2004). Web-based interventions for substance use disorders: A qualitative review. Journal of Substance Abuse Treatment, 26, Cosper, R. & Mozersky, K. (1968). Social correlates of drinking and driving. Quarterly Journal of Studies on Alcohol, Supplement No.4, Couture, S., Brown, T.G., Brochu, S., & Gianoulakis, C. (2010, August). A neurobiological pathway to a high recidivism risk in first-time DWI offenders. Presented at the 19th International Conference on Alcohol, Drugs and Traffic Safety, Oslo, Norway. 110 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

212 Couture, S., Brown, T.G., Tremblay, J., Ng Ying Kin, N.M.K., Ouimet, M.C., & Nadeau, L. (2010). Are biomarkers of chronic alcohol misuse useful in the assessment of DWI recidivism status? Accident Analysis and Prevention, 42, Crews, F., Buckley, T., Dodd, P., Ende, G., Holey, N., & Harper, C. (2005). Alcoholic neurobiology: Changes in dependence and recovery. Alcoholism: Clinical and Experimental Research, 29(8), Dauvergne, M. (2012). Adult criminal court statistics in Canada, 2010/2011. Juristat. Statistics Canada Catalogue No X. DeMichele, M., & Lowe, N. (2011). DWI recidivism: Risk implications for community supervision. Federal Probation, 75(3), DeMichele, M., & Payne, B. (2013). If I had a hammer, I would not use it to control drunk driving. Using predictive tools to respond to persistent drunk driving. Criminology and Public Policy, 12(2), doi: / DeMichele, M., Wanberg, K.W., Timken, D.S., & Lowe, N. (2013). Impaired Driving Assessment. Lexington, KY: American Probation and Parole Association. Deyle, R. (2008). Education/Treatment Intervention Among Drinking Drivers and Recidivism. Denver: Department of Human Services, Division of Behavioral Health. Donovan, D.M., Marlatt, G.A., & Salzberg, P.M. (1983). Drinking behavior, personality factors and high risk driving. Journal of Studies on Alcohol, 44(3), Drew, L., Royal, D., Moulton, B., Peterson, A., & Haddix, D. (2010). National Survey on Drinking and Driving Attitudes and Behaviors: Washington, D.C.: U.S. Department of Transportation. Dowden, C., & Andrews, D.A. (1999). What works for female offenders: A meta-analytic review. Crime and Delinquency, 45, Dugosh, K.L., Festinger, D.S., & Marlowe, D.B. (2013). Moving beyond BAC in DUI. Criminology and Public Policy, 12(2), doi: / Dunn, C., Deroo, L., & Rivara, F.P. (2001). The use of brief interventions adapted from motivational interviewing across behavioural domains: a systematic review. Addiction, 96(12), Elliott, M.R., Shope, J.T., Raghunathan, T.E., & Waller, P.F. (2006). Gender differences among young drivers in the association between high-risk driving and substance use/environmental influences. Journal of Studies on Alcohol, 67(2), European Transport Safety Council (ETSC). (2011) Road Safety Target Outcome: 100,000 Fewer Deaths since th Road Safety Pin Report. Brussels: Author. etsc.eu/documents/etsc2011_pin_report.pdf Farrow, J.A., & Brissing, P. (1990). Risk for DWI: A new look at gender differences and drinking and driving influences, experiences, and attitudes among new adolescent drivers. Health Education Behavior, 17(2), REFERENCES 111

213 Federal Bureau of Investigation (FBI). (2008). Crimes in the United States, 2007: Table Federal Bureau of Investigation (2010). Uniform Crime Reports. cjis/ucr/ucr. Feldstein, S., & Miller, W. (2007). Does subtle screening for substance abuse work? A review of the Substance Abuse Subtle Screening Inventory (SASSI). Addiction, 102(1), Ferguson, S.A., Burns, M.M., Fiorentino, D., Williams, A.F., & Garcia, J. (2002). Drinking and driving among Mexican American and non-hispanic white males in Long Beach, California. Accident Analysis and Prevention, 34, Fine, E.W., & Scoles, P. (1974). Alcohol, alcoholism and highway safety. Public Health Review, 30, Flowers, N.T., Naimi, R.S., Brewer, R.D., Elder, R.W., Shultz, R.A., & Jiles, R. (2008). Patterns of alcohol consumption and alcohol-impaired driving in the United States. Alcoholism: Clinical and Experimental Research, 32(4), Freeman, J., Maxwell, J.C., & Davey, J. (2011). Unraveling the complexity of driving while intoxicated: A study into the prevalence of psychiatric and substance abuse comorbidity. Accident Analysis and Prevention, 43(1), Gainsbury, S., & Blaszczynski, A. (2010). A systematic review of internet-based therapy for the treatment of addictions. Clinical Psychology Review, 31(3), Gebers, M., & Peck, R. (2003). Using traffic conviction correlates to identify high accident-risk drivers. Accident Analysis and Prevention, 35(6), Gendreau, P., Little, T., & Goggin, C. (1996). A Meta-Analysis of the Predictors of Adult Recidivism: What Works! Ottawa: Public Works and Government Services Canada. George, D., & Mallery, P. (2003). SPSS for Windows step by step: A simple guide and reference update (4th ed.). Boston: Allyn & Bacon. Glass, R.J., Chan, G., & Rentz, D. (2000). Cognitive impairment screening in second offense DUI programs. Journal of Substance Abuse Treatment, 19, Gould, L.A. (1989). Criminality and driving while intoxicated: A comparison of DWIs and a random sample of licensed drivers. Journal of Contemporary Criminal Justice, 5, Grant, B., & Dawson, D. (1997). Age of onset of alcohol use and its association with DSM-IV alcohol abuse and dependence: Results from the national Longitudinal Alcohol Epidemiologic Survey. Journal of Substance Abuse, 9, Greenfield, S.F. (2002). Perspectives: Women and alcohol use disorders. Harvard Review Psychiatry, 10(2), Gudrais, E. (2011). Women and Alcohol. Harvard Magazine, July-August, IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

214 Gulliver, P., & Begg, D. (2004). Influences during adolescence on perceptions and behaviour related to alcohol use and unsafe driving as young adults. Accident Analysis and Prevention, 36(5), Harlow, C. (2003). Education and correctional populations. Bureau of Justice Statistics Special Report. NCJ Available online: Harrison, P.A.,& Asche, S.E. (2001). Outcomes monitoring in Minnesota: Treatment implications,practical limitations. Journal of Substance Abuse Treatment, 21, Hasin, D., Stinson, F., Ogburn, E., & Grant, B. (2007). Prevalence, correlates, disability, and comorbidity of DSM-IV alcohol abuse and dependence in the U.S.: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. Archives of General Psychiatry, 64(7), Health Canada. (2004). Best Practices: Treatment and Rehabilitation for Driving While Impaired Offenders. Ottawa: Author. Herbert, T., Davis, H.T., Beaton, S.J., Von Worley, A., Parsons, W., & Gunter, M.J. (2012). The effectiveness of screening and brief intervention on reducing driving while intoxicated citations Population Health Management, 15(1), Higgins-Biddle, J., Hungerford, D., & Cates-Wessel, K. (2009). Screening and Brief Interventions (SBI) for Unhealthy Alcohol Use: A Step-By-Step Implementation Guide for Trauma Centers. Atlanta: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Available online: Hingson, R., & Winter, M. (2003). Epidemiology and consequences of drinking and driving. Alcohol Research and Health, 27, Hingson, R., Heeren, T., Levenson, S., Jamanka, A., & Voas, R. (2002). Age of onset driving after drinking, and involvement in alcohol-related motor-vehicle crashes. Accident Analysis and Prevention, 34, Hingson, R., Heeren, T., Zakocs, R., Winter, M., & Wechsler, H. (2003). Age of first intoxication, heavy drinking, driving after drinking and risk of unintentional injury among U.S. college students. Journal of Studies on Alcohol, 64, Horn, J.L., Wanberg, K.W., & Foster, F.M. (1990). Guide to the Alcohol Use Inventory (AUI). Minneapolis, MN: National Computer Systems. Jonah, B.A., & Wilson, R.J. (1986). Impaired drivers who have never been caught: Are they different from convicted impaired drivers? Society of Automotive Engineers Technical Paper Series, Jones, R., & Lacey, J. (2001). Alcohol and Highway Safety 2001: A Review of the State of Knowledge. DOT HS Washington, D.C.: U.S. Department of Transportation. Kelley-Baker, T., & Romano, E. (2010). Female involvement in U.S. nonfatal crashes under a three-level hierarchical model. Accident Analysis and Prevention, 42(6), REFERENCES 113

215 Khadjesari, Z., Murray, E., Hewitt, C., Hartley, S., & Godfrey, C. (2010). Can stand-alone computer-based interventions reduce alcohol consumption? A systematic review. Addiction, 106, Kong, R., & AuCoin, K. (2008). Female offenders in Canada. Juristat, 28(1). Ottawa: Statistics Canada. Kramer, A. (1986). Sentencing the Drunk Driver: A Call for Change. In S. Valle (Ed.), Drunk Driving in America (pp ). New York: Haworth Press. Kushner, M., Maurer, E., Menary, K., & Thuras, P. (2011). Vulnerability to the rapid ( telescoped ) development of alcohol dependence in individuals with anxiety disorder. Journal of Studies on Alcohol and Drugs, 72, Kypri, K., Saunders, J., & Gallagher, S. (2003). Acceptability of various brief interventions approaches for hazardous drinking among university students. Alcohol and Alcoholism, 38, Labrie, R.A., Kidman, R.C., Albanese, M., Peller, A.J., & Shaffer, H.J. (2007). Criminality and continued DUI offense: Criminal typologies and recidivism among repeat offenders. Behavioral Sciences and the Law, 25, Lacey, J.H., Jones, R.K., & Wiliszowski, C.H. (1999). Validation of Problem Drinking Screening Instruments for DWI Offenders. Technical Report. DOT HS Washington, D.C.: U.S. Department of Transportation. Lacey, J.H, Kelley-Baker, T., Furr-Holden, D. Voas, R., Romano, E., Ramirez, A., Brainard, K., Moore, C., Torres, P., & Berning, R. (2009). National Roadside Survey of Alcohol and Drug Use by Drivers: Drug Results. DOT HS Washington, D.C.: Department of Transportation. Lapham, S. (2004/2005). Screening and brief intervention in the criminal justice system. Alcohol Research and Health, 28(2), Retrieved from: publications/arh28-2/85-93.pdf Lapham, S.C., Skipper, B.J., & Simpson, G.L. (1997). A prospective study on the utility of standardized instruments in predicting recidivism among first DWI offenders. Journal of Studies on Alcohol, 58, Lapham, S.C., Skipper, B.J., Hunt, W.C., & Chang, I. (2000). Do risk factors for re-arrest differ for female and male drunk-driving offenders? Alcoholism: Clinical and Experimental Research, 24(11), Lapham, S.C., Smith. E., C de Baca, J., Chang, I., Skipper. B.J., Baum, G., & Hunt, W.C. (2001). Prevalence of psychiatric disorders among persons convicted of driving while impaired. Archives of General Psychiatry, 58, Laplante, D.A., Nelson, S.E., Odegaard, S.S., LaBrie, R.A., & Shaffer, H.J. (2008). Substance and psychiatric disorders among men and women repeat driving under the influence offenders who accepts a treatment-sentencing option. Journal of Studies on Alcohol and Drugs, 69(2), IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

216 Latessa, E., Smith, P., Lemke, R., Makarios, M., & Lowenkamp, C. (2009). Creation and Validation of the Ohio Risk Assessment System: Final Report. Cincinnati: University of Cincinnati Center for Criminal Justice Research. Lex, B.W., Sholar, J.W., Bower, T., & Mendelson, J.H. (1991). Putative Type II alcoholism characteristics in female third DUI offenders in Massachusetts: A pilot study. Alcohol, 8(4), Lex, B. W., Goldberg, M. E., Mendelson, J. H., & Lawler, N. S. (1994). Components of antisocial personality disorder among women convicted for drunken driving. Annals of the New York Academy of Sciences, 708, Lowe, N. (2013). Screening for Risk and Needs Using the Impaired Driving Assessment: Final Report. Lexington, KY: American Probation and Parole Association. (in press). Lund, A., & Wolf, A. (1991). Changes in the incidence of alcohol-impaired driving in the United States, Journal of Studies on Alcohol, 52(4), Lynskey, M., Bucholz, K., Madden, P., & Heath, A. (2007). Early-onset alcohol-use behaviors and subsequent alcohol-related driving risks in young women: A twin study. Journal of Studies on Alcohol and Drugs, 68(6), MacDonald, S. (1989). A comparison of the psychosocial characteristics of alcoholics responsible for impaired and non-impaired collisions. Accident Analysis and Prevention, 21(5), MacDonald, S., & Pederson, L.L. (1990). The characteristics of alcoholics in treatment arrested for Driving While Impaired. British Journal of Addiction, 85(1), Maldonado-Bouchard, S., Brown, T., & Nadeau, L. (2012). Decision-making capacities and affective reward anticipation in DWI recidivists compared to non-offenders: A preliminary study. Accident Analysis and Prevention, 45, Mann, R.E., Smart, R.G, Stoduto, G., Adlaf, E.M., & Ialomiteanu, A. (2004). Alcohol consumption and problems among road rage victims and perpetrators. Journal of Studies on Alcohol, 65(2), Marcoux, K., Robertson, R.D., & Vanlaar, W. (2011). Road Safety Monitor 2010: Youth Drinking and Driving. Ottawa: Traffic Injury Research Foundation. Marques, P., Tippetts, S., Voas, R., & Beirness, D. (2001). Predicting repeat DWI offenses with the alcohol interlock recorder. Accident Analysis and Prevention, 33, Marques, P., Tippetts, S., & Voas, R. (2003). The alcohol interlock: An underutilized resource for predicting and controlling drunk drivers. Traffic Injury Prevention, 4, Marques, P., & Voas, R.B. (2008). Alcohol Interlock Program Features Survey. Unpublished Survey Results from the Interlock Working Group of the International Council of Alcohol Drugs and Traffic Safety. Marques, P., & Voas, R.B. (2005). Interlock BAC tests, Alcohol Biomarkers, and Motivational Interviewing: Methods for Detecting and Changing High-Risk Offenders. In P. Marques (Ed.), International Council on Alcohol, Drugs and Traffic Safety (ICADTS): Alcohol Ignition Interlock REFERENCES 115

217 Devices Volume II: Research, Policy, and Program Status 2005, (pp ). The ICADTS Working Group on Alcohol Ignition Interlocks. Pacific Institute for Research and Evaluation. Maruschak, L.M. (1999). DWI Offenders Under Correctional Supervision. Washington, D.C.: Bureau of Justice Statistics, U.S. Department of Justice. Maxwell, S. (2005). Rethinking the broad sweep of recidivism: A task for evaluators. Criminology & Public Policy, 4(3), Maxwell, J.C., & Freeman, J. (2007). Gender differences in DUI offenders in treatment in Texas. Traffic Injury Prevention, 8, Maxwell, J.C. (2011). Drunk versus drugged: How different are the drivers? Drug and Alcohol Dependence, doi: /j.drugalcdep Mayhew, D.R., Brown, S.W., & Simpson, H.M. (2010). The Alcohol Crash Problem in Canada: Ottawa: Transport Canada, Road Safety and Motor Vehicle Regulation. Mayhew, D.R. Brown, S.W., & Simpson, H.M. (2011). The Alcohol-Crash Problem in Canada: Ottawa: Transport Canada, Road Safety and Motor Vehicle Regulation. Mayhew, D.R., Ferguson, S.A., Desmond, K.J., & Simpson, H.M. (2003). Trends in fatal crashes involving female drivers, Accident Analysis and Prevention, 35(3), Mayhew, D.R., Brown, S.W., & Simpson, H.M. (1995). Alcohol Use Among Drivers and Pedestrians Fatally Injured in Motor Vehicle Accidents: Canada, TP Ottawa: Transport Canada, Road Safety. Mayhew, D.R., Brown, S.W., & Simpson, H.M. (1996). Alcohol Use Among Drivers and Pedestrians Fatally Injured in Motor Vehicle Accidents: Canada, TP Ottawa: Transport Canada, Road Safety. Mayhew, D.R., Brown, S.W., & Simpson, H.M. (1997). Alcohol Use Among Drivers and Pedestrians Fatally Injured in Motor Vehicle Accidents: Canada, TP Ottawa: Transport Canada, Road Safety. Mayhew, D.R., Brown, S.W., & Simpson, H.M. (1998). Alcohol Use Among Drivers and Pedestrians Fatally Injured in Motor Vehicle Accidents: Canada, TP Ottawa: Transport Canada, Road Safety. Mayhew, D.R., Brown, S.W., & Simpson, H.M. (1999). Alcohol Use Among Drivers and Pedestrians Fatally Injured in Motor Vehicle Accidents: Canada, Ottawa: Transport Canada, Road Safety. Mayhew, D.R., Warren, R.A., Simpson, H.M., & Hass, G.C. (1981). Young Driver Accidents: Magnitude and Characteristics of the Problem. Ottawa: Traffic Injury Research Foundation. Mayhew, D.R., & Simpson, H.M. (1990). New to the Road Young Drivers and Novice Drivers: Similar Problems and Solutions? Ottawa: Traffic Injury Research Foundation. Mayhew, D.R., Warren, R.A., Simpson, H.M., & Hass, G.C. (1990). Young Driver Accidents: Magnitude and Characteristics of the Problem. Ottawa: Traffic Injury Research Foundation. 116 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

218 Meck, D.S., & Baither, E. (1980). The relation of age to personality adjustment among DWI offenders. Journal of Clinical Psychiatry, 36, McMillen, D.L., Adams, M.S., Wells-Parker, E., Pang, M.G., & Anderson, B.J. (1992a). Personality traits and behaviors of alcohol-impaired drivers: A comparison of first and multiple offenders. Addictive Behaviors, 17, McMillen, D.L., Pang, M.G., Wells-Parker, E. & Anderson, B.J. (1992b). Alcohol, personality traits, and high risk driving: a comparison of young drinking driver groups. Addictive Behaviors, 17, McMurran, M., Riemsma, R., Manning, N., Misso, K., & Kleijnen, J. (2011). Interventions for alcohol-related offending by women: A systematic review. Clinical Psychology Review, 31, Milkman, H.B., Wanberg, K.W., & Gagliardi, B.A. (2008). Criminal Conduct and Substance Abuse Treatment for Women in Correctional Settings: Female-Focused Strategies for Self- Improvement and Change, Pathways to Responsible Living. Thousand Oaks, CA: Sage Publications, Inc. Miller, S., & Brodsky, S. (2011). Risky business: Addressing the consequences of predicting violence. Journal of the American Academy of Psychiatry and Law, 39, Miller, W.R., & Rollnick, S., (2002). Motivational Interviewing: Preparing People to Change, (2nd ed.). NY: Guilford Press. Moore, R.H. (1994). Underage female DUI offenders: Personality characteristics, psychosocial stressors, alcohol and other drug use and driving risk. Psychological Reports, 74, Nadeau, L. (2010, October). The impact of drink driving in Quebec and the management of drunk drivers. Presented at the 11th Annual International Alcohol Interlock Symposium, Montebello, Quebec. Nagin, D.S., Cullen, F.T., & Lero Jonson, C. (2008). Imprisonment and Re-offending. In M. Tonry (Ed.), Crime and Justice: A Review of Research, Vol 23. Chicago: University of Chicago Press. National Highway Traffic Safety Administration (NHTSA). (1986). The Drunk Driver and Jail. The Drunk Driver and the Jail Problem. Volume 1. Washington, D.C.: U.S. Department of Transportation. National Highway Traffic Safety Administration (NHTSA). (1996). Traffic Safety Facts Washington, D.C.: National Center for Statistics and Analysis. National Highway Traffic Safety Administration (NHTSA). (2001). Age of Drinking Onset, Driving After Drinking, and Involvement in Alcohol-related Motor Vehicle Crashes Discussion. Washington, D.C.: U.S. Department of Transportation. National Highway Traffic Safety Administration (NHTSA). (2003). Traffic Safety Facts 2002: Alcohol. DOT HS Washington, D.C.: U.S. Department of Transportation. National Highway Traffic Safety Administration (NHTSA). (2005). Toward a Comprehensive Strategy to Stop Impaired Driving: Alcohol Screening and Brief Intervention Overview. DOT HS Washington, D.C.: U.S. Department of Transportation. REFERENCES 117

219 National Highway Traffic Safety Administration (NHTSA). (2007). Screening and Brief Intervention Tool Kit for College and University Campuses. DTNH22-02-H Washington, D.C.: U.S. Department of Transportation. National Highway Traffic Safety Administration (NHTSA). (2009). Alcohol-Impaired Drivers Involved in Fatal Crashes, by Gender and State, Traffic Safety Facts DOT HS Washington, D.C.: U.S. Department of Transportation. National Highway Traffic Safety Administration (NHTSA). (2010). Traffic Safety Facts Data: Alcohol-impaired Driving. DOT HS Washington, D.C.: U.S. Department of Transportation. National Highway Traffic Safety Administration (NHTSA). (2012). Traffic Safety Facts Data: Alcohol-impaired Driving. DOT HS Washington, D.C.: U.S. Department of Transportation. National Institute on Drug Abuse (NIDA). (2006). Principles of Drug Abuse Treatment forjustice Populations: A Research-based Guide. NIH Publication No Washington, D.C.: National Institutes of Health, U.S. Department of Health and Human Services. National Institutes of Health and National Institute on Alcohol Abuse and Alcoholism. (2005). Alcohol Alert: Brief interventions (NIAAA Publication No. 66). Available online: niaaa.nih.gov/publications/aa66/aa66.htm. Nelson, S.E., Shaffer, H., & Belkin, K. (2012, May). Developing a Computerized Assessment and Referral System. Harvard Medical School: Division on Addictions. Presented at the National Association of Drug Court Professionals Annual Conference, Nashville, TN. Newton, N., Teeson, M., Vogl, L., & Andrews, G. (2010). Internet-based prevention for alcohol and cannabis use: Final results of the Climate Schools course. Addiction, 105, Nochajski, T., & Wieczorek, W. (1998). Identifying potential drinking-driving recidivists: Do non-obvious indicators help? Journal of Prevention and Intervention in the Community, 17, Nochajski, T. H., Miller, B. A., & Wieczorek, W. F. (1989). Criminal history and DWI recidivism. The Problem-Drinker Driver Project, Research Note Nochajski, T.H., Miller, B.A., Wieczorek, W.F., & Whitney, R. (1993). The effects of a drinkerdriver treatment program: Does criminal history make a difference? Criminal Justice and Behavior, 20, Nochajski, T.H., & Wieczorek, W.F. (1997). Study shows BAC not a reliable predictor of drinking and driving recidivism, new screen more comprehensive. In D. Foley (Ed.), Impaired Driving Update, I, No. 2, Kingston, NJ: Civic Research Institute. Nochajski, T.H., & Wieczorek, W.F. (2000). Driver characteristics as a function of DWI history. In H. Laurell and F. Schlyter (Eds.), Proceedings of the 15th International Conference on Alcohol, Drugs and Traffic Safety T-2000, (pp. 103). Stockholm, Sweden: Ekonomi-Print. Nochajski, T.H., & Stasiewicz, P.R. (2006). Relapse to driving under the influence (DUI): A review. Clinical Psychology Review, 26, IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

220 Noell, J., & Glasgow, R. (1999). Interactive technology applications for behavioural counseling: Issues and opportunities for health care settings. American Journal of Preventive Medicine, 17, Oglaff, J., & Davis. M. (2004). Advances in offender assessment and rehabilitation: Contribution of the risk-needs-responsivity approach. Psychology, Crime, and Law, 10(3), Ouimet, M.C., Brown, T.G., Nadeau, L., Lepage, M., Pelletier, M., Couture, S., Tremblay, J., Legault, L., Dongier, M, Gianoulakis, C., & Ng Ying Kin, N.M.K. (2007). Neurocognitive characteristics of DUI recidivists. Accident Analysis and Prevention, 39, Owens, B. (2001). Accurate treatment reporting. Impaired Driving Update, 5(3), Palmer, E.J., Hatcher, R.M., McGuire, J., Bilby, C.A.L., & Hollin, C.R. (2011). The effect on reconviction of an intervention for drink-driving offenders in the community. International Journal of Offender Therapy and Comparative Criminology. Available online: sagepub.com.prox.lib.ncsu.edu/content/early/2011/04/20/ x full.pdf+html Parks, K.A., Nochajski, T.H., Wieczorek, W.F., & Miller, B.A. (1996). Assessing alcohol problems in female DWI offenders. Alcoholism: Clinical and Experimental Research, 20(3), Parsons, O. (1998). Neurocognitive deficits in alcoholics and social drinkers: A continuum? Alcoholism: Clinical and Experimental Research, 22(4), Peck, R., Arstein-Kerslake, G., & Helander, C. (1994). Psychometric and biographical correlates of drunk driving recidivism and treatment program compliance. Journal of Studies on Alcohol, 55(6), Peck, R., Gebers, M.A., Voas, R.B., & Romano, E. (2008). The relationship between blood alcohol concentration (BAC), age, and crash risk. Journal of Safety Research, 39, Pemberton, M., Williams, J., Herman-Stahl, M., Calvin, S., Bradshaw, M., Bray, R., Ridenhour, J., Cook, R., Hersch, R., Hester, R., & Mitchell, G. (2010). Evaluation of two web-based alcohol interventions in the U.S. military. Journal of Studies on Alcohol and Drugs, 72(3), Perreault, S. (2013). Impaired Driving in Canada, Juristat (Catalogue no X ). Ottawa: Statistics Canada. Available online: article/11739-eng.pdf Perrine, M.W. (1975). The Vermont driver profile: A psychometric approach to early identification of potential high-risk drinking drivers. In S. Israelstam and S.Lamber (Eds.), Proceedings of the Sixth International Conference on Alcohol, Drugs and Traffic Safety, (pp ). Toronto: Addiction Research Foundation. Perrine, M.W. (1990). Who are the drinking drivers? The spectrum of drinking drivers revisited. Alcohol Health & Research World, 14(1), PEW Charitable Trusts. (2012). Time Served: The High Cost, Low Return of Longer Prison Terms. Washington, D.C.: Author. REFERENCES 119

221 Poikolainen, K. (1999). Effectiveness of brief interventions to reduce alcohol intake in primary health care populations: A meta-analysis. Preventative Medicine, 28, Popkin, C.L. (1991). Drinking and driving by young females. Accident Analysis and Prevention, 23(1), Prochaska, J., DiClemente, C., & Norcross, J. (1992). In search of how people change: Applications to addictive behaviours. American Psychologist, 47, Project MATCH Research Group (1997). Matching alcoholism treatment to client heterogeneity: Project MATCH post-treatment drinking outcomes. Journal of Studies on Alcohol, 58, Rauch, W., Zador, D., Ahlin, E., Baum, H., Duncan, D., Beck, K., Raleigh, R., Joyce, J., & Gretsinger, N. (2002). Alcohol-impaired driving recidivism among first offenders more closely resembles that of multiple offenders. In D. Mayhew & C. Dussault (Eds.) Alcohol, Drugs and Traffic Safety T2002. Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety, Montreal, August 4-9, 2002 (pp ). Quebec City: Société de l assurance automobile du Québec. Rauch, W., Zador, P.L., Ahlin, E.M., Howard, J.M., Frissell, K.C., & Duncan, G.D. (2010). Risk of alcohol-impaired driving recidivism among first offenders and multiple offenders. American Journal of Public Health, 100(5), Riper, H., Kramer, J., Conijn, B., Smit, F., Schippers, G., & Cuijpers, P. (2009). Translating effective web-based self-help for problem drinking into the real world. Alcoholism: Clinical and Experimental Research, 33(8), Robertson, R.D., Holmes, E., & Marcoux, K. (2013). Female Drunk Drivers: A Qualitative Study. Ottawa: Traffic Injury Research Foundation. Robertson, R.D., & Simpson, H.M. (2002, July) DWI enforcement: Solutions to nine common problems. Police Chief Magazine, Robertson, A., Liew, H., & Gardner, S. (2011a). An evaluation of the narrowing gender gap in DUI arrests. Accident Analysis and Prevention, 43, Robertson, R.D., Marcoux, K.D., Holmes, E.A., & Vanlaar, W. (2011b). State of Knowledge: Female Drunk Drivers. Ottawa: Traffic Injury Research Foundation. Rollnick, S., & Miller, W.R. (1995). What is motivational interviewing? Behavioural and Cognitive Psychotherapy, 23, Rooke, S., Thorsteinsson, E., Karpin, A., Copeland, J., & Allsop, D. (2010). Computer-delivered interventions for alcohol and tobacco use: A meta-analysis. Addiction, 105, Royal, D. (2003). National Survey on Drinking and Driving Attitudes and Behaviors: Washington, D.C.: U.S. Department of Transportation. Schell, T., Chan, K., & Morral, A. (2006). Predicting DUI recidivism: Personality, attitudinal, and behavioral risk factors. Drug and Alcohol Dependence, 82, IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

222 Schwartz, J., & Rookey, B.D. (2008). The narrowing gender gap in arrests: Assessing competing explanations using self-report, traffic fatality, and official data on drunk driving, Criminology, 46(3), Schwartz, J., & Steffensmeier, D. (2007). The Nature of Female Offending: Patterns and Explanation. In R. Zaplin (Ed.), Female Offenders: Critical Perspective and Effective Interventions, (pp ). Boston: Jones and Bartlett. Schuckit, M. (1999). New findings in the genetics of Alcoholism. Journal of the American Medical Association, 281(20), Schuckit, M. (2009). An overview of genetic influences in alcoholism. Journal of Substance Abuse, 36(1), S5-14. Selzer, M.L, Payne, C.E., Gifford, J.D., & Kelly, W.L. (1963). Alcoholism, mental illness and the drunk driver. American Journal of Psychiatry, 120, Shaffer, H.J., Nelson, S.E., LaPlante, D.A., LaBrie, R.A., & Albanese, M. (2007). The epidemiology of psychiatric disorders among repeat DUI offenders accepting a treatmentsentencing option. Journal of Consulting and Clinical Psychology, 75(5), Shore, E.R., & McCoy, M.L. (1987). Recidivism among female DUI offenders in a Midwestern American city. Journal of Criminal Justice, 15(5), Simon, J. (2007). Governing Through Crime: How the War on Crime Transformed American Democracy and Created a Culture of Fear. New York: Oxford University Press. Simpson, H.M., & Mayhew D.R. (1991). The Hard Core Drinking Driver. Ottawa: Traffic Injury Research Foundation. Simpson, H.M., Mayhew, D., & Beirness, D. (1996). Dealing with the Hard Core Drinking Driver. Ottawa: Traffic Injury Research Foundation. Simpson, H.M., & Robertson, R.D. (2001). DWI System Improvements for Dealing with Hard Core Drinking Drivers: Enforcement. Ottawa: Traffic Injury Research Foundation. Simpson, H.M., Beirness, D.J., Robertson, R.D., Mayhew, D.R. & Hedlund, J.H. (2004). Hard core drinking drivers. Traffic Injury Prevention, Special Issue 5(3), Steer, R.A., & Fine, E.W. (1978). Mood differences in men arrested once and men arrested twice for driving while intoxicated. Journal of Studies on Alcohol, 39(5), Substance Abuse and Mental Health Services Administration (SAMHSA). (2005). Substance Abuse Treatment for Adults in the Justice System: A Treatment Improvement Protocol. TIP 44. Washington, D.C.: U.S. Department of Health and Human Services, Center for Substance Abuse Treatment. Substance Abuse and Mental Health Services Administration (SAMHSA). (2012). Results from the 2011 National Survey on Drug Use and Health: Summary of National Findings, NSDUH Series H-44, HHS Publication No. (SMA) Rockville, MD: SAMHSA. Syrcle, J., & White, W. (2006). The Illinois DUI Risk Reduction Project: ASUDS-RI Pilot Phase I (Statistical Summary). Springfield, IL: Center for Legal Studies, University of Illinois-Springfield. REFERENCES 121

223 Taxman, F.S., & Piquero, A. (1998). On preventing drunk driving recidivism: An examination of rehabilitation and punishment approaches. Journal of Criminal Justice, 26, Taxman, F.S., Perdoni, M., & Harrison, L. (2007). Treatment and adult offenders: The state of the state. Journal of Substance Abuse Treatment, 32(3), Taxman, F., Shepardson, E., Byrne, J., Gelb, A., & Gornik, M. (2004). Tools of the Trade: A Guide to Incorporating Science into Practice. National Institute of Corrections, U.S. Department of Justice; Maryland Department of Public Safety and Correctional Services. Available online: Teichner, G., Horner, M.D., Roitzsch, J.C., Herron, J. & Thevos, A. (2002). Substance abuse treatment outcomes for cognitively impaired and intact outpatients. Addictive Behaviors, 27, Timken, D.S. (2002). What Works: Effective DWI Interventions. In Harry Allen (Ed.), Risk Reduction: Interventions for Special Needs Offenders (pp ). Lanham, MD: American Correctional Association. Traffic Injury Research Foundation (2012). Internal, unpublished database query. Tsai, V.W., Anderson, C.L., & Vaca, F.E. (2008). Young female drivers in fatal crashes: Recent trends, Traffic Injury Prevention, 9(1), Tsai, V.W., Anderson, C.L., & Vaca, F.E. (2010). Alcohol involvement among young female drivers in U.S. fatal crashes: Unfavorable trends. Injury Prevention, 16, Underhill, B.L. (1986). Driving under the influence of gender discrimination. Alcoholism Treatment Quarterly, 3, Velasquez, M.M., Stephens, N.S., & Ingersoll, K. (2006). Motivational interviewing in groups. Journal of Groups in Addiction & Recovery, 1(1), Vernon, M. (2010). A review of computer-based alcohol problem services designed for the general public. Journal of Substance Abuse, 38(3), Vingilis, E.R. (1983). Drinking drivers and alcoholics: Are they from the same population? In R. Smart, F. Glasser, Y. Israel et al. (Eds.), Research Advances in Alcohol and Drug Problems, Vol.7., (pp ). New York: Plenum Publishing Corporation. Vingilis, E. (2000). Driver characteristics: What have we learnt and what do we still need to know? In H. Laurell and F. Schylter (Eds.), Proceedings of the 15th International Conference on Alcohol, Drugs, and Traffic Safety T2000. Voas, R. B., Tippetts, A. S., & McKnight, A.S. (2010). DUI offenders delay license reinstatement: A problem. Alcoholism: Clinical and Experimental Research, 34(7), Waller, J.A. (1967). Identification of problem drinking among drunken drivers. Journal of the American Medical Association, 200, Waller, P. (1997). Women, Alcohol, and Traffic Safety. University of Michigan. 122 IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

224 Wanberg, K.W., Milkman, H.B., & Timken, D.S. (2010) Driving with Care Education and Treatment of the Underage Impaired Driving Offender: An Adjunct Provider s Guide. Thousand Oaks: Sage. Wanberg, K., Milkman, H., & Timken, D. (2005). Driving With Care: Education and Treatment of the Impaired Driving Offender. Strategies for Responsible Living and Change. New York: Sage Publishing. Wanberg, K.W., & Timken, D.S. (2006). Adult Substance Use and Driving Survey Revised Illinois (ASUDS-RI). Arvada, CO: Center for Addictions Research and Evaluation. Wanberg, K. W., & Lowe, N. (2013). Screening Risk and Needs among DWI Probationers: Preliminary Results of the Impaired Driving Assessment Pilot Study. Presented at the Winter Training Institute of the American Probation and Parole Association, Phoenix, AZ. Ward, T., Mesler, J., & Yates, P. (2007). Reconstructing the Risk-Need-Responsivity model: A theoretical elaboration and evaluation. Aggression and Violent Behavior, 12, Webb, T., Joseph, J., Yardley, L., & Michie, S. (2010). Using the internet to promote health behaviour change: A systematic review and meta-analysis of the impact of theoretical basis, use of behaviour change techniques, and mode of delivery on efficacy. Journal of Medical Internet Research, 12(1), e4. Webster, J.M., Pimentel J.H., Harp, K.L.H., Clark, D.B., & Staton-Tindall, M. (2009). Substance abuse problem severity among rural and urban female DUI offenders. American Journal of Drug and Alcohol Abuse, 35(1), Wechsler, H., Lee, J.E., Nelson, T.F., & Lee, H. (2003). Drinking and driving among college students: The influence of alcohol-control policies. American Journal of Preventive Medicine, 25(3), Wells-Parker, E., Bangert-Drowns, R., McMillen, R., & Williams, M. (1995). Final results from a meta-analysis of remedial interventions with drink/drive offenders. Addiction, 90, Wells-Parker, E., Pang, M.G., Anderson, B.J., McMillen, D.L., & Miller, D.I. (1991). Female DUI offenders: A comparison to male counterparts and an examination of the effects of intervention on women s recidivism rates. Journal of Studies on Alcohol, 52(2), Wendling, A., & Kolody, B. (1982). An evaluation of the Mortimer-Filkins test as a predictor of alcohol-impaired driving recidivism. Journal of Studies on Alcohol, 43(7), Weisheit, R.A., & Klofas, J.M. (1992). Social status of DUI offenders in jail. The International Journal of the Addictions, 27(7), White, W. & Gasperin, D. (2006, unpublished manuscript). Management of the High Risk Driving Offender. Coordinated by the Institute for Legal and Policy Studies, Center for State Policy and Leadership, University of Illinois at Springfield. Retrieved from ILAPS/TrainingPrograms?DUIMonograph/ManagementHighRiskOffender.htm White, W., & Hennessey, M. (2006). Evaluating, Treating and Monitoring the Female DUI Offender. REFERENCES 123

225 Wieczorek, W.F., & Nochajski, T.H. (2004). Multiple treatment experiences as a predictor of continued drinking-driving. In J. Oliver, P.M. Williams and A.B. Clayton (Eds.), Proceedings of the 17th International Conference on Alcohol, Drugs and Traffic Safety, Paper O47. Glasgow, Scotland: Scottish Executive. Wieczorek, W.F., & Nochajski, T. (2005). Characteristics of Persistent Drinking Drivers: Comparisons of First, Second, and Multiple Offenders. In D. Hennessey and D. Wiesenthal (Eds.), Contemporary Issues in Road User Behavior and Traffic Safety (pp ). New York: Nova Science Publishers Inc. Wieczorek, W.F., Miller, B.A., & Nochajski, T.H. (1992). The limited utility of BAC for identifying alcohol-related problems among DWI offenders. Journal of Studies on Alcohol, 53(5), Wilsnack, R.W., Wilsnack, S.C., & Klassen, A.D. (1984). Women s Drinking and Drinking Problems: Patterns from a 1981 National Survey. American Journal of Public Health, 74(11), Wilson, R.J. (1991). Subtypes of DWIs and high-risk drivers. Alcohol, Drugs and Driving, 7(1), Wilson, R.J. (1992). Convicted impaired drivers and high-risk drivers: How similar are they? Journal of Studies on Alcohol, 53, Wilson, J., & Jonah, B. (1985). Identifying impaired driver among the general driving population. Journal of Studies on Alcohol, 46(6), World Health Organization (WHO). (2010). Screening and Brief Intervention for Alcohol Problems in Primary Health Care. Retrieved from: activities/sbi/en/ World Health Organization (WHO). (2013). Lexicon of alcohol and drug terms published by the World Health Organization. Retrieved from: terminology/who_lexicon/en/ Yoder, R., & Moore, R.A. (1973). Characteristics of convicted drunk drivers. Quarterly Journal of Studies on Alcohol, 34, Zador, P.L., Krawchuk, S.A., & Voas, R.B. (2000). Alcohol-related relative risk of driver fatalities and driver involvement in fatal crashes in relation to driver age and gender: An update using 1996 data. Journal of Studies on Alcohol, 61, Zinn, S., Stein, R., & Swartzwelder, H.S. (2004). Executive functioning early in abstinence from alcohol. Alcoholism: Clinical and Experimental Research, 28(9), Zuckerman, M. (2000). Are you a risk taker? Psychology Today. November-December, p.54-56; IMPAIRED DRIVING RISK ASSESSMENT: A PRIMER FOR PRACTITIONERS

226 T R A F F I C I N J U R Y R E S E A R C H F O U N D A T I O N T R A F F I C I N J U R Y R E S E A R C H F O U N D A T I O N Traffic Injury Research Foundation (TIRF) 171 Nepean Street, Suite 200 Ottawa, Ontario Canada K2P 0B4 Toll Free: Fax: Registered Charity No RR0001

227 FOURTH WAIVER AND ECPA SEARCH/SEIZURE CONSENT San Diego Superior Court Case # Defendant (print) (Full Legal Name) IT IS HEREBY ORDERED that the above identifed defendant shall: submit his/her person, vehicle, residence, property, personal effects, and anything under the defendant s custody or control to search at any time with or without warrant, and with or without reasonable cause, when requested by a Probation Officer or law enforcement officer. The defendant is ordered to volunteer and disclose to any law enforcement officer who contacts him/her that he/she has a 4 th waiver. In addition to the 4 th waiver conditions listed above, the defendant provides specific consent within the meaning of P.C et seq. to whatever government entity is seeking information protected by the California Electronic Communication Protection Act. This consent includes consent to seize and examine call logs, text and voic messages, photographs, s, and social media account contents contained on any device or cloud or internet connected storage owned, operated, or controlled by the defendant, including but not limited to cell phones, computers, computer hard drives, laptops, gaming consoles, mobile devices, tablets, storage media devices, thumb drives, Micro SD cards, external hard drives, or any other electronic storage devices, by whatever government entity is seeking the information. The defendant shall also disclose any and all passwords, passcodes, password patterns, fingerprints, or other information required to gain access into any of the aforementioned devices or social media accounts. Dated: Judge of the Superior Court I acknowledge and accept the above described 4 th waiver conditions and consent as indicated. Dated: Defendant FOURTH WAIVER ORDER AND ECPA SEARCH/SEIZURE CONSENT

228 T R A F F I C I N J U R Y R E S E A R C H F O U N D A T I O N FEMALE DRUNK DRIVERS: A QUALITATIVE STUDY

229 The Traffic Injury Research Foundation The mission of the Traffic Injury Research Foundation (TIRF) is to reduce traffic-related deaths and injuries. TIRF is a national, independent, charitable road safety research institute. Since its inception in 1964, TIRF has become internationally recognized for its accomplishments in a wide range of subject areas related to identifying the causes of road crashes and developing programs and policies to address them effectively. This report was made possible by a grant from The Century Council. Traffic Injury Research Foundation (TIRF) 171 Nepean Street, Suite 200 Ottawa, Ontario Canada K2P 0B4 Toll Free: Fax: Registered Charity No RR0001 August 2013 Copyright 2013 ISBN:

230 FEMALE DRUNK DRIVERS: A QUALITATIVE STUDY HISTORY AND EXPERIENCES IN THE SYSTEM Robyn D. Robertson Erin Holmes Kyla Marcoux

231

232 ACKNOWLEDGEMENTS TIRF gratefully acknowledges the assistance of the following individuals who provided assistance in organizing focus groups and recruiting participants for this project. Marilyn Gibson Drug Court Administrator Greene County Drug Court Missouri Bob Iusi Chief Warren County Probation New York Julia Scott Senior Office Coordinator Collaborative Justice Courts San Joaquin County California Mary Ellen Still Director Dutchess County Probation New York Edmundo Varela Assistant Commissioner Westchester County Department of Probation New York FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Acknowledgements iii

233 TIRF also wishes to thank the following individuals for their assistance and input to inform the development of discussion guides for justice and treatment professionals. Mary Ann Mowatt Research Associate American Probation and Parole Association Ward Vanlaar Vice President Research Traffic Injury Research Foundation TIRF extends its gratitude to the 186 women who participated in interview focus groups, in-depth interviews, and the survey who willingly shared their experiences, perspectives, and insights in relation to drinking and driving. Without their courage and cooperation this work would not have been possible. TIRF also thanks the reviewers of this report who graciously assisted in its development and who provided feedback and comments on earlier drafts. Their knowledge and perspectives allowed us to create a user-friendly and useful report that can benefit practitioners across the country. Honorable Peggy Davis Drug Court Commissioner Greene County Drug Court Missouri Francine Perretta Deputy Commissioner Westchester County Department of Probation New York Judge Richard Vlavianos Superior Court of California County of San Joaquin California Officer Sheryl Day Forward Motion Female DWI Offender Program Westchester County Department of Probation New York iv Acknowledgements FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

234 Judge Susan Jonas 58th District Court Ottawa County, Michigan Norma McGuire Probation Officer Dutchess County Office of Probation and Community Corrections New York Carl Wicklund Executive Director American Probation and Parole Association Sam De Alba DUI Court Compliance Officer/Case Manager Superior Court of California County of San Joaquin California FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Acknowledgements v

235 ACKNOWLEDGEMENTS EXECUTIVE SUMMARY TABLE OF CONTENTS III VIII 1. INTRODUCTION 1 2. OBJECTIVES AND METHODS 5 3. FEMALE DRUNK DRIVER RESULTS Demographic information Age Education and employment Marital and family status Family history and environment Physical health Criminal history General profiles of female drunk drivers Young women Recently married women with children Divorced older women and/or empty nesters Substance use Mental health Life situation prior to offense Experiences in the criminal justice system Arrest Court process Sentencing and supervision Experiences in the treatment system Emotional effects of their experience Summary PRACTITIONER INTERVIEW RESULTS General characteristics of female drunk drivers who are supervised or treated Arrest and crash characteristics 47 vi FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

236 4.1.2 Demographic characteristics Substance use Mental health Other driving and criminal history Experiences of practitioners Criminal justice system Experiences with supervision and treatment Lessons learned What to expect supervising and treating female drunk drivers Strategies for supervising and treating female drunk drivers RECOMMENDATIONS Prevention Criminal justice system recommendations Recommendation from offenders Recommendations from practitioners Treatment system recommendations Recommendations from offenders Recommendations from practitioners CONCLUSIONS REFERENCES APPENDIX - CASE STUDIES Jurisdictional profile: California Jurisdictional profile: Missouri Jurisdictional profile: New York Jurisdictional profile: Michigan 113 FEMALE DRUNK DRIVERS A QUALITATIVE STUDY vii

237 EXECUTIVE SUMMARY Introduction While males constitute a significant portion of the impaired driving problem, there is evidence of a growing number of DWI 1 arrests among females, and incremental increases among female drivers testing positive for alcohol in fatal crashes in some jurisdictions in the U.S. This suggests that women are an important part of the problem and warrant attention. Although an examination of female self-report data on drinking and driving shows it has remained stable at 10-20% 2, and crash data from the Fatality Analysis Reporting System (FARS) reveals incremental changes in drinking and driving among females in the past three decades (12% in the 1980s to 14% in the 2000s), there has been a dramatic increase in the number of women arrested for drunk driving in just the past decade. To illustrate, the number of female DWI arrests has risen nationally by 28.8% between 1998 and A majority of the available research examines male drunk driving offenders. Women account for a much smaller proportion of the problem (approximately 20%), making it difficult to conduct meaningful research about this population. Moreover, much of this research is outdated. Hence, it does 1 The abbreviation DWI (driving while impaired or intoxicated) is used throughout this report as a convenient descriptive label, even though some states use other terms such as OUI (operating under the influence) or DUI (driving under the influence), and in some states they refer to different levels of severity of the offense. We have used DWI not only to maintain consistency throughout the report but also because it is more descriptive of the offense usually associated with drunk drivers. 2 Drew et al. 2010; Royal 2003; Schwartz and Rookey 2008; Wilsnack et al NHTSA 2009; Schwartz and Steffensmeier 2007; Lapham et al. 2000; Schwartz and Rookey viii Executive Summary FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

238 not provide an accurate or complete picture of female drunk drivers today; nor does it identify effective programs and interventions specific to this population. What is known is that there are important differences between male and female DWI offenders. To summarize: > Many female drunk drivers have substance abuse issues which they tend to develop later in life than male counterparts. 4 > A majority of these women are likely to be single, separated, divorced, or be living with a partner with an alcohol problem. 5 > Female offenders tend to be older than males and have higher levels of education but lower paying jobs. 6 > Female DWI offenders have significantly higher co-morbidity relative to males. 7 > Several factors contribute to their alcohol use including a family history of alcoholism, history of abuse or trauma, mental health issues, and relationship problems. 8 > Little is known about effective programs and interventions for convicted female drunk drivers, although data suggest that they account for 15-25% of DWI offenders in traditional drunk driving programs such as alcohol monitoring and DWI Courts. Available research mainly focuses on treatment effectiveness among substance abusing females. Features of effective programming include provision of childcare and transportation options in conjunction with access to treatment, 9 customized treatment to meet individual risks and needs, 10 individual counseling, 11 and womenonly programs or women-only group therapy. 12 Purpose and objectives The findings stemming from TIRF s 2011 literature review on this topic revealed important gaps in knowledge pertaining to their profile and characteristics, experiences in the criminal justice and treatment systems, and the types of 4 White and Hennessey Argeriou et al. 1986; Chang et al. 1996; McMurran et al. 2011; Shore and McCoy Chalmers et al. 1993; Shore and McCoy Maxwell and Freeman White and Hennessey Green Freeman et al Sun Grella and Greenwell FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Executive Summary ix

239 strategies and interventions that are most effective with this population. To address these gaps and increase understanding of this problem, a follow-up qualitative study using a case study approach was conducted in The objectives of the study are as follows: > Create a foundation that could inform the development of much needed research initiatives as well as prevention efforts and effective interventions tailored towards female drunk drivers (i.e., hypothesis-generating as opposed to hypothesis-testing). > Explore the life histories of convicted female drunk drivers and the ways that their history may contribute to their offending. > Examine women s experiences in the criminal justice and treatment systems. > Explore the experiences of criminal justice and treatment professionals in supervising this offender population. Case studies were conducted in four sites (San Joaquin County, California; Greene County, Missouri; Ottawa County, Michigan; and Dutchess, Warren, and Westchester Counties in New York). A multi-faceted research design was utilized that included interview focus groups with 154 first and repeat offenders to explore their attitudes, behaviors, characteristics, risk and needs, and pathways to offending as well as their experiences in the criminal justice and treatment systems. Key informant interviews were also conducted with 36 experienced criminal justice and treatment professionals to identify how female drunk drivers are managed within these systems and what has been learned from their experiences. Lastly, a survey was administered to 28 female offenders in California who were unable to attend the focus groups. The data obtained from these sources were used to identify lessons learned and to formulate recommendations to improve the supervision of female offenders and the delivery of services to them. Female drunk driver results Data gathered during interview focus groups and the survey of offenders, and interviews with practitioners revealed highly consistent findings across the four jurisdictions in this study. These data are estimates based on the data collected. x Executive Summary FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

240 Demographics > Women ranged in ages from late teens to mid-60s, suggesting that women of all ages drink and drive. Early onset and late onset drunk driving behavior were common. > Many participants attained a high school education or its equivalency (three-quarters) and approximately one-third reported having initiated or completed some type of post-secondary education. Occupations generally included nurses, dental assistants, paralegals, teachers, corporate employees, self-employed entrepreneurs, and bartenders. Approximately one-third had worked in bars and restaurants. > Home environments equally included those that were stable with no history of substance use and those that were dysfunctional or abusive where substance use was prevalent and acceptable. > More than half of the women were single, separated, or divorced and the majority had children. Just one-third of them reported that they had some type of support network. > Almost all women reported that their impaired driving arrest was precipitated by a major life stressor such as a domestic argument, the end of a relationship, the loss of a job or child custody, or the illness or death of a parent or other family member. > More than three-quarters of the women used one or more prescription medications for anxiety, depression, post traumatic stress disorder (PTSD), and other disorders. Undiagnosed mental health issues and histories of trauma and/or abuse (both physical and sexual) were not uncommon. > It is estimated that less than 20% of participants reported arrests for other offenses, in addition to impaired driving, such as drug manufacturing/ distribution, theft, and/or fraud. > Less than one-third of participants reported use of illicit substances of which marijuana and methamphetamines were the most common drugs of choice. Profiles With regard to the characteristics of female drunk driving offenders, three different profiles of this population emerged: FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Executive Summary xi

241 1. Young women who drink in order to fit in and consume alcohol and/ or binge drink at house parties and bars; 2. Recently married women with children who drink following the birth of their children as a means for coping with loneliness; and, 3. Divorced older women and/or empty nesters who begin to drink later in life (after age 40) following a catalyst such as the death of a parent, end of a marriage, or children leaving home. Characteristics of female drunk drivers reported in the literature were common across study participants including mental disorders, family history of substance misuse, multiple impaired driving arrests, trauma history, failed relationships, and feelings of shame and guilt. Experiences in the criminal justice system Overall, several interview focus group participants reported generally negative experiences in the justice system at some point, although there were exceptions. Many women defined their experiences in terms of emotional reactions such as shame, frustration, anger, depression, anxiety, and fear. > A majority of women reported that they were arrested within a few blocks of their residence. An estimated one-quarter of the women were arrested as a result of a crash, and a very small minority were involved in severe crashes. > A number of women reported that their blood alcohol content (BAC) was shockingly high and BACs ranging from.16 to.42 were not uncommon. > Women expressed concern about the focus of the system on their offense and the overlooking of the circumstances or underlying factors that contributed to the behavior. > Overall, study participants reported mixed arrest experiences but almost all of them agreed that being incarcerated in jail or prison was a frightening experience. At least one-third of participants spent time in jail or prison and reported it had negative effects. > Many women felt that their sentence or conditions of probation failed to take into account either their life circumstances or address their issues. They also noted that long delays in the court process prevented them from focusing on the future. xii Executive Summary FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

242 > A majority of women reported that information about their period of supervision, conditions, eligibility for certain privileges, or consequences for non-compliance were generally unclear. > Most women consistently agreed that accountability is essential, and that the level of respect, communication, support, and encouragement provided by criminal justice practitioners can make the difference between their success and failure. > Almost all study participants reported that the overall cost of their arrest and subsequent supervision were quite substantial and could be overwhelming, even if they were employed. > More than half of women underscored challenges to comply with random testing requirements due to the lack of available services or extended hours, and testing costs. > Approximately three-quarters of participants reported that securing transportation was very challenging and made it difficult to comply with supervision, treatment, and testing conditions and maintain employment. This was more problematic in rural jurisdictions. > Between one-half and two-thirds of study participants reported having little or no support system to help them manage and meet all of the requirements of their supervision. > Almost all of the women acknowledged that it was difficult to make and maintain lifestyle changes because they felt they lacked the tools, skills, and support system to do so. Experiences in the treatment system While many of the female offenders in the interview focus groups reported that treatment was very beneficial, they also expressed varying degrees of frustration with it as a whole. Satisfaction levels were correlated with their perceptions of whether the intervention they received was specific to their needs. A majority responded favorably to treatment programs that were tailored to women and that included individual counseling and group therapy. > More than 80% of study participants consistently reported insufficient services in relation to substance use treatment programs and mental health services. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Executive Summary xiii

243 > Generic substance use treatment groups were not perceived as beneficial by many of the participants. Individualized and female-specific approaches were identified as the most likely to produce positive and lasting outcomes because they fostered a supportive, understanding, and safe environment that enabled them to share their experiences. > Individual counseling is preferred initially as it provides women with an opportunity to explore issues and share experiences without fear of being judged by others. > The availability of treatment services varies considerably and many women are forced to participate in interventions that are not suited to their needs as a function of cost or access. The presence of insurance typically determines their options. Practitioner interview results Key informant interviews with 36 criminal justice professionals (judges, defense attorneys, probation officers, and alcohol education and treatment providers) were conducted in four states. Many of the themes that emerged from these interviews and many of the reported experiences were highly similar across professions. Data collected from practitioners regarding the profile and characteristics of female drunk drivers were generally consistent with the data provided by female offenders, and are described in the full report. Additional data are summarized below. > Practitioners reported physical health issues were fairly common among women entering the system (e.g., eating disorders, early menopause, hormonal issues), although the nature of these issues varied in accordance to age. > Practitioners indicated that female and male drunk drivers were equally likely to be uninsured and this was a significant problem in relation to the affordability of treatment. > Practitioners agreed that more young female drunk drivers were entering the system. Early onset drinking and significant substance use problems were not uncommon. xiv Executive Summary FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

244 > They also agreed that many female offenders were either in denial about the severity of their drinking or unaware of the extent of their alcohol use. They noted that women seem to experience more shame about their drinking and tend to minimize their use. Experiences of practitioners > Practitioners reported that police are more likely to charge all impaired drivers, that fewer diversion programs are available today, and that while treatment is more readily available, the quality and diversity of these services varies substantially. > There was a high level of consensus among practitioners that female drunk drivers are most often required to participate in mixed-gender treatment in a group setting comprised of individuals with diverse backgrounds and histories of substance use. They also agreed that there are fewer specialized treatment services available today. > Practitioners agreed that younger women failed to acknowledge the seriousness of their first DWI offense. By comparison, practitioners stated that women in their 30s and older more often acknowledge the arrest and court processing as a wake-up call. > A majority of practitioners agreed that the arrest and court process is more likely to be traumatic for women than it is for men. However, they are also more accepting of the process and are less resistant to it than their male counterparts. > More than half of practitioners agreed that the supervision of female drunk drivers can be more effective in a DWI Court setting because of the sharing of information and team approach. > All practitioners agreed that the actions and attitude of probation officers or case managers is an important factor in the successful completion of supervision requirements by female drunk drivers. > A large majority of practitioners reported that female drunk drivers are more receptive to female-only treatment groups; more than half also agreed that individualized treatment or counseling at the outset appears to be more helpful for them. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Executive Summary xv

245 > Practitioners generally agreed that female drunk drivers are more likely to become compliant sooner than males and half of them indicated that female drunk drivers are more likely to successfully complete supervision/ treatment than men and in less time. > Practitioners identified several barriers to successful completion of supervision and treatment that were especially pronounced among female drunk drivers, including limited financial resources, limited access to affordable childcare services, and transportation. Lessons learned Based on the collective experiences of practitioners in the supervision and treatment of female drunk drivers, there were a number of lessons learned that can help inform the development of specific strategies and interventions to better serve this population. > Women are more likely to try and manipulate the system and avoid the requirements of supervision and/or treatment at the outset of the process. > Female drunk drivers often come to supervision and treatment with a wide range of issues including substance misuse, mental health problems or a history of trauma. > Female drunk drivers generally experience more pressure to succeed due to financial and/or childcare responsibilities. While children are often a motivator for female drunk drivers to successfully complete supervision and treatment, this is not true in all cases. > Female drunk drivers often are more emotional about their situation and want to talk about it more so than their male counterparts. > It can take a long period of time for female drunk drivers to recognize their chaotic thinking and to develop strategies to help them defend against extremes in behavior. > Female drunk drivers are more likely to need assistance with relationship dynamics and this is important to their success. > There is not one program or intervention that will work universally for all female offenders; what strategies/interventions are most likely to result in successful outcomes is often a function of the complexity of their drinking problem. xvi Executive Summary FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

246 > Spouses of female drunk drivers are generally less likely to contact practitioners about risk of relapse, unlike the spouses of male drunk drivers. > Anniversaries of traumatic events can trigger intense emotions among female drunk drivers and the potential for relapse should be monitored particularly at these times. > Female drunk drivers are more likely to be pro-social and compliant than males. They often feel pressure to succeed and have more responsibilities for childcare and finances. > Female drunk drivers are more likely to be successful than males in completing their supervision and treatment requirements, particularly if family/friends are supportive of their sobriety. Strategies for supervising and treating female drunk drivers The following strategies can be useful guidelines for practitioners to consider when supervising and managing female drunk drivers: > Drug screens for suspected female drunk drivers at the time of arrest can be a very helpful source of information to identify potential issues with drugs, particularly as women may be less forthcoming about this at the outset. > Accountability and intensive monitoring and support are particularly useful for female drunk drivers at least at the beginning of their supervision period. > It is important to remove issues of ego, shame, and guilt from conversations with female drunk drivers and to focus on the individual and avoid making judgments. > Female drunk drivers often experience stress and anxiety about the many supervision requirements. Informing them at the outset that there is a lot of information and it can be reviewed again in subsequent appointments can help minimize these feelings. > Applying a more comprehensive approach to supervision and treatment can benefit female drunk drivers who often suffer from a broader range of life issues. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Executive Summary xvii

247 > The use of strategies that include incentives to encourage and reinforce compliance can better motivate female drunk drivers and help to build their self-esteem. > Learning to read between the lines in relation to female drunk drivers can help practitioners to more quickly identify custody concerns, domestic violence, and mental health issues that may affect outcomes with this population. Potential flags should be monitored and explored. The supervision of female drunk drivers who experience domestic violence may require additional sensitivity and precautions. > Strategies that assist female drunk drivers in recognizing a lack of boundaries in relationships, and how this contributes to their addiction or behaviors that are connected to their offending are helpful. > It is often useful to work to engage the family of female drunk drivers in their supervision and treatment where possible as they can help support the offender s sobriety. Recommendations Data collected during the interview focus groups and individual interviews with female drunk drivers, and the key informant interviews with practitioners revealed a number of recommendations that can inform efforts to strengthen prevention initiatives and the criminal justice and treatment systems for dealing with female drunk drivers. Recommendations include: Prevention recommendations > Women need to learn at an early age what constitutes normal drinking versus excessive or binge drinking. Other important areas of education include how alcohol is metabolized differently by women and the effects of alcohol on driving performance. > Increased awareness and efforts to address earlier difficult living arrangements, emotional problems and mental disorders can help prevent issues that can contribute to the development of substance use problems and drinking and driving. Criminal justice system recommendations > Provide guidance and assistance to female drunk drivers to help them manage life issues related to their offending. xviii Executive Summary FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

248 > Provide clear information about the conditions of probation, requirements of sentencing, and any additional responsibilities during the initial meeting with a probation officer. > Recognize the individual value and accomplishments and/or progress of offenders. > Be honest, clear and follow through in your interactions with female drunk drivers. > Work with female drunk drivers to achieve a good balance in how their time is scheduled and filled. Consider increased flexibility in the scheduling of probation appointments, testing, and treatment for female offenders as appropriate. > Identify the most appropriate and best suited treatment intervention for each offender as this is a key to their success in recovery. > Minimize stress and discomfort in situations when a female drunk driver is assigned to a new probation officer. > Minimize situations in which female drunk drivers must constantly re-live their story and review the circumstances which led to their offense. Find constructive strategies to deal with emotions and emotional situations. > Work to talk, listen, connect, and develop rapport with the women that are supervised but maintain boundaries. Seek to build the self-esteem and the trust of the women that are supervised. > Manage perceptions around the inconsistent application of sanctions across individual offenders by making clear why sanctions are applied and what factors are considered. > Increase the availability of education for criminal justice practitioners about substance abuse and dependence. > Increase education for criminal justice practitioners about female offender issues and about alcohol monitoring technologies such as ignition interlocks. > Learn more about social work and increase the availability of social services, educational and vocational services for offenders generally to the extent possible. > Develop a different protocol for home visits for female probationers. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Executive Summary xix

249 Treatment system recommendations > Increase the availability of better and more holistic assessments (including substance use, mental health, trauma) at the time of first offense along with referrals to appropriate and intensive treatment interventions for those who require them. > Increase the availability and quality of treatment services; this is in relation to both male and female DWI offenders. Among women, provide more affordable services and offer the option of women-only treatment programs where possible. > Provide women with opportunities to integrate real life responsibilities into treatment so they learn how to cope with stress, and manage their life in conjunction with sobriety. > Counselors should be discouraged from sharing their personal stories about substance use. > Do not require participation in treatment for women who do not have alcohol abuse or dependence issues (as determined by screening and/or assessment). > Increase funding and resources to help women who lack insurance or cannot afford more robust treatment programs to enter programs that can better address individual risks and needs. > Make available to practitioners more research about which treatment interventions and strategies work best with female drunk drivers. > Make available more training and resources to improve the quality of treatment. Provide more training for clinicians and strengthen state treatment certification protocols as appropriate. > Use alcohol-intake instruments that acknowledge and identify a history of trauma as this can have implications for treatment (e.g., it could assist practitioners in making better referrals to more appropriate services). Increase screening for co-occurring disorders. > Provide more outpatient services outside of regular business hours (e.g., in the evening and on weekends). > Increase skills among alcohol education counselors in relation to the moderation of group sessions. xx Executive Summary FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

250 > Increase the availability of aftercare which is important to success following the completion of treatment. Conclusions This study adds to the body of knowledge about female drunk drivers and provides greater insight into their pathways to offending. It also uncovered three distinct and unique profiles of female drunk drivers that begin to shed light on the confluence of factors that play a role in their drunk driving behavior. Today, there are important gaps in existing criminal justice and treatment systems that can make it more challenging for female drunk drivers to successfully complete their sentence and comply with the conditions imposed upon them. There are also important gaps in the interventions and services that are available to women and a need to integrate interventions with relevant community and social services. The results of this study clearly demonstrate the importance of focusing efforts to begin to better understand this problem and to develop more effective strategies both to prevent and manage it. Also of importance, this study sheds light that can inform future research initiatives. In particular, the data collected suggest opportunities for inquiry and further exploration into issues that pertain to the supervision and treatment of female drunk drivers. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Executive Summary xxi

251 1. INTRODUCTION There is no doubt that males constitute a significant proportion of the driving while impaired (DWI 1 ) problem (Argeriou et al. 1986; Jones and Lacey 2001; Zador et al. 2000; Mayhew et al ) and this has been the case for more than three decades. However, evidence of a growing number of DWI arrests among females, and incremental increases among female drivers testing positive for alcohol in fatal crashes in some jurisdictions in the United States suggest that women are an important part of the problem that is worthy of our attention. This is perhaps best illustrated by recent news reports involving women arrested in high-profile drunk driving crashes which have captured the public s attention. Most notable of these was Diane Schuler who drove, with her children in her van, the wrong way down the Taconic State Parkway in New York and killed eight people (four of whom were children) and Carmen Huertas who rolled her car causing a high-speed wreck with seven children in the vehicle. Her BAC was a.12 and the crash resulted in the death of one child. These cases ultimately resulted in the passage of Leandra s Law in New York. Today, our understanding of the impaired driving problem is largely derived from research that has studied predominantly male offenders, often for practical reasons. While there is some research to suggest that female DWI offenders may possess some different characteristics and have different treatment needs, a majority of this research was conducted more than two decades ago. 1 The abbreviation DWI (driving while impaired or intoxicated) is used throughout this report as a convenient descriptive label, even though some states use other terms such as OUI (operating under the influence) or DUI (driving under the influence), and in some states they refer to different levels of severity of the offense. We have used DWI not only to maintain consistency throughout the report but also because it is more descriptive of the offense usually associated with drunk drivers. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Introduction 1

252 Since that time, much has changed, including public attitudes and knowledge about addiction, and the use of laws, programs and policies to target this problem. For these reasons, a closer examination of female involvement in this issue is warranted. Without more research to increase understanding of drunk driving among women, it will be challenging, if not impossible, to develop more refined approaches to prevention, detection, sentencing, supervision, and treatment of this problem among a female population. As a first step towards addressing this gap, the Traffic Injury Research Foundation (TIRF) reviewed some three decades of research on this issue in a 2011 literature review entitled State of Knowledge: Female Drunk Drivers under funding from The Century Council. The report describes the magnitude of the female drunk driver problem, the characteristics of these offenders, the involvement of female drivers testing positive for alcohol in fatal crashes, and what is known about effective strategies. Key findings from the report are briefly summarized below. Magnitude of the problem. Self-report data from different sources indicate that a fairly small percentage (10-20%) of females report drinking and driving and this number has been stable for many years (Drew et al. 2010; Royal 2003; Schwartz and Rookey 2008; Wilsnack et al. 1984). Conversely, arrest data shows DWI arrests among women have risen nationally, and especially in some jurisdictions in this same period (NHTSA 2009; Schwartz and Steffensmeier 2007). In fact, DWI arrests among women increased by 28.8% between 1998 and 2007 (Lapham et al. 2000; Schwartz and Rookey 2008). These increases are believed to be a result of changes in the societal roles of women (e.g., more women driving and entering the workforce), changes in social norms that make it more acceptable for women to drink, and changes in social control mechanisms (e.g., lowering the legal breath alcohol concentration (BAC) limit from.10 to.08). Analyses of data from the U.S. Fatality Analysis Reporting System (FARS) reveal the portion of female drivers in alcohol impaired road crashes has incrementally increased. Women accounted for just 12% of alcohol impaired drivers in the 1980s, 13% in the 1990s, and 14% in the 2000s. Since 2006, the percent of women drivers who tested positive for any amount of alcohol in fatal crashes has averaged 14% and in 2011 of the drivers involved in fatal crashes with a BAC of.08 or greater, 1,567 were female (NHTSA 2012). 2 Introduction FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

253 It has been argued that this increase among women is a result of sharper declines in male compared to female rates of DWI from the 1980s to the 1990s (Schwartz and Rookey 2008). Characteristics of female offenders. For the most part, female drunk driving offenders differ somewhat from their male counterparts, yet they also share some common characteristics. A majority of female drunk drivers experience alcohol problems and the gravity and complexity of those problems is not insignificant (White and Hennessey 2006). Unlike men, they tend to develop substance abuse problems when they are older, and to develop them in a shorter period of time such that they require medical intervention on average four years earlier (Green 2006; McMurran et al. 2011). Older research indicates the average age of female drunk drivers is 31(Shore and McCoy 1987), although younger females are a growing concern (Peck et al. 2008) due to their increasing involvement with alcohol, their inexperience driving, and more recent propensity for risky driving (Lynskey et al. 2007; Tsai et al. 2010). Research about the level of education and employment among female drunk drivers is unclear. Generally, female drunk drivers are older than males and have higher levels of education but lower paying jobs (Chalmers et al. 1993; Shore and McCoy 1987). Female offenders are more likely to be the primary caretaker of children at the time of arrest (Bloom et al. 2003). A significant proportion of these women are single, divorced or separated, or more likely to be living with a partner with an alcohol problem (Argeriou et al. 1986; Chang et al. 1996; McMurran et al. 2011; Shore and McCoy 1987). Research also reveals more female offenders may suffer from mental health problems and have significantly higher psychiatric co-morbidity relative to their male counterparts (Maxwell and Freeman 2007). Diagnoses of anxiety, depression, and post-traumatic stress disorder (PTSD) are common, as is the use of drugs among this population (Maxwell and Freeman 2007; SAMHSA 2005). Many female DWI offenders who were admitted to addiction treatment had multiple factors that contributed to their alcohol use including a history of alcoholism within the family, experience with abuse or trauma, anxiety and depression, and family and personal relationships that encouraged heavy drinking (White and Hennessey 2006). FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Introduction 3

254 Also of concern, it appears that among adult male and female drunk drivers, the risk of recidivism may be similar (Lapham et al. 2000; Rauch et al. 2010). There are also some data which suggest that at least a portion of these women have a history of other traffic offenses or criminal offenses (e.g., drugs, assault, theft), although more research into this topic is needed (Caldwell-Aden et al. 2009). Effective programs and practices. Perhaps of greatest concern, little is known about the effectiveness of programs and interventions for convicted female drunk drivers, although data illustrate that women account for 15-25% of DWI offenders in traditional drunk driving programs such as alcohol monitoring and DWI Courts. Available research focuses mainly on treatment effectiveness among substance abusing females. In brief, once in treatment, there is little difference between males and females generally in terms of effectiveness and this is true for different measures including program retention, completion, and outcomes (Greenfield et al. 2007). A review of studies examining substance abusing women in treatment found that certain characteristics are associated with better outcomes in terms of treatment retention and completion for both men and women. These characteristics include lower levels of psychiatric symptoms, higher income, being employed, having higher levels of education, and social supports, as well as having personal and social stability (Greenfield et al. 2007). However, many of these predictors vary by gender and have been found to be associated with women s retention in substance abuse treatment (Greenfield et al. 2007). Features of effective treatment programs include access to childcare, family services and transportation (Green 2006); customized treatment plans (Freeman et al. 2011); individual counseling (Sun 2006); womenonly programs or groups (Grella and Greenwell 2004); and programs which address the needs of different sub-groups (Tsai et al. 2010). Summary. This research clearly demonstrates that female drunk drivers are a problem that is worthy of our attention and concern. In light of outdated research, renewed efforts to examine the life histories and experiences of this population, and their experiences in the criminal justice and treatment systems can do much to inform prevention efforts and to guide the development of effective programs and practices targeted towards the risks and needs of female offenders. 4 Introduction FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

255 2. OBJECTIVES AND METHODS The findings from TIRF s 2011 literature review on female drunk drivers revealed important gaps in knowledge about this problem. In particular, the research on their characteristics is quite dated and the profile of these women is not well understood. Hence, it is not known if women who are currently convicted of drunk driving possess the same characteristics as their counterparts who were studied more than 20 years ago, or if these women share similar or distinct profiles in relation to male drunk drivers today. There has also been limited research to examine the experiences of female drunk drivers who are processed and supervised in the criminal justice and treatment systems; systems which have undergone tremendous changes in terms of policies and programs in the past two decades. Perhaps of greatest concern is the paucity of research in relation to the types of strategies and interventions that are most effective with this population. These important gaps in knowledge are a pressing limitation to policymakers and practitioners alike because such gaps are a barrier to the development of effective practices and strategies to address and subsequently, reduce this problem. Approach. To further build the knowledge base and increase understanding of this problem, TIRF selected a qualitative approach to this research. This type of approach was most amenable to the exploration of this issue, and, in particular, to examining the dynamic features of the personal situations of female drunk drivers and their interactions in the criminal justice and treatment systems which are not easily measured or quantitated (Creswell 2013). Of importance, this work also builds upon TIRF s prior work on system improvements. Since 2000, TIRF has worked closely with criminal justice practitioners as part of a comprehensive U.S.-based research effort designed FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Objectives and Methods 5

256 to improve the effectiveness and efficiency of the impaired driving system for dealing with hard core or persistent drinking drivers (see Simpson et al and Williams et al for more information about hard core drinking drivers). The goal of this work was to examine priority problems that professionals face at each phase of the justice system and identify practical ways to address these problems. One of the key findings was that similar problems exist at all phases of the justice system and that fixing just one can have positive reverberations throughout it. This work has informed the development of a system improvements paradigm (Robertson et al. 2009). In this respect, system refers to the context in which strategies and countermeasures are implemented and delivered (e.g., goals of scheme, how processing of offenders occurs, levels of communication, information-sharing protocols) and structures or entities used to deliver these countermeasures to a designated target group (e.g., agencies/stakeholders involved in the delivery, the legal system, treatment setting). A priority recommendation emanating from this research emphasized the importance of an intimate understanding of the systems in which measures are implemented as a pre-requisite for successfully applying them to any target population (see Simpson & Robertson 2001; Robertson & Simpson 2002a,b; 2003a,b). A successful implementation strategy is based upon streamlined delivery of the countermeasure, communication, and cooperation among various stakeholders, well-designed information exchange strategies, and accountability among agencies as well as offenders (Robertson et al. 2009). As such, the information collected in this study about female drunk drivers and their experiences in the criminal justice and treatment systems has been investigated against this paradigm of system improvements and conclusions have been drawn and recommendations formulated accordingly. This was deemed essential to guide the development of programs and policies to best serve this population. Study design. An exploratory case study approach using multiple cases was identified as the most practical strategy to investigate this issue. Studies that involve a single case are vulnerable to external validity threats because the focus of the study is specific to the context of the selected site. An important goal for this study was to be able to generalize the findings so the use of a multiple case study approach enabled TIRF to reveal a broad cross-section of perspectives on this issue (Maxfield and Babbie 2006). 6 Objective and Methods FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

257 Goals and objectives. The goal of this study was to increase understanding of the experiences of convicted female drunk drivers in the context of their life history as well as their experiences in the criminal justice and treatment systems. The outcomes of this study can contribute to a broader foundation of knowledge that could inform the development of much needed research initiatives as well as prevention efforts and effective interventions tailored towards female drunk drivers. It is underscored that this study was designed as a hypothesis-generating exercise as opposed to a hypothesis-testing one. Hypotheses can be generated in exploratory research whereas empirical data are gathered to test existing hypotheses in confirmatory research. The current study and its applied methods were conducive to the generation of such hypotheses, rather than testing them. To truly test them (i.e., reject or accept them), more data are needed. The main objectives of the study were three-fold. First, it was designed to explore the life histories of convicted female drunk drivers and how their history was linked to their offending. Second, the study was structured to examine what were the subsequent experiences of women in the criminal justice and treatment systems. Finally, the third objective of this study was to investigate the observations of criminal justice and treatment professionals supervising not only the women participating in the interview focus groups, but a much larger sample of women. Observations were gathered as a separate measure to determine whether the perspectives of practitioners were consistent with those reported by women participating in the study. Collectively, these data can be used to enhance our knowledge of female drunk drivers. Site selection and sampling design. There were four sites that were included as cases in this study. Three of these sites represented one county each from three different states (Greene County, Missouri; San Joaquin County, California; and Ottawa County, Michigan) and the fourth site represented three counties selected from a single state (Dutchess, Warren and Westchester Counties in New York). A complete description of each of these cases is provided in Appendix A. These sites were carefully chosen to ensure that they were geographically diverse (sites represented north-eastern, central and south-western regions of the United States), and that these sites represented a balanced cross-section of rural and urban areas. Access to the sites was facilitated with the cooperation of judges, court administrators, and deputy commissioners and chiefs of probation in the selected sites. They FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Objectives and Methods 7

258 were provided with complete information about the purpose and nature of the study, the research objectives, the criteria for the site selection, and the criteria to guide the sampling of convicted female drunk drivers and practitioners who could provide observations as part of the request. Purposeful maximal sampling was employed to select sites because the intention was to include locations that could purposefully inform understanding of the female drunk drivers and to reveal different perspectives of the problem (Creswell 2013). In particular, two of these sites were selected because of the presence of DWI Academy Courts, meaning that these courts are leaders in relation to applying the principles of DWI Courts and are welldeveloped and specialized in dealing with impaired driving offenders. The third site was chosen because its DWI Court program was rather large and had recently been evaluated; the fourth site was selected because it has developed specialized programs for female DWI offenders. Each of these sites further included traditional probation and courts as well as a combination of specialized, gender-specific and traditional treatment programs and support groups. Hence, these sites were considered to be robust and comprehensive in relation to the topic, and each site was able to provide a good crosssection of female DWI offenders and experienced practitioners representing the criminal justice and treatment systems. Stratified purposeful sampling was utilized to select individuals to participate in the study and the sample was carefully chosen to ensure it would illustrate sub-groups and facilitate comparisons (Creswell 2013). Individual selection criteria of female DWI offenders included varying ages (younger, middle-aged and older) and offense histories (i.e., first and repeat offenders; offenders with just drunk driving offenses and those with other types of criminal and/or traffic offenses), and a diversity of life experiences and issues. It was equally important that the practitioners included in the study represented a crosssection of practitioners working in courts, probation, alcohol education, and treatment programs who were able to share their observations drawn from supervising and treating the women participating in the interview focus groups, as well as a larger population of women who were not included in the groups. Criteria used to select these participants included their varied roles and responsibilities (i.e., line staff and supervisors), their length of experience in the field (i.e., limited experience or considerable experience), 8 Objective and Methods FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

259 their varied levels of experience with female offenders, their involvement in different types of programming, and their availability. Development of guides and survey questions. Interview focus group guides were developed to structure discussion in both offender interview focus groups and key informant (practitioner) interviews, and to elicit relevant information regarding priority topics that had been identified in the literature review conducted previously in The guides for offenders included questions relating to: demographic information; history of substance use; factors leading up to the offense; experiences with police; experiences with the courts; experiences with probation/parole; experiences with assessment and/or treatment programs; and, re-integration into the community. Priority topics for key informant interviews with practitioners included questions regarding their observations based on experiences working with female drunk drivers generally in relation to their experiences with males, and their observations regarding supervision and treatment strategies or interventions that have resulted in positive responses and/or successful outcomes, and those that have not produced positive responses or successful outcomes. Strengths and challenges associated with current approaches and lessons learned were also part of the interviews. Drafts of the interview guides for offenders and practitioners were peerreviewed by an experienced researcher and an experienced probation officer to ensure the questions and language contained within them were appropriate, sensitive, and clear to the intended audiences. Due to time limitations, it was not possible to pilot test the guides. The survey of female drunk drivers contained 64 items to explore a range of factors related to personal history, history with substance use, the DWI offense, characteristics of females at the time of the DWI offense, experiences with the criminal justice system, probation, and treatment, and re-integration into the community. This survey was developed based upon the interview focus group guide for offenders. Subject recruitment. Female drunk drivers were invited to participate in the interview focus groups by the criminal justice practitioner responsible for their supervision (typically a probation officer). When the invitation was extended, participants were informed that participation was strictly voluntary. Women were offered the opportunity to attend a focus group in lieu of a regularly scheduled meeting with the practitioner so that they received something FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Objectives and Methods 9

260 of value in exchange for their participation. Women were also provided with information about the agency conducting the study and the funder of the project, the purpose and objectives of the research and the use of its results in the form of a report. It was underscored that all information shared during the focus groups would be confidential, anonymous and not directly attributed to any individual. Key informants were similarly invited to participate in the study by the judge, court administrator, chief probation officer, or deputy commissioner in each of the counties involved in the study. Data sources and collection methods. The multi-faceted research design used for this study included data collected from two main sources: 1) female offenders who had been convicted of drinking and driving and who were currently under supervision or had recently completed it; and, 2) key informants who worked in the criminal justice and treatment systems and were involved in supervising and treating female drunk drivers. In particular, the inclusion of the observations of practitioners permitted the collection of data involving a much larger population of female drunk drivers (beyond the sample of offenders in the study) and over a much longer timeframe. These observations also served as an independent measure of the experiences of women involved in the study and illustrated the convergence of the findings from these two different groups to strengthen the study. Moreover, the interviews with practitioners also permitted the collection of data in relation to lessons learned about female drunk drivers and the operation of the two systems from the experiences of practitioners. Data from female offenders was gathered using a variety of methods, including interview focus groups, individual in-depth interviews, and surveys. These methods were selected to facilitate the collection of data from a broader group of participants than would have been possible using just one method, and to accommodate the schedules of participants and control project costs. Data from key informants were collected using interview focus groups and phone interviews for the same reasons as noted above. Generally speaking, qualitative research tends to have a lower degree of validity than quantitative research due to the level of subjectivity and interpretation involved in data collection and analysis (Gray et al. 2007). However, triangulation is a strategy that can be used to strengthen confidence in the validity of measures that are being used (McDavid and Hawthorn 2006). No single method of data collection is, or can be, designed to capture all of the 10 Objective and Methods FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

261 factors associated with a problem, however, the use of multiple methods can provide more empirical information that can be accumulated and analyzed to identify any similarities, patterns, and common themes. Hence, the use of multiple data sources or measurement processes can reduce the uncertainty involved in the interpretation of results. A large portion of the data were collected from interview focus groups with first and repeat offenders to explore their attitudes, behaviors, characteristics, risks and needs, and pathways to offending as well as their experiences within the criminal justice and treatment systems. This was augmented by a survey of a separate group of offenders that covered the same material as the interview focus group guides. This survey questionnaire was administered to female offenders in California who wanted to participate in the study and were unable to attend the focus groups at the scheduled times. Interview focus groups were utilized as a primary method of data collection for several reasons. First it was important to gauge the interactions of the participants to gain additional information and to provide context for their experiences. Second, the interviewees had shared a similar experience (i.e., conviction for drinking and driving and subsequent supervision) and were cooperative with each other in light of the fact that they had had contact with other participants through their involvement in court and treatment programs. Third, there was concern that one-on-one interviews would create a less comfortable environment and participants may be less forthcoming. Finally, there was a strong desire to ensure that a sufficient number of women were included in the study, however, time on site to collect data was limited. Similarly, a combination of interview focus groups and one-on-one interviews were also utilized with practitioners to maximize participation and to accommodate their respective schedules. Data collection. There were a total of 15 interview focus groups which ranged in size from five to 15 participants. Each group lasted approximately two hours and involved two researchers. Information provided at recruitment was again re-stated at the beginning of each focus group to ensure that the women understood that their participation was strictly voluntary. It was emphasized that women were encouraged to share information that they were comfortable sharing, and that each of them was free to exit the group at any time. The benefits of their participation, including the opportunity to help improve the experiences of women who found themselves in a FEMALE DRUNK DRIVERS A QUALITATIVE STUDY 11 Objectives and Methods

262 similar situation in the future, and the opportunity to help prevent other women from drinking and driving were highlighted. A copy of the 2011 literature review report that was previously conducted was also available for participants to examine at the time the groups were conducted. Women were further informed that copies of the final study report would be made available to them through the requisite court or probation agency responsible for their supervision. Data were collected between July and November 2012 with cooperation from the National Center for DWI Courts 2 (NCDC) and the American Probation and Parole Association (APPA). Interview focus groups and interviews were conducted in four sites California (San Joaquin County), Michigan (Ottawa County), Missouri (Greene County), and New York (Dutchess, Warren, and Westchester Counties). Slightly more than half of the offenders who participated in this study were located in urban areas whereas slightly less than half represented rural jurisdictions. In contrast, key informant interviews almost equally represented urban and rural jurisdictions. The interview focus groups and interviews with practitioners in Missouri were conducted at a hotel that was located very near the courthouse and drinks and snacks were provided during the meeting. This location was selected because it was convenient for the women to reach, and taxi vouchers were also made available to participants who required them. Conversely, in California and New York the interview focus groups were held either at the courthouse (in the jury room) or at the probation department in private rooms. These locations were identified as preferred settings because they were known to the participants and accessible to them. Drinks and/or snacks were also provided at these meetings. The interviews conducted with Michigan participants were conducted by phone due to cost. While several practitioners were involved in the study, only a small number of women from this site ultimately participated. It is believed that this was because they were more reluctant to provide interviews one-on-one by phone. 2 Specialty problem-solving courts such as DWI Courts and Drug Courts are more widespread in the United States than in Canada. For more information about these courts please see the National Association of Drug Court Professionals and the Nation DWI Court Center at www. nadcp.org and 12 Objective and Methods FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

263 The use of a recording device and associated software to capture and analyze discussion in each of the offender groups had been considered as part of the original study design, however, this was not pursued in order to protect and preserve the anonymity of the participants. In addition, frontline practitioners and researchers were in agreement that this would create a potentially uncomfortable environment and may discourage participants from disclosing information or speaking freely about their life history and experiences in the criminal justice and treatment systems. It was believed that this would ultimately impede the research study, influence interactions among participants, and detract from the quality and amount of data that was gathered. Two researchers were present in each of these interview focus groups and they each recorded their own notes to document discussion. Throughout the course of the focus group interviews, researchers were also careful to ensure that all of the participants were provided with the opportunity to speak, that discussions were constructive, and that individual participants did not dominate the discussion. Depending on the scheduling of the groups, researchers reviewed their notes immediately following each interview focus group or at the first available time to incorporate additional information and clarify details gathered during each session. In-depth interviews with female drunk drivers were conducted primarily by phone with one being conducted in-person. The interviews were conducted by one researcher and lasted approximately one hour. The interviews involved four offenders (two from Michigan and two from California). The anonymous survey questionnaire was distributed to female drunk drivers by their treatment provider in California because these women wanted to participate in the study but were unable to attend the focus groups. The surveys were collected by the treatment provider and returned to TIRF. The questionnaire took approximately 30 minutes to complete and participation was voluntary. In-person key informant interviews were conducted by two researchers and lasted approximately two hours, whereas phone interviews with practitioners involved only one researcher and lasted approximately 90 minutes. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY 13 Objectives and Methods

264 Study participants. In total, there were 154 convicted female drunk driving offenders (both first and repeat) who participated in the interview focus groups. There were four additional in-depth individual interviews with female drunk drivers also drawn from these sites. Focus group participants ranged in age from late teens to women in their mid-60s. Approximately one-third of the participants were first offenders; the balance were repeat offenders and the number of prior impaired driving offenses reported by them varied from one to seven with the average being two or three. Approximately half of the participants had been processed in traditional courts and the balance of them represented DWI Court participants or graduates. A select survey of 28 female drunk drivers in California was also included in the study. These surveys were returned to TIRF (21 from the El Concilio Drinking Driver Program and seven from His Way Recovery Home both located in California). It should be noted that these survey results are meant to provide a snapshot as opposed to a representative sample of all female drunk drivers. Furthermore, every question asked was not answered by all survey respondents. Therefore, results cannot be generalized and are shared to provide some additional insight into this issue. The key informant interviews involved a total of 36 individuals representing judges (3), defense attorneys (2), probation officers (24), alcohol education providers (3), and treatment counselors (4). The practitioners included in this study represented frontline professionals as well as managers and supervisors. While some practitioners had just two or three years of experience in their field, the majority of them had between ten and 30 years of experience. Data analysis. Following the conclusion of data collection at each site, researchers independently reviewed their notes to begin coding the data using both pre-figured codes that were comprised of data that researchers expected to find based upon the literature, and also emergent codes based upon surprising information that researchers did not expect to find, and information that was conceptually interesting or unusual (Creswell 2013). Summaries were independently produced by each researcher for each site, and then the summaries were compared. Once data collection was completed at each of the four sites, the summarized and coded data was analyzed to identify key themes, issues, and patterns in the data. Overall, 14 Objective and Methods FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

265 the notes produced by each researcher were highly consistent and contained similar themes, information, and observations. The data that are reported in the results sections are estimates that were developed based upon the detailed notes gathered by both researchers reflecting discussion in each of the focus groups. The results from the interview focus groups, survey 3, and key informant interviews with practitioners were synthesized and organized according to themes and priority issues described in the literature, and those that were unexpected or conceptually interesting. This information was used to identify lessons learned and to formulate recommendations to improve the supervision of these female offenders and the delivery of services to them in the criminal justice and treatment systems. The results that are reported are based upon a cross-case comparison because the larger purpose of this study was to provide a more holistic perspective of the issue so that generalizations could be drawn. The cross-case comparison revealed categories and themes emerging from each case study were highly similar across all four study sites; only very minor differences were evident in the composition of participants included in each case (e.g., slight differences in urban/rural participation, socio-economic status of participants, or offense histories) and these minor differences are noted where appropriate, however, they did not add any additional insight into this issue. Within-case comparisons were highly similar also and are omitted to avoid repetition. Peer-review. Earlier drafts of this report have been peer-reviewed by several knowledgeable practitioners who offered their feedback and input on the content of the report, the findings of the study, and the recommendations put forward. Of interest, the information obtained from the interview focus groups with offenders and key informant interviews with practitioners will be available to APPA to inform the development of a training module on female drunk drivers for probation and court practitioners, and also to NCDC to examine opportunities within the DWI Court model to better address female drunk driver issues. 3 Frequencies of response categories for the 28 surveys were calculated where possible (i.e., for the questions that received a high response rate). FEMALE DRUNK DRIVERS A QUALITATIVE STUDY 15 Objectives and Methods

266 3. FEMALE DRUNK DRIVER RESULTS More than 150 first and repeat female drunk drivers in four states (California, Michigan, Missouri, and New York) participated in 15 focus groups that ranged in size from five to 15 participants each. There were also an additional four in-depth interviews conducted and a select survey of 28 female drunk drivers from California. All of the participants either had previously been or were currently involved in the criminal justice system and had been processed in traditional courts, DWI Courts, or both. All of these participants were informed that their participation was strictly voluntary and that they were free to exit the group or interview at any time. Women were not asked to provide their names or any identifying information. They were also informed that any information they chose to disclose during discussions would be anonymous and not shared with any persons involved in their court case or supervision. Interview focus group discussion was structured according to a discussion guide that was prepared for this purpose. A range of topics were addressed in each group, including: demographic information, history of substance use and mental health, factors leading up to the offense, experiences with police, experiences with the courts, experiences with probation/parole, experiences with assessment and/or treatment programs, and re-integration into the community. Results of the interview focus groups, interviews, and survey are summarized below according to the following topics: demographic information; general profiles of offenders; substance use; mental health; life situation prior to offense; experiences with the criminal justice system; and, experiences with the treatment system. While many of these findings were consistent 16 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

267 across states, some differences were noted and these are highlighted where appropriate. The data that are reported in this section are estimates that were developed based upon the detailed notes of the researchers who conducted the focus groups and a comparison of their notes reflecting discussion in each of them. 3.1 Demographic information Age The female offenders who participated in the interview focus groups, interviews, and survey ranged in age from late teens to mid-60s, suggesting that women of all ages are involved in impaired driving. A majority of the participants were between an estimated 20 and 40 years of age. For example, the median age of survey respondents was 33. The number of college-aged young women present in each of the groups was higher than expected and accounted for perhaps one-quarter of participants. Similarly, there was also a not insignificant proportion of women older than age 40 in each age group, and a number of women acknowledged that they had grandchildren of varying ages. Survey responses reflected this trend as the results indicate that the age range for a first DWI charge was 18 to Education and employment With regard to level of educational achievement, it is estimated that more than three-quarters of the study participants reported having completed high school or their General Equivalency Diploma (GED). At least one-third of these women also reported having initiated and/or completed some type of post-secondary education to obtain a professional degree, license, or certificate. The majority of survey respondents (24 individuals) indicated that they had either finished high school or attained some level of post-secondary education. Only four respondents had not finished high school. More than half of the study participants reported that they were employed and common occupations that were represented included nurses, dental assistants, other health professionals, paralegals, administrative staff, casino staff, teachers, corporate employees, real estate agents, childcare providers, self-employed entrepreneurs, and bartenders. Less than one-quarter of women who participated in the study represented either trade professions (e.g., roofer, construction worker, welder, railway mechanic) or military FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 17

268 personnel. Of the survey respondents who answered the question about employment, 11 indicated that they had full-time jobs, eight had part-time employment, and six were unemployed. At least one-third of women reported that they had worked in bars and restaurants, either currently or previously, and this often facilitated a lifestyle of increased alcohol consumption, socializing, and driving after drinking, usually at the end of a shift. The women in the study who lacked employment generally found it difficult to be hired as a result of their prior arrests generally, and/or their felony convictions in particular. Almost all of the participants reported that gaining and/or maintaining employment was often challenging because of the scheduling demands placed on them as part of their sentence (e.g., random testing for alcohol and drugs, treatment or counseling, and court or probation appointments during regular business hours). These challenges were further compounded by a lack of available transportation options Marital and family status The lack of stable and supportive relationships among women was a common characteristic across the focus groups, interviews, and survey. It was estimated that more than one-half of women were single, separated, or divorced at the time of the study, and approximately one-quarter of women were currently in a relationship. Of those involved in a relationship, the majority of women reported having a partner or spouse who drank frequently and/or had a drinking issue whereas a minority of women reported having a sober, healthy relationship. Almost all college-age participants were single, although many of these women reported having prior relationships, and many acknowledged that drinking was a factor in those relationships. Most often, the relationship ended when the woman was either arrested for impaired driving, or made a decision to enter treatment. A history of unhealthy relationships was reported as common among the majority of study participants in their mid-20s to mid-40s. At least one-third of the women reported getting married and having children in their late teens or early 20s. A very small percentage of women reported having a stable or long-term relationship, and among those with this type of relationship, most 18 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

269 often the woman s drinking was a contributing factor to discord and periods of instability in the relationship. It is estimated that more than three-quarters of the women participating in the study had children ranging in age from toddlers to adult children, and approximately one-fifth of the participants reported having grandchildren. Although a significant majority of participants reported having custody of their children at the time of the interview focus group, an estimated onequarter of these women acknowledged having temporarily lost custody of their child(ren) for some period of time. In almost all cases, the loss of custody was a result of their heavy drinking and/or an impaired driving arrest, particularly for those women who had their child(ren) in the vehicle at the time of their arrest Family history and environment The reported family history of women who participated in the study varied considerably. It is estimated that slightly more than half of women reported a history of dysfunctional family relationships combined with prevalent alcohol and drug use and/or abuse to varying degrees. To illustrate, one 25-year old participant who had been sentenced for her fifth DWI offense reported I did my first line with my dad. More than half (17) of survey respondents indicated that there was a history of alcohol abuse in their family. Among these women, reports of coming from homes with unhealthy relationships between parents, divorced or single-parent homes, or having been raised by relatives for a period of time were common. While specific questions regarding a history of trauma were not explicitly asked, stories that were shared by more than half of participants spontaneously suggested or referred directly to a history of some type of abuse (e.g. emotional, physical, or sexual). They also often acknowledged having siblings who also had problems with alcohol or drug use, and family members (particularly male family members) who also engaged in and/or had been arrested for impaired driving. At least two focus group participants reported that their fathers had been drinking and driving for more than 20 years but had never been caught. In one case, a participant acknowledged that she never thought impaired driving was a big deal because she observed her father do it on a regular basis. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 19

270 In contrast, it is estimated that slightly less than half of the study participants reported coming from a relatively stable home environment with no recognized history of alcohol or drug abuse. A common denominator among these women was that they often reported no awareness of a relative (e.g., aunt/uncle or grandparent) suffering from alcohol problems until after they had been arrested for impaired driving. To this end, many of these women suggested that having been made aware earlier of the history of alcohol problems in their family would have enabled them to consider their own drinking behavior more objectively as a problem instead of dismissing it as normal. Teresa grew up in a traditional home where both of her parents drank, but were not heavy drinkers. She had her 1 st DWI conviction in her early 20s while in college and her 2 nd in her early 30s. Her 2 nd offense was a felony because her children were in the vehicle. She admits she only had them with her because she had been drinking less than her husband, who was an abusive alcoholic. The seriousness of this offense ultimately motivated her to end the relationship. She was sentenced in Drug Court and notes that the stigma associated with being an alcoholic was devastating. Her experiences in the criminal justice and treatment systems have been generally positive but frustrating at times, particularly because the conditions of supervision and treatment made it challenging for her to juggle her other responsibilities to family and her employer. Looking back, she admits that assessment and treatment were essential to help her admit her problem and to be compliant with the conditions of her sentence. Teresa said that more information about alcohol, normal consumption, and impaired driving laws earlier on would have made her more aware of the dangers of drinking and driving and perhaps helped her to avoid her first offense Physical health An estimated one-third of the study participants reported experiencing either past or current physical health issues. These included eating disorders and weight issues, gastric bypass surgery, cancer, fibromyalgia, diabetes, high blood pressure, and high cholesterol. In addition, it is estimated that approximately 10% of participants had undergone surgeries as a result of a car crash, often caused by drinking and driving. Many of the women in the 20 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

271 focus groups indicated that they suffered from insomnia or had a difficult time sleeping due to anxiety or emotional issues. Often they would consume alcohol as a way to self-medicate in an attempt to fall asleep Criminal history It is estimated that less than 20% of participants reported prior involvement in other criminal activities in addition to their impaired driving arrest(s). Most often, their involvement in criminal activities was linked to an existing relationship with a male partner, or a group of friends engaged in criminal activity. Almost half of survey respondents reported that they had family members who had been arrested for criminal offenses other than DWI or friends who had involvement with the criminal justice system. However, only three women indicated that they had been arrested for a non-dwi related offense. Focus group participants most often indicated involvement in offenses related to drug manufacturing and distribution, theft, fraud, or writing bad checks. 3.2 General profiles of female drunk drivers An examination of the characteristics of female drunk driving offenders in the study, revealed a number of commonalities among the majority of participants. First, a significant majority of women reported the presence of a life stressor immediately prior to their drinking and driving offense (e.g., death of a parent, spouse or sibling, serious illness of family member, fight with spouse, end of relationship, job loss, and/or financial problems). Second, it is estimated that more than half of women either specifically acknowledged or alluded to a history of trauma, although the nature and extent of it varied considerably. In addition, more than half of the study participants reported drinking and driving regularly, with some engaging in this behavior for more than a decade before their first arrest. Almost all of these women acknowledged and accepted personal responsibility for their behavior and the situation resulting from it (in some instances, to their own detriment as they continued to blame themselves and dwell on their actions which further lowered their self-esteem). This is in sharp contrast to male offenders who often deny responsibility or minimize the seriousness of their behavior. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 21

272 Three distinct profiles of female drunk drivers also emerged, and it is estimated that more than three-quarters of the study participants matched one of these three profiles which are described in more detail below. > Young women, often college-aged, involved in socializing and social activities; > Recently married women with young children; and, > Divorced older women, empty nesters, or women who had a parent with a debilitating health problem or who had recently died Young women It is estimated that at least one-quarter of the study participants were women under the age of 25, some of whom had accumulated multiple impaired driving offenses in a rather short period of time. In fact, one participant had served one year in prison following her fourth offense at the age of 24. These young women reported that they did most of their drinking in bars or at house parties and that they had attempted to drive home from those locations when they were arrested. They often reported drinking to relax, to feel comfortable, or to fit in in social settings. Moreover, many of them reported that they felt pressure to keep up with male friends or boyfriends in terms of the amount of alcohol that they consumed. Young women who had grown up in a stable home environment also reported drinking in order to cope with the high expectations of family members and what they perceived as the pressure to succeed. Daily alcohol consumption and binge drinking was not uncommon among this subgroup and this is consistent with research findings identifying binge drinking among college-age women as a phenomenon of growing concern (CDC 2013). These women tended to be single or had a partner who also drank heavily and facilitated and/or encouraged their use of alcohol Recently married women with children This group of female drunk driving offenders reported that their drinking did not become a problem or take off until after their children were born. In some instances, these women suffered symptoms of postpartum depression and drank as a coping mechanism or as a result of feelings of isolation and loneliness. Much of the alcohol consumption occurred with family or friends at home (e.g., they would drink while they did household chores, while on 22 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

273 the phone, or with friends or their partner). If a spouse was present, more often than not, they would also drink heavily which in some cases led to incidences of domestic violence. Of note, most of the women who fit into this profile stated that they did not have a drinking problem prior to entering into the relationship with the partner who abused alcohol and/or prior to the birth of their children. The circumstances that led up to the arrest of these women were often characterized by running errands close to the home such as picking up their children from school, buying groceries, or going to get gas. Many of these women were convicted of felony 4 impaired driving offenses on account of their children being passengers in the vehicle at the time of their arrest (this was especially common in New York due to the passage of Leandra s Law 5 ). While a majority of the women acknowledged that they were aware that they should not be driving after drinking with their children in the vehicle, it was often perceived as the only or the safest option (e.g., they were the more sober partner or childcare was not available). Kathy has been surrounded by substance use her entire life. Her mother drank heavily after her father left and her sister abused alcohol and drugs. Although she admits to repeatedly driving impaired when she was younger, she was able to get away with it in the past and didn t receive her 1 st DWI until age 48. Leading up to her offense, she suffered from depression and anxiety attacks as a result of having a child and a husband with health problems. Although she was able to remain sober for lengthy periods of time, many of her relationships were negative due to a love of chaos and excitement. Kathy was sentenced in Drug Court and noted that successful graduation is possible if a participant becomes committed to recovery. Kathy is now in the process of attending school to be an addictions counselor. For her, the hardest part of sobriety is people not understanding why you don t want a drink. 4 In the United States, criminal offenses are categorized as misdemeanor and felony offenses. In Canada, offenses are categorized as summary conviction and indictable offenses respectively. 5 Leandra s Law was passed in This law made any DWI conviction where a child 16 years of age or under was present in the vehicle at the time of the arrest a felony. This law also provided for mandatory ignition interlocks for a minimum period of six months for all misdemeanor and felony DWI convictions. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 23

274 3.2.3 Divorced older women and/or empty nesters Women who were not convicted of drunk driving until later in life typically reported that they developed a drinking problem in their late 30s or early 40s. Catalysts for their drinking included divorce or failed long-term relationships, shared custody arrangements or grown children leaving home, or parental illness/death. These women most often drank at home when they were alone and reported depression or feelings of isolation. Some of these women also reported drinking to feel comfortable in social settings, such as bars, because it had been a very long time since they had engaged in social activities of this nature. In particular, the women who fit this profile reported that they had more intense feelings of embarrassment and shame as their children were old enough to appreciate the stigma associated with their offending behavior, and in some cases, were also called to bail them out of jail following the arrest. While it is estimated that a small minority of participants did not fit into one of these three profiles, a majority of them possessed many of the characteristics frequently reported in the scientific literature including failed relationships, mental health problems, history of alcohol abuse within the family, multiple impaired driving arrests, history of trauma, and feelings of shame, guilt, and embarrassment. 3.3 Substance use The reported extent of substance use varied substantially across study participants. It is estimated that almost one-half of women reported early onset of drinking with many experimenting with alcohol and/or drugs in their early or mid-teen years; the lowest reported age of onset drinking was nine years old. In many cases, they indicated that their first exposure to alcohol and drugs was either in their own home, with relatives, or with friends. Conversely, it is estimated that between one-quarter and one-third of women did not begin to regularly use or develop a problem with alcohol or drugs, or begin to drive after using these substances, until they were in their 30s or 40s. Three common themes emerged among these women, regardless of the age at which they began to lose control of their substance use. First, many reported that their drinking did not become problematic until they became involved with a boyfriend or spouse who drank or used drugs more regularly 24 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

275 or extensively than they did. A number of women acknowledged that it was at this point in their life when their drinking and/or drug use changed in that it became more frequent, and/or involved the consumption of larger quantities of alcohol or other substances. Second, a majority of women stated that they were not aware that their substance use was more pronounced in relation to normal use and that they did not recognize that their level of use was problematic until an intervening event (e.g., an arrest) occurred. As one woman remarked, No one ever talked to me about addiction. Third, it is estimated that more than three-quarters of study participants used alcohol in combination with prescription drugs for mental health issues that were either medically diagnosed or undiagnosed of which anxiety and depression were reported as being the most common. It is estimated that study participants equally reported patterns of daily drinking or binge drinking. Approximately one-quarter of participants reported drinking heavily for a brief period which was followed by an extended period of sobriety that could last several months. A universal theme that emerged in all of the focus groups was that women reported that they drank for emotional reasons, or that alcohol consumption was a coping mechanism to help them manage their emotions and stress. To illustrate, one participant noted that Men drink for social reasons; women drink for emotional ones. In terms of their typical drinking scenario, more younger and middle-aged women reported that they were social drinkers in that they would do their drinking at a bar or restaurant or at the home of a friend or family member. These women reported that they typically engaged in binge drinking on weekends, and/or during holidays or celebrations. It was not uncommon for these women to report drinking to feel comfortable or to fit in. Conversely, some middle-aged and older women reported drinking at home alone and drinking on a daily basis. Among all of these women, it was not uncommon for them to report blackouts as a result of their drinking, and a minority of women recalled coming to during their arrest or waking up in jail or the hospital and having no recollection of how they arrived at that location. Those women who had been arrested multiple times for impaired driving or had previously entered treatment programs aptly described the challenges associated with maintaining sobriety, noting that not only is alcohol easily FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 25

276 accessible, but that Drinking is just so acceptable; it s easier to be a recovering addict. These women were also more likely to report learning to hide their drinking from their family or spouse. In terms of drug use, less than one-third of the participants reported use of illicit substances. Among many of these women, marijuana and methamphetamines were the most common drugs of choice, although use of cocaine, hashish, and ecstasy was also reported. Often the drug use was connected to the presence of a partner or spouse who also used drugs. There was also a very small minority of focus group participants who reported that alcohol was not their drug of choice and that they did not have a problem with alcohol. 3.4 Mental health Mental health issues were also frequently reported among this study population. It is estimated that three-quarters of the study participants reported using one or more prescription medications for disorders such as anxiety, depression, PTSD, bi-polar disorder, and schizophrenia. A small number of participants acknowledged sexual assaults or abortions as influencing their mental state, and some also indicated prior suicide attempts. Ten survey respondents indicated that they had past experiences with trauma of an emotional/psychological, physical, or sexual nature. There was also a minority of women who reported having mental health issues but not being on medication due to cost. This was succinctly described by one participant who stated, I can t afford to be depressed; I don t have insurance. Among those women with prescriptions for mental health issues, there was universal agreement that prescriptions were extremely easy to obtain and a large majority of them agreed that the doctor who provided them with the prescription never inquired about their alcohol use or informed them that they should not use alcohol in conjunction with the medication, although several women acknowledged that this information was printed clearly on the box of medication. 3.5 Life situation prior to offense More than three-quarters of the women that participated in the study reported that their impaired driving arrest was precipitated by a major life stressor such as a domestic argument, the end of a relationship, the loss of a job or child custody, the illness or death of a parent or other family member, 26 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

277 or financial problems. These women frequently associated this event as contributing to their arrest for drinking and driving, reporting that, Men don t stuff stuff in emotionally; women do, meaning that women may be more likely to suppress or ignore negative emotions instead of dealing with them. Feelings of low self-esteem, worthlessness, shame, guilt, failure, or loneliness were often reported as being the motivating factor that led to their alcohol consumption and subsequent DWI arrest. 3.6 Experiences in the criminal justice system Participants were asked about a broad range of issues pertaining to their experiences in the criminal justice system that covered their impaired driving arrest, their processing through the court system, and their experiences with supervision. Each of these is described in more detail below Arrest Of interest, almost all of the study participants reported that they were arrested within a few blocks of their residence and that they were either on their way home from an evening out or they were running an errand. An estimated one-quarter of women had their child in the vehicle at the time of their arrest, although it is estimated that more than half of the focus group participants reported drinking and driving with their child in the vehicle on one or more occasions. It is estimated that approximately one-quarter of participants were arrested as a result of a crash that ranged in severity from very minor to major. Nine survey respondents reported that they had been in a DWI-related crash. A small minority of women were involved in crashes resulting in serious injuries to themselves and/or others; between five and ten women acknowledged that they were responsible for the death of other vehicle occupants (either in their vehicle or the vehicle that they struck) or that one of their children had been killed as a result of their drunk driving crash. With regard to the reported BACs of study participants, the reported presence of high BACs ranging from.16 6 to.42 was not uncommon and many women acknowledged being well over the legal limit of.08. For example, in the New York interview focus groups, more than half of the women had BACs of.16 or higher. It is estimated that at least one-quarter of the women who 6 In most states, a BAC of.16 or greater is classified as a high-bac and is often considered an aggravating factor which can result in the imposition of more severe sanctions. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 27

278 participated in the study were unable to recall their arrest which may suggest BACs in higher ranges. Of interest, when participants did reference their BAC level, it was often included as an afterthought as opposed to a focal point of their story. In this regard, the women did not talk about their BAC in the same way as male offenders, and their BAC was generally a source of shame or embarrassment as opposed to pride. Overall, the study participants generally reported mixed arrest experiences. A not insignificant number of women reported what they perceived to be harsh treatment, particularly by female officers, although an equal number of women acknowledged that their behavior towards police officers was aggressive, resistant or abusive, and some of them reported being subjected to a Taser or additional charges, such as resisting arrest, or assault on a police officer as a result. The majority of survey respondents (24) classified their experience with police as calm/uneventful whereas the remainder described it as physical or scary. More than half of study participants reported spending time in jail immediately following their arrest. The majority of survey respondents (21) similarly reported that they were placed into custody as a result of their arrest. Almost all of the women found this to be a very frightening experience that was often coupled with the humiliation of having parents, siblings, partners, or adult children bail them out of jail. As one woman noted, It was not my proudest moment Court process It is estimated that a majority of study participants reported that the circumstances leading up to their impaired driving arrest were generally overlooked during the court process and that there appeared to be little consideration of how these circumstances may have contributed to the offending behavior. It was the perception of many participants that the focus of the system was almost exclusively on the offense without any regard for the circumstances that led up to the behavior. This was more often the case among participants in traditional courts, and less often among DWI Court participants. As one woman who was processed in traditional courts reported, following four impaired driving arrests in a relatively short span of time, Everyone looked at what I did but no one asked why. The perceived failure of the system to identify or acknowledge underlying factors that contributed to their impaired driving behavior (i.e., the pathways to offending) beyond 28 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

279 alcohol use was viewed as problematic by many participants. They often felt that their sentence or conditions of probation failed to take into account their life circumstances or address their issues. It is estimated that more than half of participants reported that they were represented by a court-appointed attorney during the court process, and at least one-third of them were represented by private attorneys. Almost all of the study participants acknowledged that they were more likely to successfully delay their case and/or ultimately avoid a conviction when they had a private attorney. One participant summarized this consensus stating, Having your own lawyer makes it a totally different experience. Approximately threequarters of women also reported significant delays in obtaining an assessment and in initiating the court case resulting from the arrest. It would often take an extended period of time (in some instances, months or even in excess of a year) to finally resolve the case. Women indicated that these long delays were a barrier to them moving forward and often referred to their experience with the court system as being in limbo which prevented the women from putting their offense behind them and focusing on the future Sentencing and supervision Study participants consistently identified a myriad of challenges associated with the sentence imposed for their impaired driving offense, and/or the period of court or probation supervision they received. These challenges and perceived barriers (described in more detail below) led to feelings of tremendous frustration among the female offenders. Negative effect of jail and prison. It is estimated that at least one-third of participants reported spending time in jail or prison as a result of their sentence for impaired driving, and/or for subsequent violations of court or probation supervision. The amount of time incarcerated generally ranged from a week to 30 days, although a not insignificant number of women reported spending between one and five years in prison for their offenses. With few exceptions, almost all of the women reported that jail and/or prison is a frightening experience that merely desensitized them, meaning that they shut down emotionally and became disconnected or lacked empathy for others. The exceptions reported that, at that time, jail or prison was the only thing that would get their attention and force them to stop drinking. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 29

280 Lack of information. Perhaps most notably, a majority of participants reported that information about the requirements of their supervision, the length of their supervision, their eligibility for certain privileges (e.g., a driver s license), or the potential consequences for non-compliance generally lacked clarity or were unclear. The women also indicated that the information they received often varied depending on the probation officer or judge to whom their case was assigned. It is estimated that more than half of the participants reported that they did not feel that they received adequate information to enable them to understand the process or what was expected of them, or that they received what they considered confusing or contradictory answers to questions. This is perhaps best illustrated by the fact that during the interview focus groups in each jurisdiction, women frequently asked other participants for information about different requirements or processes. The lack of clarity in relation to requirements or conditions became a source of great frustration for several participants who violated the terms of their supervision. They stated that they were not intentionally breaking the rules but instead were unaware that certain actions would lead to the imposition of sanctions. Quality of supervision. A large majority of study participants consistently agreed that the level of respect, communication, support, and encouragement provided by criminal justice practitioners (e.g., judges, case managers, probation officers) can make the difference between their success and failure. Women who reported being able to establish a rapport with the practitioner assigned to their case acknowledged that this was a source of motivation and important to their belief that they could successfully complete the requirements of their supervision and maintain sobriety. A significant majority of participants reported that when a practitioner treats them with respect, makes an effort to understand them and the circumstances of their case, and takes the time to listen it has very positive effects for their self-esteem and sense of self-worth and is a tremendous source of encouragement. For many of these women, this is often the first time that they have received positive feedback and support from individuals in their life and as one participant described it, A helpful and supportive probation officer makes you want to try harder; you don t want to disappoint them. At the same time, a majority of participants also underscored the importance of being held accountable throughout the course of their supervision, agreeing that, Probation officers are not counselors; they are there to hold 30 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

281 you accountable. A not insignificant number of women indicated that when they initially began their period of supervision, they tried to manipulate the system and made excuses to avoid doing what was required of them. In such cases, these women acknowledged that this only prolonged their period of supervision and/or resulted in additional offenses, and that their life only began to improve once they learned to adhere to the requirements of their supervision. At the same time, many women also emphasized the importance of fairness and balance in relation to the approach adopted by the practitioner, and the value of practitioners working with them to help them problem-solve and manage their risks to reduce the likelihood of re-offending. A minority of participants indicated that their experiences with supervision were mainly negative and that they were not able to develop a rapport with the practitioner supervising their case. In these instances, participants indicated they did not feel like they received enough information to be able to comply with the requirements of supervision and reported feeling unmotivated, discouraged, and depressed as a result of the process. One participant noted that, For a person to get well, they need to feel worthy; I never left my probation officer feeling good about myself. Finally, while an estimated one-quarter of the study participants noted that it was preferable to have a female practitioner supervise them, a larger proportion reported that it was not the gender of the practitioner that was an issue, but instead their ability to develop a good rapport with them. Some participants also stated that having a male probation officer who is supportive had a positive impact in their life because they had previously had only negative or abusive relationships with men. Cost. Almost all of the study participants indicated that the overall cost of their arrest and subsequent supervision were quite substantial, and a majority of them reported that, combined, the payment of lawyer fees, court and supervision costs, fees for services, and fines were overwhelming and difficult to handle. It was not uncommon for participants to report that there were often hidden or unexpected costs associated with their supervision requirements. To illustrate, more than a few women reported that they were surprised to find out they had to pay a fee in order to complete their community service requirement. The ongoing fees associated with ignition interlock devices and urinalysis were highlighted as particularly cumbersome. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 31

282 Inadequate services for random testing. It is estimated that more than half of the participants reported challenges specifically in relation to compliance with random testing requirements. In many of the jurisdictions, women reported that there was only one testing facility available, and this facility was typically only open during regular business hours and often could not be easily accessed using public transportation. According to participants, this made it quite difficult for them to reach the facility either before or after work. Moreover, they often reported that they had to have cash on hand to cover the costs of both testing and/or transportation in the event they were randomly selected. A minority of participants also reported challenges in identifying the reason why some of their test results came back as dilutes. 7 This was a source of concern given that this result could result in sanctions. A number of participants indicated that such results were due to diets and cleanses, and that, while some probation officers provided assistance to help them avoid such results, others were just told to figure it out. Additionally, a few older women reported considerable discomfort with the random testing process and being observed when providing a urine sample to probation officers. They noted that older women may experience bladder control issues and find it difficult and embarrassing to provide a sample for testing immediately while under observation. They indicated that more sensitivity in such situations would be helpful to mitigate this problem. Limited transportation options. An estimated three-quarter of study participants reported that finding or obtaining transportation, which was essential to their ability to complete the requirements of their sentence and supervision, was very challenging. They further noted that the additional costs for transportation (as a result of loss of licensure) were also substantial, particularly in rural jurisdictions where public transportation options are limited. A few women reported having to sell their house and move into town in order to access limited public transportation because they were unable to afford the costs of taxis, and they could not rely upon family members, friends, or neighbors for rides. 7 Dilution is the process of reducing the concentration of drug or drug metabolites in a urine sample. This result may occur intentionally or unintentionally, although sanctions may be imposed on offenders for such a result, particularly if it is a repeated occurrence. 32 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

283 Many women also acknowledged that they were forced to take public transportation late in the evening when it was dark, or to accept rides from persons they did not know well just in order to attend urine testing, treatment programs, or Alcoholics Anonymous (AA) meetings. In this regard, participants were very sensitive to the potential risk to personal safety that was associated with these situations. It is estimated that at least one-quarter of study participants reported that they found it difficult to maintain family relationships, particularly with those who lived outside of their county, because of limited transportation options in conjunction with restrictions resulting from their supervision (i.e., they required permission from their probation officer to travel outside of their county or state). Lack of employment. While it is estimated that more than half of participants were employed at the time of this study, it was universally acknowledged that securing employment following their conviction for impaired driving was challenging, particularly if they had been convicted for a felony offense. Among those women in the study who lacked employment, they reported it was difficult to find a job as a result of the prior conviction(s) generally, and/or their felony conviction(s) in particular, and due to the necessity of finding employment in a location near their residence as a result of limited transportation options. In addition to their criminal record and limited transportation options, women reported that the difficulty associated with obtaining and maintaining regular employment was compounded by the scheduling demands placed on them as part of their court or probation supervision. In particular, women noted that most of their scheduled appointments (i.e., appointments with probation officers, random testing, treatment, and court appearances) occurred during regular business hours. As a result, even when women were able to secure employment, they found it difficult to retain their position due to the frequency with which they had to request time off of work. Moreover, the time off work they required was often reported as being two hours or more because they generally had to rely on public transportation to get to and from their appointment. As a consequence, many women were forced to resort to pink-collar professions in restaurants, bars, call centers, or casinos because the shift work was easier to manage in conjunction with their requirements of FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 33

284 supervision. However, the women expressed concern that these professions are not always the most conducive for the maintenance of sobriety. Limited support system. With regard to their current situation, between onehalf and two-thirds of study participants reported having little or no support system that they could rely upon following their impaired driving arrest to assist them in the completion of the requirements of their supervision. Of those survey respondents who answered the question about support systems, only 12 indicated that they had a lot of support. A minority of focus group participants identified family members, neighbors, or friends who were able to provide transportation, childcare, or financial support, and noted that a support system was essential to their ability to meet the requirements of supervision and also maintain employment. Most frequently these participants reported that the members of their support system were female, and a not insignificant number of participants identified other female drunk drivers as being an important part of their support network. A number of women identified a positive aspect of AA support meetings was that it enabled them to meet other women who were more understanding of their situation and who could potentially be relied upon to provide assistance. Of interest, among those participants involved in stable marital or commonlaw relationships almost none of these women reported that their spouse or partner provided significant support in terms of transportation, childcare, or helping them maintain sobriety. This is in sharp contrast to the case of male impaired driving offenders who often rely upon their female partner for such support. Change in lifestyle. It is estimated that a large majority of participants reported that it was challenging to completely change their lifestyle although this was often deemed essential to their ability to complete the requirements of supervision and/or maintain sobriety. Approximately 20% of participants specifically reported that they had to distance themselves from family members in order to maintain sobriety, and in some cases, this involved not permitting their own children to reside in and/or visit their residence. Many of these women also had to end relationships with partners who engaged in heavy drinking or drug use as continuing to live in this sort of environment would not allow them to maintain their sobriety. 34 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

285 Women almost universally acknowledged that it was extremely difficult to make and to adhere to these changes because they felt they lacked the tools, skills, and support system to do so. As a consequence, it was not uncommon for women to report returning to unhealthy relationships and environments, and this often resulted in relapse as well as subsequent drinking and driving arrests. 3.7 Experiences in the treatment system While almost all of the females participating in this study reported that treatment had a positive impact on their lives, they also identified significant gaps in the treatment system which are described in greater detail in this section. Overall, it is estimated that more than half of the participants acknowledged that, generally speaking, most people do not enter treatment when they are at their bottom and ready to stop drinking and to be serious about treatment. In other words, they have not yet reached the readiness for change stage in recovery and are often in denial about the severity or extent of their substance abuse and/or dependency. In this regard, they believe that access to services that match the needs of individuals provides a greater chance that such services will be effective. Furthermore, the women consistently agreed that a majority of female offenders require more time in treatment than men before they feel as though they are able to connect to others, to disclose their life histories, and to identify and understand personal issues that contribute to their drinking. Survey respondents identified the sharing of personal experiences among women as one of the more helpful aspects of treatment In addition, although AA is a support group and not a form of treatment, the majority of participants described their participation in AA in conjunction with their treatment experiences. As such, their experiences with AA are also described here for convenience. Insufficient and inadequate services. Most notably, more than 80% of study participants consistently reported insufficient services not only in relation to treatment for alcohol and other substance use, but also in relation to mental health services. The women repeatedly stated that the types of services available to them were often limited and failed to meet their needs; this was especially the case in rural jurisdictions. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 35

286 Moreover, the general lack of service openings frequently resulted in delays in enrolment and/or long waiting times to receive services. This was identified as problematic for a not insignificant number of participants who reported that they were unable to access these services when they were highly motivated to achieve sobriety. This could also result in delayed case dispositions or extended probation in instances where the successful completion of treatment was a condition of sentencing and/or supervision. In addition, mental health services were rarely offered or available to participants. For the limited number of participants who were able to access mental health services, they believed that these services were an essential component to their ability to manage their drinking and establish a stable, healthy, and positive lifestyle. It is estimated that almost all of the participants reported attending generic substance abuse treatment services or programming delivered in mixedgender group sessions at some point. This type of treatment program was the most commonly reported as available across jurisdictions, and this was even more pronounced in rural areas. It was generally agreed by participants that these generic services delivered to a broad cross-section of individuals including both males and females were often not helpful and, in many cases, were considered to be detrimental. Survey respondents identified mixedgender group therapy as one of the least helpful aspects of treatment. In relation to mixed-gender treatment services, study participants frequently reported that such services had little or no benefit. In particular, they felt as though they could not relate to the experiences of individuals with different substance dependency issues, levels of dependence, or offense histories. To illustrate, a woman with alcohol dependence resulting from postpartum depression reported difficulty attending group therapy sessions that are comprised of a wide spectrum of individuals with substance abuse issues such as long-term drug abusers. Women also generally agreed that it was uncomfortable being female in male-dominated treatment groups and that they were unable to address personal issues related to trauma (which is often intimately linked to substance abuse) in these settings. For these reasons, a majority of participants indicated that a tailored and female-specific approach was preferred and that they believed this approach increased the likelihood of positive and lasting treatment outcomes. 36 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

287 It is estimated that between one-third and one-half of participants were able to access female-only treatment services and/or women-only AA support groups at some point. All of these women reported higher levels of satisfaction with gender-sensitive treatment services and also female-only AA support groups. They generally agreed that female-only treatment groups fostered a supportive, understanding, and safe environment that enabled them to better relate according to common experiences such as a history of abuse, parental and family issues, and concerns about children. In addition, a majority of participants also reported a strong preference for individual counseling at the outset of their treatment regimen as it provides them with an opportunity to identify and explore issues connected to their substance use, to discuss their experiences in greater detail, and to open up without fear of being judged by others. Women reported that individual counseling equipped them to better engage in group treatment by helping them to understand their issues so they could begin to address them. Approximately half of participants reported that they had completed either outpatient or inpatient treatment programs. An estimated one-third of participants reported that they had, at some point in their drunk driving career, been able to access specific treatment services tailored to a drunk driver population or to those with alcohol use issues, or programs that utilized a gender-sensitive approach. These programs that target the risks and needs of either drunk drivers or female offenders were frequently reported as being more beneficial and effective. However, according to participants, the availability of these types of services is limited or has been reduced substantially in recent years due to reduced funding. While there was no distinct trend in terms of the type of treatment that worked for study participants (e.g., inpatient, outpatient, individual, or group therapy), the level of satisfaction that women reported in relation to the treatment they received appeared to be correlated with their perceptions of whether the treatment they received was delivered in a setting in which they felt comfortable to disclose their experiences and whether the services provided were specific to their individual needs. At least one-quarter of participants were neither Caucasian or African- American and self-identified as a range of other ethnic and cultural populations including Asian, Hispanic, Mexican, and Samoan. Among these participants, the lack of culturally-sensitive treatment services was not raised FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 37

288 as an issue, suggesting that this may not be perceived as an important barrier in comparison to some of the other issues that were identified. Other reported gaps in treatment services related to the limited hours during which programming was delivered, limited transportation options, and the lack of available childcare. It is estimated that more than half of focus group participants reported difficulty in attending treatment either because of limited transportation options, the need to take time off work to attend treatment during regular business hours, or because of the lack of childcare services offered (i.e., there was no one available to watch their children while they attended treatment). Survey respondents identified these three issues as substantial barriers to treatment and the majority (18) noted that no childcare or transportation was offered as part of the treatment they attended. A significant majority of women also agreed that the length of treatment programs is often insufficient to enable them to understand the nature of their substance use and, more importantly, to learn the tools and skills necessary to manage it. To illustrate, with regard to inpatient treatment services, one woman noted that 28 days is like putting a band-aid on a boo-boo; this is much bigger than that. It was estimated that up to one-half of participants relied upon AA support groups to supplement the limited availability of treatment services. Several women relied on AA as their primary source of treatment or therapy either due to a lack of services or because the services available did not meet their needs. Cost. The cost associated with treatment services was an important factor that determined the type of treatment that at least three-quarters of participants were able to access. In particular, cost often limited their ability to access services that best suited their needs, as did transportation options. More intensive and specific treatment programs are often costly and it is not uncommon for women to be forced to participate in a program that does not meet their needs because that is all they can afford. In addition, some women also reported having to select treatment options that were accessible in terms of transportation as opposed to services that best addressed their needs. Women further reported that such services are counterproductive and make it difficult for them to maintain sobriety. Participants noted that the availability of insurance coverage typically determines what treatment options are available to them, reporting that, when it comes to treatment services, you get what you pay for. The cheapest 38 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

289 option is normally the only option that many of these women can afford as approximately three-quarters of them do not have insurance. They further note that because the quality of services provided is so poor, the treatment itself has little or no value and/or benefit. They perceived their involvement in these types of programs as a waste of time. Conversely, those participants who reported having insurance coverage expressed frustration about the length of time they were retained in treatment as treatment agencies tended to keep them enrolled in programs or therapy longer than they felt was needed. One woman stated that If you don t have insurance you can t get into treatment, and if you have insurance you can t get out. 3.8 Emotional effects of their experience Undoubtedly, the most prevalent theme that emerged and was consistent across all of the study participants was that women often defined their experiences in both the criminal justice system and the treatment system in terms of the emotional effects. All study participants reported a pronounced sense of shame, humiliation, and embarrassment as a result of their offense, not only at the time of and immediately following their arrest, but also throughout the entire court process and subsequent supervision period. These feelings were pervasive not only in relation to their children and immediate family members, but also in relation to their neighbors, friends, and peers, and also the broader community which included organized groups and school associations. It is estimated that almost one-quarter of participants reported that their arrest was either published in newspapers or publicized on television and this was reported to be a very traumatic experience. Their public humiliation resulted in decreased self-esteem and, in some instances, contributed to increased substance use as a result of depression and anxiety. It is estimated that almost all of the study participants reported fear about the unknown associated with the process (i.e., not understanding the requirements they would have to complete, the conditions they would have to abide by, and how to successfully adhere to both). To illustrate, one woman claimed her experience in the criminal justice system was akin to being held hostage in terms of not knowing what was expected and not knowing when the process would come to an end. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 39

290 Study participants also reported feelings of anxiety, concern, and stress in relation to the effects of their arrest and involvement in the system on their children. Moreover, these emotions were also connected to their apprehensions about their ability to complete the multitude of requirements of supervision. A myriad of factors including a lack of information about the process, challenges associated with completing all of the requirements of supervision, lack of support systems, inability to afford the costs associated with supervision and treatment, lack of appropriate treatment options, and inadequate transportation options compounded the angst that these women experienced. A majority of participants also reported feelings of dread in relation to relapse (returning to substance use and the consequences associated with this), and to having to put themselves in risky situations by taking public transportation or cabs at night or taking rides from persons they did not know very well due to a lack of alternative transportation options. Other emotions that were commonly reported by study participants included frustration and anger. These emotions were mainly connected to the focus of the criminal justice system on their offense alone (as opposed to the circumstances leading up to it) and also with being forced to repeatedly discuss their offense, as well as not being able to move forward with their life because of the requirements associated with their supervision. Several women reported that they felt as though they were continually struggling to keep their head above water. An intense feeling of pressure was also indicated by study participants. This was most often in relation to sustaining sobriety, fulfilling parental roles, not disappointing family members, continuing to pay bills, and meeting the demands placed upon them by family and employers. A number of women referred to feeling as though they were a burden on family and friends because of their ongoing need for assistance with transportation and childcare. Several women also acknowledged feelings of depression and of being overwhelmed by the process. In many cases, they reported feeling, at some point, like they would not be able to succeed or reach the end of the supervision and requirements. Such feelings appeared to be more pronounced among women with no support system or transportation options, or those who had more intensive supervision arrangements as a result of their offense. When there was no perceived end in sight to supervision, some of the women 40 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

291 questioned whether success was even possible or if everything they did was an exercise in futility. This greatly reduced their motivation for self-betterment. Finally, almost all of the women involved in the interview focus groups acknowledged a tremendous sense of relief and were extremely thankful that they had not injured or killed someone as a result of their impaired driving behavior. This appeared to be an important focus of attention for women following their offense, suggesting that they had not recognized or considered the possibility of such consequences at the time of their behavior. In this regard, they demonstrated insight into the severity of their actions and also admitted that their situation could be much worse. 3.9 Summary The interview focus group, survey, and in-depth interview data collected from more than 150 first and repeat female drunk drivers revealed highly consistent findings across the four jurisdictions that participated in this study. Demographics. Interview focus group participants ranged in age from late teens to mid-60s, suggesting that women of all ages are involved in impaired driving. A majority of them had attained a high school education or its equivalency and approximately one-third reported having initiated or completed some type of post-secondary education. Most were employed in diverse occupations and approximately one-third of the women reported that they had previously or currently worked in bars and restaurants which often facilitated a lifestyle of increased alcohol consumption, socializing, and driving after drinking, usually at the end of a shift. Women who lacked employment generally found it difficult to be hired as a result of their criminal record. Some women came from stable home environments where there was no history of alcohol or drug misuse whereas others were the product of broken or abusive homes where substance use and abuse were prevalent and acceptable. A lack of stable and supportive relationships was another common characteristic among focus group participants. More than half of the women were single, separated, or divorced and the majority had children, some of which were in the vehicle at the time of their impaired driving arrest. Approximately one-third of the participants stated that they had some type of support network in the form of either family or friends that FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 41

292 enabled them to comply with the demands of parenting, employment and the requirements of supervision. A significant majority of women who participated in the interview focus groups reported that their impaired driving arrest was precipitated by a major life stressor such as a domestic argument, the end of a relationship, the loss of a job or child custody, or the illness or death of a parent or other family member. It is estimated that less than 20% of participants reported prior involvement in other criminal activities in addition to their impaired driving arrest(s). Typical offenses included drug manufacturing and distribution, theft, and/or fraud. Profiles. With regard to the characteristics of female drunk driving offenders, three different profiles of this population emerged: 1. Young women who drink in order to fit in and consume alcohol and/ or binge drink at house parties and bars; 2. Recently married women with children who drink following the birth of their children as a means for coping with loneliness; and, 3. Divorced older women and/or empty nesters who begin to drink later in life (after age 40) following a catalyst such as the death of a parent, end of a marriage, or departure of children. Many of the characteristics identified in the literature were common across all groups including trauma history, failed relationships, mental disorders, history of alcohol and/or drug misuse within the family, multiple impaired driving arrests, and feelings of shame, guilt, and embarrassment. Substance use and mental health. Less than one-third of participants reported use of illicit substances of which marijuana and methamphetamines were the most common drugs of choice. More than three-quarters of the focus group participants were on at least one or more prescription medications for disorders such as anxiety, depression, PTSD, and bi-polar disorder; there was agreement among the women that these prescriptions were extremely easy to obtain. There were also many instances of undiagnosed mental health issues and histories of trauma and/or abuse (both physical and sexual). Experiences in the criminal justice system. Overall, several focus group participants reported generally negative experiences with the justice system from the point of arrest through to completion of probation, although 42 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

293 there were exceptions. Many women emphasized the emotional reactions associated with their experiences in terms of shame, frustration, anger, depression, anxiety, uncertainty, and fear. In relation to their arrest, a large majority of focus group participants reported that they were arrested within a few blocks of their residence. It is estimated that approximately one-quarter of the women were arrested as a result of a crash, and a very small minority were involved in significant crashes. A number of women reported that their BAC was shockingly high and BACs ranging from.16 to.42 were not uncommon. Overall, study participants reported mixed arrest experiences but agreed that being placed in custody was a frightening experience. There were a number of concerns associated with the court process. It was the perception of many participants that the focus of the system was almost entirely on the offense without any regard for the circumstances that led up to the behavior, and this was perceived as problematic. Many women felt that their sentence or conditions of probation failed to take into account either their life circumstances or address their issues. Women also indicated that long delays in the court process prevented them from putting their offense behind them and focusing on the future. Challenges associated with sentencing and supervision practices were also highlighted. At least one-third of participants spent time in jail or prison and reported that it had a negative effect. A majority of participants reported that information about requirements and consequences was generally unclear, and a large majority of participants consistently agreed that the way that practitioners respond to women can make the difference between their success and failure. Almost all study participants indicated that the overall cost of their arrest and subsequent supervision were quite substantial and could be overwhelming. More than half of them reported challenges in relation to compliance with random testing requirements. Securing transportation was reported as very challenging by an estimated three-quarters of participants and this made it difficult to meet supervision, treatment, and testing conditions as well as maintain employment. Compounding this problem, between one-half and two-thirds of study participants reported having little or no support system to assist them. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 43

294 Experiences in the treatment system. While many of the female offenders in the focus groups reported that treatment had a positive impact on their lives, they also expressed a great deal of frustration. More than 80% of study participants consistently reported insufficient services not only in relation to treatment for alcohol and substance use, but also for mental health. The general lack of availability of services led to delays in program enrolment. Generic substance use treatment groups were not perceived as beneficial by many of the participants because they reported they were not able to relate to mixed groups that included offenders who have different substance use problems and offense histories. Subsequently, the tailored and femalespecific approach was cited as the most likely to produce positive and lasting outcomes because it fostered a supportive, understanding, and safe environment that enabled them to better relate and share their experiences. The women also expressed a strong preference for individual counseling at the outset of their treatment plan. Women also indicated that the availability of treatment services varies considerably and many women are forced to participate in interventions that are not suited to their needs as a function of cost or access. This was especially pronounced in rural jurisdictions. The quality of available treatment was also reported as a concern. The availability of insurance coverage often determines what treatment options are available to them. Emotional effects of their experience. All study participants reported a pronounced sense of shame, humiliation, and embarrassment as a result of their offense throughout the entire process. These feelings were pervasive in relation to their children, family, friends, neighbors, and communities which were exacerbated in one-quarter of the cases as a result of media coverage. The women further reported fear about all of the unknown variables and conditions associated with their involvement in the criminal justice and treatment systems. Additional emotions that participants identified included: > Anxiety, concern, and stress over how to discuss their arrest/conviction with their children; > Frustration due to a lack of information about the process, lack of support systems, inability to afford costs associated with supervision and treatment, lack of appropriate treatment options, and inadequate transportation; 44 Female Drunk Driver Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

295 > Dread in relation to the possibility of relapse; > Pressure to remain sober, meet all supervision conditions, meet the demands of family and employers, fulfill parental roles, and pay bills; > Depression due to a perceived inability to succeed as a result of being overwhelmed and lacking support; and, > Relief that they had not injured or killed someone as a result of their impaired driving behavior. Beth grew up in a small town. Drinking was common because there wasn t really much to do. Her father was an abusive drunk and many of her siblings also struggled with alcohol use. To escape, Beth joined the Air Force and was a soldier serving in combat overseas. Upon return from two tours, she struggled with PTSD. Her family did not understand and the estrangement from her siblings and nieces/nephews contributed to her depression and binge drinking. Beth subsequently accumulated four DWIs in a few short years and ultimately lost custody of her son who went to live with his father in another state. She knew that getting sober was the only way to get her son back. She has since regained custody of her son and is in a healthy relationship with a man who does not drink. She made the difficult decision to separate herself from family members because of their continued drinking which she acknowledges would have a negative impact on her life. She notes that she struggled in several treatment programs and credits her probation officer for helping her to find a Veterans Affairs group where she could get help for her drinking and PTSD. While thankful for her sobriety, she continues to be frustrated with her limited employment opportunities. A combination of being overqualified and being a felon has made her feel as though she cannot contribute to the society she once served. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Female Drunk Driver Results 45

296 4. PRACTITIONER INTERVIEW RESULTS Key informant interviews with practitioners in the criminal justice and treatment systems were conducted in four states - California (San Joaquin County), Michigan (Ottawa County), Missouri (Greene County), and New York (Dutchess, Warren, and Westchester Counties). These states were chosen to provide a geographically representative sample of the United States. Key informant interviews almost equally represented practitioners in urban and rural jurisdictions. Key informant interviews involved a total of 36 individuals representing judges (3), defense attorneys (2), probation officers (24), alcohol education providers (3), and treatment counselors (4). While a majority of these interviews were conducted in-person with two researchers present, a minority of these interviews took place via phone. Some of the interviews were conducted with just one key informant, whereas others took place in small groups of three to five practitioners and this was often a result of scheduling opportunities. For example, in New York, key informant interviews with probation officers were conducted in groups in each of the three counties included in the study. The practitioners included in this study represented frontline professionals as well as managers and supervisors. While some practitioners had just two or three years of experience in their field, the majority of them had between 10 and 30 years of experience. The practitioners involved represented both criminal justice professionals and alcohol education and treatment professionals. However, many of the themes that emerged from these interviews and many of the reported experiences 46 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

297 were highly similar across professions. As such, the results in this section are structured according to the reported characteristics of female offenders to whom these professionals provide services, their experiences in delivering supervision or treatment services to these women, and the lessons that they have learned as a result of their experiences. 4.1 General characteristics of female drunk drivers who are supervised or treated Arrest and crash characteristics It was agreed by practitioners who participated in this study that the arrest characteristics of female drunk drivers are, by and large, similar to those of male drunk drivers in terms of time of day and day of the week during which arrests are most likely to occur (i.e., evenings and weekends). However, many of them reported that women are much more likely than men to have children in the vehicle at the time of arrest. There was also considerable agreement among practitioners that Child Protective Services (CPS) was rarely called, and only became involved if there was no family member to take custody of the children in the event that the woman was placed in custody or admitted to inpatient treatment. This generally occurs because it is widely recognized among justice professionals that children are better accommodated within their family setting as opposed to being placed in the custody of the State. Practitioners reported differences in relation to the average BAC of women they encountered at the time of arrest. An estimated two-thirds of practitioners reported that women often have BACs equal to and also higher than men on average. Conversely, approximately one-third of them reported that the BACs among female drunk drivers are generally lower than those of men. However, a number of these practitioners also acknowledged that a not insignificant number of women are arrested with lower BACs, and this is generally believed to reflect the fact that women are more likely to combine alcohol with their prescription medication or other substances. Hence, they exhibit signs of impairment that may not be consistent with their BAC level. There were also reported diametrical differences in relation to crash involvement. Whereas approximately half of practitioners indicated that women were more likely than males to be involved in some type of crash (e.g., either minor or major) at the time of arrest, an equal portion also reported FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 47

298 that women were less likely to be involved in a crash. This is consistent with the focus group results which revealed that reports of crash involvement were more prevalent in some groups than in others. There were no apparent differences across jurisdictions. There were also reported differences among practitioners regarding the behavior of female drunk drivers at the time of arrest. Whereas some of them reported that women were more likely to be combative with arresting officers or during the booking process, others reported that this behavior was rare among women. Again, this is consistent with results from the focus groups where only some of the women reported aggressive behavior on their part or having blackouts and/or no recollection of the arrest. There were no apparent differences across jurisdictions. There was a high level of consensus among practitioners that drunk drivers often report a lack of consideration of the possibility of killing or injuring themselves or others when they get behind the wheel after drinking. However, after the arrest, almost all offenders have the realization that they could have easily killed or injured someone and express this fact, or experience intense and ongoing guilt because they did kill or injure someone. This is generally true both for female and for male drunk drivers Demographic characteristics Age. Practitioners consistently agreed that, in their experience, the majority of female drunk drivers entering the criminal justice system are between the ages of 21 and 45. They further noted that a smaller proportion of women who entered the system were either older or younger, and these findings are consistent with the literature (Shore and McCoy 1987). However, there was a high level of consensus that a growing number of younger, collegeaged women were entering the system for drunk driving than was previously the case, and that there is often more drama associated with the lives of younger women. They also indicated that the presence of young women who were pregnant at the time of arrest was becoming somewhat more common and that this was a concern, although this is not the case in every jurisdiction. Physical health. Practitioners reported that many women who entered the system were identified as having physical health issues, although the nature of these issues varied in accordance with the age of offenders. They noted that 48 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

299 younger women were more likely to suffer from eating disorders and other addiction issues whereas older women were more likely to indicate hormonal issues, early menopause, or significant illness. Of note, practitioners reported that insomnia and sleep disorders were common among a significant number of offenders as were undiagnosed medical issues, which is consistent with interview focus group findings. Of interest, in Michigan, multiple cases of female drunk drivers who had undergone gastric bypass surgery were reported. Marital status. All practitioners reported that the majority of female drunk drivers they supervise or treat are currently single with many of them coming out of unhealthy or abusive relationships. They also agreed that those women who were in relationships were most often involved with a partner who used and/or abused alcohol and/or drugs. This was frequently reported as being a factor in a female offender s substance use problem. As one practitioner noted Women desire independence but they do not like to be alone. Support network. Approximately half of the practitioners agreed that women were less likely than male drunk drivers to have a support network that they could rely upon to help them with transportation, childcare, and provide emotional support and encouragement in relation to their sobriety. The balance of practitioners reported that this was equally the case for females and males, or that males were less likely to have a support network. Education. A majority of practitioners agreed that, in their experience, female drunk drivers represent all levels of education and professions. They reported that most of these women have at least some high school education and many of them had either finished high school or obtained their GED. They further agreed that a minority of female drunk drivers have at least some college education. These findings are consistent with the literature (Peck et al. 2008). Of interest, it is estimated that at least half of the practitioners agreed that female drunk drivers were more likely to have some college education and a professional career relative to males. This is also consistent with the literature (Chalmers et al. 1993; Shore and McCoy 1987). Employment. With regard to employment, findings were mixed. Whereas approximately half of the practitioners reported that more women than men had low-paying jobs and were either minimally employed or looking for work, an equal portion reported that women were more likely to have a professional career, were more likely to be working than men, and were FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 49

300 better able to afford supervision and treatment costs. These mixed findings are also consistent with the literature (Chalmers et al. 1993; Shore and McCoy 1987). Insurance coverage. Practitioners reported that both female and male drunk drivers were equally likely to be uninsured, and this was noted as a significant problem in relation to the affordability of treatment services for all offenders. They further noted that, even for those women with insurance, rarely are all of the associated costs of treatment or health services covered by the policy, meaning that cost can also be an issue even for those who do have coverage. Moreover, some practitioners underscored that those women with insurance were unlikely to utilize their insurance coverage to pay for treatment services because they did not want to have to report their impaired driving offense for fear that their employer would find out and they would lose their job Substance use It is estimated that more than half of practitioners reported that they are encountering more female drunk drivers who are younger and they often have significant substance use problems at a young age. They agree that these younger women have an earlier age of onset of drinking that is comparable to that of men, and that they are consuming larger quantities of alcohol than has historically been the case. These findings are consistent with the literature (Popkin 1991; Brady and Randall 1999; Gudrais 2011). According to some practitioners, this is sometimes referred to as the Sex in the City Syndrome, meaning that these young women report frequently drinking in social settings with their friends (who often encourage their alcohol consumption) and that these women are less likely to stop drinking or say no to alcohol. In addition, young women generally have fewer responsibilities (i.e., children or full-time employment) and this facilitates drinking on a regular basis. In contrast to younger female drunk drivers, a majority of practitioners reported that, in their experience, older female drunk drivers are more likely to indicate that they drink at home alone, often as a result of depression. However, they acknowledge there is also a proportion of older women (i.e., age 40 and older) who are newly single and who go out with friends to bars in order to meet new people. Practitioners report that these women are more likely to drink heavily in these settings in order to relax and feel comfortable. 50 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

301 Practitioners also generally agreed a majority of female drunk drivers meet the clinical criteria for a diagnosis of alcohol abuse with many of them indicating early onset of dependence and a smaller proportion reporting adult onset. Among this population alcohol consumption is believed to be a natural segue to drugs (more often illicit). Subsequently, poly-substance use issues are reported by practitioners as being common among female drunk drivers. Interestingly, whereas some practitioners noted that women exhibit more poly-substance use than males with drinking coming first, others equally indicated that poly-substance use was more prevalent among males. Similarly, some practitioners reported that prescription drug use was more prevalent among women over age 35 and others noted that prescription drugs were common among all age groups. Some treatment professionals also reported that male drunk drivers are often likely to have prescription medication as well. Almost all of the practitioners acknowledged that it was fairly easy for female drunk drivers to obtain prescription medications from physicians or psychiatrists. It is estimated that at least half of the practitioners indicated that many female drunk drivers are likely to be in denial about the severity of their drinking or unaware of the extent of their alcohol use. According to interviews, many of these women view themselves as social drinkers, particularly if they are able to maintain employment and care for children and, as such, lack insight into their level of dependency. Practitioners generally agreed that female drunk drivers are often not aware of their limits in relation to drinking and that women are less likely than males to have a good understanding about how much they can drink. In particular, they noted that female first offenders are often quite shocked by their BAC reading, and the amount that they drank to reach this level. They reported that many female offenders commonly suggest that they had just one glass of wine. They also frequently acknowledged that women experience more shame in relation to their drinking and tend to minimize their alcohol use. However, practitioners reported that in alcohol education and treatment settings, females are more willing to disclose past use than male drunk drivers Mental health In the experience of practitioners there was agreement that mental health issues were very common among the female drunk driver population. They also agreed that in some cases these issues were diagnosed whereas in FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 51

302 others they were not yet formally identified by a mental health practitioner. The mental health issues most prevalent among these women according to practitioners included anxiety, depression, PTSD, and bi-polar disorder, and this finding is consistent with the literature (McMurran et al. 2011; LaPlante et al. 2008; Shaffer et al. 2007). By comparison they report that males are more likely to suffer from bi-polar disorder and anger management issues. Almost all of the practitioners reported that mental health issues are generally more common among older women, although a not insignificant number of younger women are likely to experience these issues. Many practitioners also acknowledged that a majority of female drunk drivers consume alcohol in order to help cope with emotional issues. Some practitioners suggested that this is because women in treatment are often of lower socio-economic status and cannot afford medication so they use alcohol to self-medicate. In this regard, women drunk drivers are reported by treatment providers as using alcohol and marijuana to cope with anxiety. Trauma. It is estimated that more than half of practitioners indicated that it is not uncommon for female drunk drivers to have experienced domestic violence at some point in their lives. They noted that, in some cases, it is following a domestic violence event in which the woman leaves the residence that she is arrested for drunk driving. In particular, treatment professionals report that past trauma (e.g. verbal, physical, or sexual abuse) is quite prevalent among female drunk drivers, with rates estimated to be as high as 80% in some jurisdictions. This trauma is often believed to be a consequence of familial relationships. They noted that many women develop low self-esteem as a result of these experiences and past trauma is often a trigger for substance use problems. Treatment professionals indicated that rape and incest, violent deaths and loss of siblings, abandonment issues, grief, and estrangement issues were not uncommon experiences among this offender population. Interestingly, at least one-third of practitioners also reported that past trauma was a common issue among male drunk drivers as well, but they noted that such issues are less likely to be identified or addressed for males as a result of their hesitancy to disclose and discuss abuse. 52 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

303 4.1.5 Other driving and criminal history It is estimated that half of the practitioners reported approximately 60% of female drunk drivers have other driving violations (e.g., speeding, careless driving, tailgating, driving while suspended or revoked) and that such offenses are as common among female drunk drivers as they are among males. They also generally agreed that a minority of female drunk drivers have other criminal offenses on their record and this is more common among repeat female drunk drivers. The most often reported offenses include drugs, domestic violence or assaults, shoplifting, and retail fraud. 4.2 Experiences of practitioners Changes in society. A number of practitioners who have worked in their field for several years reported that some general changes in society over time may have contributed to increased drinking and driving among women. Not only do more women (and younger women in particular) have greater access to vehicles, but also alcohol is more easily accessible today than it has been in the past. Moreover, it is agreed that social attitudes towards alcohol consumption have also changed and alcohol is often present and available at social gatherings, at celebrations, and at meals. As one practitioner noted, Alcohol is not socially unacceptable, so there are more social reasons to drink than to not drink. These practitioners also acknowledged that there is more societal breakdown and/or disintegration in relation to community and familial relationships. In many cases, there are more single-parent families, and there are greater pressures on families economically due to today s difficult financial climate. Some practitioners also identified an increased trend towards personal selfindulgence and the demand for and expectations of possessing luxuries and items that reflect social status. In stark contrast, practitioners reported that public attitudes towards drunk driving have become increasingly punitive and, despite research clearly demonstrating that substance use is prevalent among drunk drivers, public support for treatment and rehabilitation for this population is quite low. However, they did acknowledge that it is more socially acceptable for women to participate in treatment for substance abuse now as compared to the past. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 53

304 4.2.1 Criminal justice system Practitioners. With regard to their experiences working in the criminal justice system and how it has changed over time, practitioners agreed that police officers are more likely to charge all impaired drivers as compared to in the past when it was more common for drunk drivers generally, and women in particular, to be sent home or given a ride home as opposed to being arrested. They further noted that the presence of high-profile cases involving drunk drivers has increased concern about liability and this has contributed to changes in attitudes and law enforcement practices along with changing social norms. Similarly, practitioners across jurisdictions also reported that police officers are still likely to endeavor to make arrangements to ensure that children (who were present in the vehicle when a female driver was arrested) are placed in the care of an appropriate family member as opposed to involving CPS. This is often considered to be in the best interest of the children. There was agreement that the involvement of CPS tended to be only in those extreme cases when there was no one else who was able to take custody of the children. All practitioners reported that fewer diversion programs are available to drunk drivers today, whereas in the past such programs were common and frequently utilized. They noted that this change in practice has also likely contributed to more female drunk drivers becoming involved in the criminal justice system, especially since women are now more likely to be formally arrested and processed than they were in the past. There was considerable agreement among practitioners that more criminal justice professionals are now sensitive to and knowledgeable about issues of trauma and victimization, mental health issues, and self-esteem issues than they were previously. They also noted that younger practitioners are more likely to be familiar with these issues in relation to those who have worked in the system for more than two decades, and that they are more likely to be aware that these issues are relevant to both male and female offenders. Finally, there was also a high level of consensus among practitioners that there are more female police and probation officers, judges, and attorneys working in the system today than was the case historically. As a result of this increased level of professional equality, female offenders may be less likely to benefit from being female to obtain leniency. 54 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

305 In relation to perceptions about treatment availability, practitioners generally agreed that while treatment is more often available today than has been the case previously, they also noted that the quality and diversity of treatment services and the qualifications of treatment providers vary substantially. In all jurisdictions, it was reported that those treatment agencies that are eligible to accept insurance coverage are generally staffed with more qualified providers. Moreover, practitioners agreed that fewer specialized or comprehensive services are generally available to drunk drivers than was the case a decade ago. Case processing. Practitioners consistently agreed that their colleagues generally endeavor to supervise and treat male and female drunk drivers in an equal fashion using the same strategies. It is estimated that half of the practitioners reported that female drunk drivers are not treated differently than males and that women are offered the same types of pleas and receive the same sentences as men. They also noted that women are more likely to receive sentences involving a period of custody today than was the case in the past. Of interest, some practitioners also noted that in the case of jury trials, female jurors are more likely to be harder on female drunk drivers than they are on males. Conversely, the balance of practitioners indicated that although male and female drunk drivers are likely to be processed in the same way, courts still tend to be more lenient with females. It was noted that perceptions about women remains an influencing factor although to a lesser degree than has been the case historically. These practitioners also agreed that females were more likely to plea to a lesser charge or receive a suspended sentence, and acknowledged that this is often to avoid children being placed in the custody of the State. Some practitioners suggested that women are more often able to manipulate the system and that some female drunk drivers will try to use their children (e.g., bring them to court) in order to avoid more serious sanctions. Among practitioners, probation officers reported that one of the limitations of current practices in the justice system is that they often do not receive any information about the arrest of drunk drivers and, unless offenders are screened by the probation department, they often do not have any objective history of substance use and must rely solely on the information that offenders choose to disclose. Hence, this means that intake and initial FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 55

306 meetings with probationers can take longer because officers have to try to identify relevant issues before being able to determine the most appropriate strategies to manage them. A majority of practitioners agreed that the biggest barrier to participation in treatment for female drunk drivers is the fear of losing custody of their children, particularly if they are required to attend inpatient treatment and they are a single parent. They further noted that it was not uncommon for women to refuse to attend treatment or to try to postpone attending treatment for this reason. There was a high level of consensus among practitioners that female drunk drivers are most often required to participate in mixed-gender treatment in a group setting comprised of individuals with diverse backgrounds and histories of substance use. There was also considerable agreement that this is generally problematic for females not only because of the diversity of participants, but also because of the small numbers of females in these groups. As such, they report that females are often uncomfortable and unable to discuss or focus on issues of trauma, abuse, or relationships which are frequently contributing factors to their substance use because the group is predominantly male. It was further noted by many practitioners that most treatment agencies strongly promote group counseling only and this makes it difficult for individuals to receive the type of treatment that would benefit them most, regardless of whether they are male or female. A majority of practitioners reported that there are fewer gender-sensitive or female-only treatment services available to drunk drivers today than was the case even a few years ago, and this is perceived to be a significant issue of concern. It was generally acknowledged that while placement of male and female drunk drivers in an appropriate treatment setting is important to their ability to succeed, the availability of such services is lacking, particularly in rural jurisdictions. In all jurisdictions, practitioners acknowledged that previously available female-only treatment services have been shut down or discontinued due to funding and resource issues. They indicated that this was also the case in relation to mental health services and, overall, expressed concern about the shrinking or erosion of the range of essential services that are available to all drunk drivers. This results in offenders being placed in treatment programs that are available as opposed to those that are most appropriate. 56 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

307 Interestingly, some practitioners reported that they perceived that treatment professionals with their own past history of substance use can be more effective than those without personal substance abuse experience. They suggested that it is more difficult for drunk drivers to manipulate these counselors. It was also believed to be important because female drunk drivers in particular require motivation and it is beneficial for them to see a success story (i.e., a person who had struggled with addiction issues and managed to gain control of those issues and build a better life). However, some practitioners also expressed concern about practitioners in recovery as there were instances when there seemed to be a void with counselors not sharing important information with the probation officer when an offender reported a relapse during a treatment session in a perceived effort to protect the therapeutic relationship and prevent the offender from having a violation recorded (although it should be noted that a report of relapse will not necessarily result in a violation). Another issue of concern expressed by many practitioners was the lack of trauma-informed clinicians in treatment settings. Of equal concern, it was also noted that most alcohol intake instruments that are used in treatment settings do not include questions about trauma, meaning that these issues which may be important contributors to or triggers of alcohol use, particularly among a female offender population, are frequently not identified or are overlooked. Many practitioners reported that generic treatment programs targeted towards all types of substance use are often not consistently equipped to deal with trauma issues. In these instances, practitioners acknowledged that counselors often do not want to hear about trauma and/ or abuse and this affects the overall effectiveness of the treatment. More than half of practitioners agreed that the supervision of female drunk drivers can be more effective in a DWI Court setting because the entire team of practitioners (e.g., prosecutors, judges, probation officers, and treatment professionals) receives a greater amount of information that is shared among them. It is believed that this contributes to a higher level of accountability and also provides a stronger support network for women in particular. They also reported that DWI Courts are more likely to have linkages to the mental health system and services and this enables women to deal with these issues in an appropriate setting. For this reason, practitioners working in these settings report that they feel better equipped to manage female drunk FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 57

308 drivers. However, it was acknowledged that this is equally the case for male drunk drivers. There was also consensus among many practitioners that alcohol monitoring technologies (e.g., alcohol ignition interlock, continuous transdermal alcohol monitoring devices) are beneficial for both female and male drunk drivers. Not only did they agree that these devices contributed to a higher level of accountability among offenders, but they also enable counselors to more easily identify and address relevant issues. They acknowledged that technologies make it easier for treatment professionals to work with clients and note that, Technology is the client s conscience to get them sober Experiences with supervision and treatment Reactions to arrest and court processing. According to the experience of practitioners, among many of those women who are college-aged and single, the initial arrest and court processing that results from a drunk driving offense is not generally perceived as a serious situation. They reported that these women often fail to fully appreciate the magnitude of the consequences of their offense, particularly in terms of having a criminal record or the extended loss of their driving privileges. It was also suggested by practitioners that one explanation for this may be that younger women are in denial about their drinking behavior, and/or they have less to lose at this point in their life because they are more often single and do not have significant family or employment responsibilities which are frequently a source of concern among older women. In general, they reported that the attitude of these younger women often conveys the sentiment that I don t need help; I got this. Although it is estimated that more than half of practitioners agreed that this reaction is much more pronounced among first offenders, they also reported that if young women continue to accumulate subsequent impaired driving offenses, their attitude does change rather dramatically as they begin to appreciate the consequences associated with their behavior. They reported that women also begin to experience a much higher level of concern about their future and the opportunities that will be available to them, particularly if they are convicted of felony offenses. These observations are similar to focus group findings as some of the young women stated that they failed to appreciate the severity of their offending until their third or fourth offense, at which point there was no longer leniency and they instead encountered severe sanctions such as jail time. 58 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

309 Most practitioners reported that in comparison, women in their 30s and older are much more likely to appreciate the seriousness of their circumstances, and to acknowledge the arrest and court processing as a wake-up call to the fact that they are in trouble (in relation to their drinking) and need help. They suggested that one reason why older women take their arrest and court processing seriously the first time is because they have more to lose in terms of child custody arrangements, family disapproval, community stigma, and loss of employment. Practitioners acknowledged that, in many cases, women indicated that they are mad at themselves particularly given the burden that their offending can place upon their family. Regardless of the age of female offenders, it is estimated that a majority of practitioners agreed that the arrest and court process is more likely to be traumatic for women than it is for men. Furthermore, almost half of them reported that women also appeared to find the court process more intimidating and confusing than men. They proposed that one explanation for this may be because women are less likely to have been previously involved in the criminal justice system for other offenses combined with the additional stress resulting from women frequently being the primary or sole caregiver of children. They further reported that, overall, women tend to experience more shame in relation to their offense, particularly if they have children, and that they are also more likely to try to hide their arrest from family members and/or their spouse whereas with males this is less often the case. Generally speaking, practitioners indicated that women appeared to be more embarrassed to be in court than men, and that women expressed more concern about the well-being of others, and the impact that their arrest/ conviction will have on those close to them, which is in sharp contrast to men. It was acknowledged by at least half of practitioners that more women rely upon a public defender than do men. However, it was also noted that those female drunk drivers who could afford a private attorney would utilize one, although this was more often the case for women charged with a misdemeanor and less often the case for women charged with a felony. Moreover, it is estimated that many practitioners agreed that women are more accepting of the process and are less resistant to it, whereas men generally have more anger about it. It was further suggested that these differences may result, in part, from the fact that women are much more likely to accept FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 59

310 responsibility for their offense in court and are more likely to plead guilty, whereas men tend to make excuses or attempt to justify their behavior in court and are less likely to plead guilty. There was also some discussion among practitioners in some jurisdictions with regard to the experiences of ethnic minorities. In particular, some practitioners noted that drunk drivers who are Hispanic (both male and female and particularly first generation) are more likely to accept responsibility for their offense and are motivated to fulfill requirements imposed upon them to reach closure of the process. At the same time, it was also noted that this population of offenders appears to be somewhat more reluctant regarding self-examination and behavior change in relation to their drinking, and this was noted in particular in DWI Courts. One possible explanation that was offered by practitioners is that this particular group of offenders is concerned about ongoing involvement in the criminal justice system due to concerns about immigration status and deportation issues. Supervision experiences. Perhaps the most significant finding in relation to supervision experiences of female drunk drivers is that all practitioners agreed that the actions and attitude of probation officers or case managers is an important factor in the ability of female drunk drivers to successfully complete the requirements of their supervision. While this is likely true in the case of male drunk drivers as well, this was noted particularly in relation to females because this population appears to have heightened concerns about a broader range of issues and they may face more significant challenges as the sole caregiver or breadwinner as a single parent. Subsequently, practitioners reported that women tend to place greater demands on probation officers and case managers in terms of time and emotional support. These issues are described in more detail below. Overall, it is estimated that more than three-quarters of practitioners agreed that they spend more time in appointments with female drunk drivers than they do with males, and more time during these appointments is spent listening to offenders discuss their issues and/or concerns. This is especially the case once women have established a rapport with their probation officer or case manager. They noted that women generally have more questions and seek to talk more about their current situation as well what may happen in the future. To illustrate, female drunk drivers ask more questions about and are more focused on long-term scenarios, also referred to as the what-if 60 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

311 phenomenon (e.g., Will I be able to be present when my grandchild is born in six months?; When will I get my license back?) as opposed to immediate supervision issues. Practitioners reported that men, on the other hand, tend to want to get out of their appointments as quickly as possible and do not feel the need to have lengthy conversations with their probation officer or case manager. It was also agreed by many practitioners that women require more guidance and support in relation to practical issues as well as emotional issues. In this regard, some practitioners indicated that female drunk drivers generally are more emotional about their circumstances, and it is not uncommon for them to express intense emotions during these appointments. They reported that women can be more sensitive to feedback, and for these reasons, appointments with women can be more emotionally draining for officers. Practitioners reported that female offenders tend to sense when probation officers are not being genuine with them and therefore, it is important to assign practitioners who want to work with this population and have the requisite skillset. Practitioners acknowledged that necessary attributes include a caring demeanor, an excellent grasp of motivational interviewing techniques, a willingness to help female offenders be successful, and the ability to hold them accountable in the event of violations. It is estimated that a majority of practitioners acknowledged that female drunk drivers generally tend to be less forthcoming, particularly in relation to their past experiences as well as current living or family situation than males. For example, women generally will not report domestic violence unless there are obvious signs or family members get involved. They consistently agreed that practitioners have to listen for flags and ask more questions at the outset to gain a complete understanding of the situation of female drunk drivers. Male drunk drivers are the opposite in that they are much more likely to volunteer information and be an open book. As one practitioner noted, Men are straightforward. Women can be exhausting to deal with. Overall, there appeared to be a high level of consensus among practitioners that the reasons why female drunk drivers have a problem with alcohol and/ or other substances are often more complex than is the case for males. As a consequence, practitioners indicated that they frequently need to spend more time gathering information about the offender s life history or current situation when they are female. They also spend more time trying to identify FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 61

312 key issues and triggers (e.g., Do they have issues with authority?; Do they have a history of abuse?) as well as optimal strategies to manage female drunk drivers. Researchers estimated that most practitioners agreed that time management (i.e., scheduling of priorities and responsibilities) is more often an issue for female drunk drivers than it is for males. Whereas they reported that male drunk drivers often have fewer responsibilities (e.g., no children), or may have a spouse to provide support (e.g., drive them around), female drunk drivers generally have more responsibilities in terms of childcare, employment, finances, and community involvement, particularly if they are a single parent. This can make it more of a challenge for them to fulfill all of the requirements related to their supervision and treatment, and, for this reason, it is essential that female drunk drivers develop a plan to enable them to manage these responsibilities. At the same time, practitioners also acknowledged that women require more flexibility in scheduling in order to enable them to successfully cope with these increased demands. Practitioners suggested that probation officers should recognize the hardships and demands that female offenders face in their daily lives and work with them in a collaborative fashion to assist them in meeting their supervision requirements. There appeared to be a high level of consensus among practitioners that women are often busier, particularly at the beginning of supervision as they learn to handle the increased number of appointments and conditions imposed upon them. Some of them also suggested that female drunk drivers being busy can be an important factor in their remaining sober as it keeps them out of trouble and eliminates distractions that contribute to their drinking behavior. An estimated half of the practitioners also acknowledged that female drunk drivers are more employable than males, and some of them also suggested that this may be because they more often seek low-level, minimum wage jobs which are more readily available. A number of practitioners reported that it was not uncommon for females to be more fearful about losing their teaching/nursing license and are more concerned about the embarrassment of having to report their offense as part of the re-licensing process. Practitioners noted that, as a consequence, females are also less likely to renew their professional license once it expires and this can result in them seeking alternative employment at a lower wage. 62 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

313 There was a high level of agreement among half of the practitioners that female drunk drivers are not more or less likely to have support networks to assist them during their period of supervision than men, although the balance of them disagreed and indicated that women were less likely to have a support network. As evidence of this latter finding, it was noted that, in relation to offenders who are incarcerated, males frequently get more visitors than females. They agreed that this may be because females do not have a spouse and their family is unable to visit regularly. They also noted that female drunk drivers often discourage visits because of the shame and embarrassment of having their children and/or family see them in jail or prison. Similarly, some practitioners reported that women are less likely to be picked up or dropped off at appointments or classes by their spouse, in sharp contrast to the situation with males. In cases where a female drunk driver is receiving transportation support, it is most often another woman who picks her up and drops her off. Women are also much more likely to miss appointments or classes, or to appear at them with their children, than men because of their inability to arrange childcare. Among those female drunk drivers that are in a relationship and/or have a spouse, some practitioners noted that the male partner is likely to complain to the woman about the effects of their situation, and be less willing to help with transportation or childcare. This is usually not the case if the situation is reversed as women are likely to be supportive and accommodating when their boyfriend/husband is the offender. In this regard, it is suggested by some practitioners that it should not be assumed that female drunk drivers with a spouse are more likely to have support and that it may, in fact, be likely that these women have to cope with the frustration of their spouse in addition to the stress they already face. In fact, it was acknowledged by a number of practitioners that women who are in relationships are actually more likely to stay, even if the relationship is unhealthy or destructive, because of financial reasons or because they need a source of transportation or childcare to successfully complete their sentence. They emphasized that this contributes to their victimization and should be a consideration with regard to supervision. Several practitioners also expressed concern about the ability of women to stay sober in these environments as the male spouse will often not be supportive of sobriety and instead, encourage drinking because FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 63

314 they perceive the woman to be calmer or less high-strung when she is drunk. There were mixed findings with regard to practitioners perspectives on the likelihood that women will seek to build a support network if they do not have one. Some of the practitioners reported that women may actively engage in AA or other support groups more in an effort to establish some type of support network or to meet individuals who can provide assistance with transportation. However, it was equally noted that some women will intentionally avoid any offers of assistance and resist developing a support network because they do not know how to ask for help, they are too embarrassed to ask for help, or they will insist on managing all of their responsibilities independently. A handful of practitioners referred to this type of female offender as the super mom who was accustomed to handling a plethora of responsibilities on their own even when the situation became untenable to manage. Practitioners perspectives with regard to the financial situation of female drunk drivers and their ability to cope with the financial demands associated with arrest, case processing, and supervision were also mixed. It is estimated that approximately half of them reported that women have greater concern about and struggle more with the financial implications of involvement with the criminal justice system than their male counterparts. Practitioners reported that this is largely due to the fact that many female offenders are single parents, and as such, frequently bear all of the financial burden associated with caring for children as few of them receive child support payments from their spouse. The other half of practitioners reported that women did not struggle more financially than male drunk drivers. In the same vein, it is estimated that more than half of them indicated that cost is a greater barrier to treatment participation for women than it is for men, and that perhaps 50% of women can afford to pay for treatment. However, a number of them also noted that it is not so much an issue of affordability as it is of willingness to pay, and proposed that women are less likely than men to be willing to pay for treatment. Of interest, some practitioners suggested that female drunk drivers are much more motivated to stop drinking because of the costs associated with the arrest, case processing, and supervision. 64 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

315 Finally, in relation to supervision, approximately one-third of practitioners reported that female drunk drivers are more likely to require and to receive referrals for mental health services and other community services, particularly in relation to housing. It is not uncommon for the drunk driving arrest to precipitate thoughts about or actual departure from unhealthy relationships. As such, practitioners noted that, in particular, women are more likely to require assistance with transitional housing. A minority of practitioners also indicated that female drunk drivers more often require family services, although involvement with CPS on an ongoing basis is rare and is more common in cases where there are several charges in addition to drunk driving (e.g., drug possession, assault). Practitioners in some jurisdictions acknowledged that there is often a deferred response for family services in relation to child protection issues, and women are more likely than men to use this service in which counselors go to the home for visits and oversight purposes. Treatment experiences. With regard to alcohol education sessions, a minority of practitioners reported that having men and women in the same alcohol education class or treatment group was not a problem for either gender. In fact, some of these practitioners further suggested that mixed settings for education and/or treatment are beneficial to help women to better understand male experiences so that they can critically examine their own issues, particularly those pertaining to relationships and addiction. It was proposed that this may be useful to provide females with other perspectives and insight into their own situation. Conversely, it is estimated that at least half of practitioners indicated that it is more difficult for women to engage in and benefit from mixed-gender education and treatment services, as well as support groups like AA. They noted that this is often related to the fact that women in general are more likely to have had negative experiences with abusive male partners or sexual relationships. Of concern, they suggested that not only are women less likely to share their experiences or receive support in these settings, but also that men are often a source of distraction from their recovery. Female focus group participants echoed these sentiments and acknowledged that some males in these settings made a habit of seeking emotional and or other relationships from women, and referred to them as the thirteen steppers. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 65

316 Some treatment providers try to manage this by requiring appropriate attire by both sexes at meetings. In relation to the response of female drunk drivers to education and treatment, it is estimated that more than half of practitioners agreed that women tend to be more skeptical at the outset when they enter treatment than males. However, they equally agreed that, once women attain a level of comfort, they are more likely to open up about their experiences and bond with counselors and peers more easily than men. This is particularly true in the case of female-only services. Many practitioners generally agreed that, in femaleonly programs, women exhibit a higher level of empathy and provide support for others, and they are more likely to share intimate details about loss. In these settings they report that women share more personal information and tend to take the classes seriously whereas men are more likely to make jokes and laugh about alcohol education classes and treatment. They view it as a requirement that they must complete as opposed to an opportunity from which to learn or derive benefit. As a caveat, practitioners highlighted that women who engage seriously in treatment are more likely to exhibit extreme behavior and frequently pursue cleanses, juices, and diets to improve their health which often have implications for random testing results. Overall, a large majority of practitioners acknowledged that female drunk drivers are more receptive to female-only treatment groups, with some noting that men prefer this also. There was also a high level of agreement among practitioners that female-only groups are rarely available and that this is a critical gap in the treatment system. Lastly, it is also estimated that more than half of practitioners concurred that individualized treatment or counseling for female drunk drivers appears to be helpful, particularly at the outset because women may have supressed or not be aware of how past experiences have shaped their behavior. Practitioners generally agreed that this approach has the added benefit of enabling practitioners to get a more accurate history to inform treatment and supervision strategies. Some jurisdictions with DWI Courts indicated that there is more one-on-one counseling available to female drunk drivers in these settings and this is important to help them feel comfortable and encourage them to talk about their past experiences. 66 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

317 Supervision and treatment outcomes. It is estimated that a majority of practitioners agreed that female drunk drivers are more likely to become compliant sooner than males, although they acknowledged that a significant majority of women attempt to manipulate the system or avoid requirements of supervision at the outset, particularly if they have no or few prior offenses. According to these practitioners, women more often express the sentiment that they just want to complete the process and get out of the system. They further reported that female drunk drivers generally have a higher level of fear about going back to court than do male offenders, and that they are more cognizant of the shame and cost associated with their offending, as they are about its overall effects on their children. As such, these practitioners reported that women are less likely to be violated for non-compliance, and when they are violated it is often in relation to missing appointments due to transportation or childcare issues. Approximately half of practitioners also acknowledged that female drunk drivers are more likely to successfully complete supervision/treatment than men and in a shorter period of time. Moreover, it was agreed that they have lower levels of absenteeism in groups and are less likely to repeat treatment. It is estimated that there was a high level of consensus among many practitioners regarding the presence of specific barriers to the successful completion of supervision and treatment by female drunk drivers. While these barriers also exist for male offenders, they are not as pronounced for a male population. These reported barriers include: > Female drunk drivers are more likely to struggle financially to meet the costs associated with arrest and supervision. Not only may married women not have access to money depending on the nature of their relationship with their spouse, but as single parents they are more likely to bear all of the costs associated with child rearing. > Female drunk drivers are more likely to have full custody of any minor children, and this is particularly true if a child has special needs. A lack of access to affordable childcare services can prevent these women from attending supervision appointments or treatment sessions. > Female drunk drivers are more likely than men to struggle with transportation issues not only as a result of limited availability in rural areas and/or a lack of a support network, but also because of the safety FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 67

318 concerns that many women have about using public transportation at night, and/or accepting rides from persons they do not know well. In terms of outcomes, it is estimated that more than half of practitioners noted that women are more likely to accumulate multiple offenses in a fairly short period of time as compared to male drunk drivers, although, in their experiences, men are more likely to re-offend overall. One explanation that was proposed to explain this phenomenon was that women can be overconfident about their sobriety and then relapse because they have not dealt with the underlying issues that contribute to their drinking (e.g., mental health issues, trauma). 4.3 Lessons learned Based on the collective experiences of practitioners in the supervision and treatment of female drunk drivers, some of whom have specialized in this area, there were a number of lessons learned that they shared which can be useful to inform the development of specific strategies and interventions to better serve female drunk drivers. These are briefly highlighted below and structured according to what practitioners can expect, as well as what strategies have proven to be useful and successful in many cases What to expect supervising and treating female drunk drivers > Women are more likely to try to manipulate the system and avoid the requirements of supervision and/or treatment at the outset of the process. This may be because these women have been able to successfully manipulate other persons in their life and this is a strategy that they have learned is effective and have relied upon in the past. In these cases, it often takes women longer to get past the games, to open up about their experiences, and to approach their situation with more seriousness. For this reason, practitioners agreed that whereas supervising and dealing with male drunk drivers is very black and white with these offenders being fairly transparent, the supervision and treatment of female drunk drivers is described as being more shades of gray. > Female drunk drivers often come to supervision and treatment with a wide range of issues in addition to substance abuse and/or dependence such as mental health problems or a history of trauma. In most cases, they have not sought help to deal with any of these issues and only do 68 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

319 so because they are forced to as a result of the court process, probation, or treatment. More time and effort are needed on the part of supervision and treatment professionals as compared to male drunk drivers to help them unpack these issues. > Female drunk drivers generally experience more pressure to succeed because of their financial and/or childcare responsibilities. It is not unusual for these women to convey that they have the feeling that they cannot fail in relation to their diverse responsibilities. Females are more likely to have high expectations of themselves to perform or succeed because they feel that others are counting on them to fulfill different roles. As such, these women tend to be preoccupied with trying to show that everything is okay. > Whereas children are often a motivator for many female drunk drivers to be successful in completing their supervision and treatment requirements, this is not true in all cases. However, for those women who are motivated by their children, it is essential to recognize that their children can be a significant influence on their behavior and play an important role in their decision-making. For these female drunk drivers, having their children show pride in or be proud of them as a parent is very important. In addition, it can be very challenging for these women to focus on supervision or treatment requirements if they are concerned about the welfare of their children. > Female drunk drivers often are more emotional about their situation and frequently expect and need practitioners to let them talk and also to listen more so than male offenders. Women are more likely to shut down in these situations if they feel rushed or ignored. It is not unusual for women to share their emotions during appointments and this can be exhausting for practitioners. In this regard, women frequently are seeking opportunities to establish a rapport or connect with the practitioner. On a positive note, female drunk drivers are also more likely to be able to help themselves and to problem solve but they need emotional support to do this. Generally speaking, once they can establish this connection, they will work harder to dig themselves out and move forward. > It should be underscored that, in many cases, it takes a long period of time for female drunk drivers to recognize their chaotic thinking and to develop strategies to help them defend against extremes in behavior. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 69

320 Practitioners can benefit from exercising more patience when dealing with female drunk drivers, and from taking steps to ensure that they are able to manage any frustration they may experience. > Female drunk drivers are more likely to need assistance with relationship dynamics and this is important to their success. In many cases, it can be easier for women to manage their situation and successfully complete their probation and treatment requirements if they are single. This enables them to deal with drinking issues which can be hard to do when they are focused on or distracted by a relationship in order to make it work, or if they are involved with a partner who also has a substance use problem. > There is not one program or intervention that will work universally for all female drunk drivers; what strategies/interventions are most likely to result in successful outcomes is often a function of the complexity of their drinking problem. Practitioners need to have the determination to find the right program for women and to never stop trying to get the light switch to come on. As such, female drunk drivers tend to be more honest and comfortable in female-only groups so this should be a consideration. Women who also have a history of trauma may benefit from inpatient treatment at the outset. > Whereas the wife of a male drunk driver is more likely to call a practitioner if they believe their husband is at risk for relapse or does relapse, the husband or male spouse is less likely to call if a female drunk driver is at risk of relapse or does relapse. In some cases, when a male spouse does report this, it may be a reflection of an unhealthy relationship or domestic violence situation. For this reason, it is important that practitioners are more sensitive to potential flags when dealing with female drunk drivers, and to also take steps to ensure that these women are not re-victimized by their spouse. > For female drunk drivers in particular, anniversaries of traumatic events (e.g., arrest, death, end of a relationship, and so forth) can trigger intense emotions and they may be more likely than men to be at risk for and/or experience a relapse at these times. 70 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

321 > Women are more likely to be pro-social and compliant than men, most often because of the pressure they feel to succeed and the broader range of responsibilities that they have in relation to childcare and financial support. > Female drunk drivers are more likely to be successful than males in completing their supervision and treatment requirements particularly if they have family/friends that are supportive of their sobriety Strategies for supervising and treating female drunk drivers During the course of the interviews, each of the practitioners shared what they had learned based on their individual experiences in relation to more appropriate strategies to both supervise and treat female drunk drivers. These strategies can be useful guidelines for other practitioners to consider and are briefly summarized below. > Drug screens for suspected female drunk drivers at the time of arrest can be a very helpful source of information to identify potential issues with drugs, particularly as women may be less forthcoming about substance use in addition to alcohol at the beginning of their involvement in the criminal justice system. Practitioners further acknowledged that making specific inquiries about a female drunk driver s prescriptions and continually checking their prescriptions can also help identify problems with prescription medications as well as tendencies towards doctor shopping. > The use of a presenting assessment tool during intake can assist practitioners in the identification of substance use and mental health issues at the outset and, subsequently, help them make informed and appropriate referrals to services initially. > Intensive monitoring and support are particularly useful for female drunk drivers at least at the beginning of their supervision period, in order to create a high level of accountability. At the same time, it is important to balance supervision and accountability with compassion (as appropriate) and rehabilitation. To this end, the matching of probation officers and treatment counselors with appropriate skillsets to best meet the needs of female drunk drivers should be a consideration as part of case assignment. To better accommodate some of the specific challenges that FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 71

322 female drunk drivers face, it is helpful if probation officers and treatment counselors can exercise a certain degree of flexibility and discretion to address the individual circumstances of each female, and to help women be successful. In addition, practitioners that are prepared to serve as a resource and source of support for females are perhaps better suited to manage this population. > It is important to remove issues of ego, shame, and guilt from the conversation with female drunk drivers and to start with the individual and refrain from making judgments (either real or perceived). Initial conversations with females should gauge their current level of selfesteem and also what factors contribute to or detract from their selfesteem. According to practitioners, Some women fake it well but have low self-esteem, and it is helpful to understand where female drunk drivers are at in relation to their individual perceptions of self-worth. This understanding helps to ensure that practitioners are better able to recognize and utilize strategies that contribute to the building of selfesteem as part of their approach to supervision and treatment for this population. > The creation of a safe environment that women perceive as welcoming can assist in the establishment of rapport and help women to open up (i.e., setting up an office so that there is good lighting, plants and/or flowers, artwork, pamphlets that contain information relevant to women, and so forth, in an effort to create an inviting as opposed to a sterile atmosphere). > The stress and anxiety of female drunk drivers can often be eased by informing them at the beginning of the process that they will receive a lot of information and paperwork about their case and supervision during initial discussions and appointments. It should be emphasized that there is recognition that it is a lot to absorb and that one of the roles of supervision professionals is to help them figure it out and ensure that they manage their conditions accordingly. By delivering this information in manageable packages as appropriate, practitioners can make the process seem less overwhelming for female drunk drivers and can make them feel as though they are better equipped and capable of accomplishing what is expected of them. In this regard, officers should be prepared to provide ongoing guidance and support. 72 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

323 > Consideration of a more comprehensive approach to supervision and treatment can benefit female drunk drivers who frequently need assistance to address a broader range of factors in relation to their life situation beyond drinking and driving. In addition, a key factor that is useful to inform these decisions is that many female drunk drivers are unable to focus on supervision or treatment until they attain a reasonable level of comfort in relation to the living situation of their children and their ability to continue to access and care for them. > The selection of strategies that involve incentives to encourage and reinforce compliance, that are designed to motivate female drunk drivers, and that can help build self-esteem are highly beneficial for this population. By supplementing these strategies with verbal praise and demonstrations of support, practitioners can increase a woman s belief that she can be successful. At the same time, it cannot be overlooked that female drunk drivers are more likely to perceive their probation officer or treatment counselor as a positive role model, and this may be the first time they have had not only a role model but also someone who is supportive of their success in life. As such, it is underscored that practitioners should be sensitive to these perceptions and take steps to not only be a source of support and guidance but also to maintain appropriate boundaries and a professional relationship with female drunk drivers. > Practitioners should take extra steps to learn to read between the lines in relation to female drunk drivers in order to better identify custody concerns, domestic violence, and mental health issues that may affect supervision and treatment outcomes. Of importance, male practitioners report that, whereas male drunk drivers are more likely to be up front about their issues, female drunk drivers are less likely to spontaneously disclose them. Conversely, female officers indicate that female drunk drivers are equally likely to disclose such issues. As such, the gender of the practitioner may be a factor in relation to the willingness of female drunk drivers to disclose personal or sensitive issues. > It is essential that practitioners be sensitive to potential flags that may indicate sources of concern that should not be overlooked. These flags may include the status of a female drunk driver s relationship with their significant other (whether they are supportive or not); as well as whether their family is supportive or not. In some cases, the arrest and supervision FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Practitioner Interview Results 73

324 requirements often force female drunk drivers to choose between sobriety and maintaining a personal relationship. Not only is this situation difficult emotionally, but it can contribute to relapse. > The supervision of female drunk drivers who experience domestic violence may require additional sensitivity and precautions. It is not only important that women believe that they can trust the practitioner, but the practitioner also has to protect against letting the spouse abuse the situation and continue the victimization. In this regard, efforts for supervision and treatment professionals to create a non-threatening environment for female drunk drivers who are victims of domestic violence and/or have a past history of trauma are important. To illustrate, the size or set-up of the room for appointments with women and the tone of voice used by the practitioner could potentially be perceived as threatening and must be taken into consideration. In some instances, female drunk drivers may be more receptive to discussing personal issues with a female practitioner as opposed to a male, although this is not a factor for all women. > Strategies that assist female drunk drivers in recognizing their potential lack of boundaries in relationships, and how this contributes to their addiction or behaviors that are connected to their offending are helpful to those offenders for whom this is an issue. It is also equally important to consider the use of strategies that can assist female drunk drivers in learning how to set boundaries and better protect themselves from unwanted or negative experiences. > It can be beneficial to try and engage a female drunk driver s family in supervision and/or treatment, to the extent possible, so that they can be a source of support for the offender s sobriety. Explanations about addiction and its effects, as well as the needs of individuals seeking sobriety can contribute to female drunk drivers experiencing a higher level of support and ultimately, success. At the same time, ongoing encouragement for female drunk drivers to attend AA support groups in order to help them build a network of support can also be a contributing factor in their success. 74 Practitioner Interview Results FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

325 5. RECOMMENDATIONS Based on feedback gathered during the interview focus groups and individual interviews with female drunk drivers, and the interviews with practitioners representing probation officers, judges, defense attorneys, alcohol education counselors, and treatment professionals, a number of recommendations emerged that can inform efforts to strengthen prevention initiatives as well as improvements to the criminal justice and treatment systems for dealing with female drunk drivers. These recommendations are discussed in more detail below. 5.1 Prevention The first step toward reducing incidents of impaired driving is to increase education among women about alcohol. In particular, prevention efforts should be targeted at young women and delivered in a school setting before they begin drinking and it should be ongoing throughout high school. It is important that women learn at an early age what constitutes normal drinking and what is classified as excessive or binge drinking. Other important areas of education include how alcohol is distributed in the body differently for each sex (i.e., when a man and a woman drink exactly the same amount of alcohol under the same circumstances, the woman s BAC will be higher) and how drinking effects driving performance. In addition, increased awareness of and attention to the identification of emotional problems, mental disorders, or difficult living circumstances among women is also very important from a prevention perspective as these issues often compound a woman s situation and can contribute to substance misuse as well as drinking and driving. As one offender put it, We have issues even before we begin drinking. Young women who face these challenges should FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Recommendations 75

326 be encouraged to seek counseling or treatment and be taught that turning to alcohol and/or drugs is not a good coping strategy and is likely to increase problems as opposed to alleviate them. Furthermore, efforts are needed to breakdown stigma associated with these issues that can prevent women from seeking the help they need. 5.2 Criminal justice system recommendations The following are recommendations to improve the experiences of female drunk driving offenders processed through the criminal justice system as well as recommendations relating to the management of their supervision. The recommendations highlight strategies that practitioners may wish to consider and/or utilize when monitoring this offender population. Suggestions are also put forward as to the most appropriate and effective interventions as identified by both offenders and experienced practitioners Recommendation from offenders > Provide guidance and assistance to female drunk drivers to help them manage life issues related to their offending. In particular, women are seeking assistance in the identification and use of strategies to ask for forgiveness from children following their conviction for impaired driving. This was cited by almost all women as an area where probation officers or treatment professionals could be of help to offenders as they frequently reported that this was a priority issue to them that they felt unequipped to manage. > Provide clear information about the conditions of probation, requirements of sentencing, and any additional responsibilities during initial meetings with probation officers. Female offenders expressed frustration with the lack of information they received or inconsistencies in response to their questions. These women noted that the supervision process could be simplified if they know exactly what the expectations of them are and how to avoid violating their conditions. In the interest of fairness, these women also feel as though they should equally be made aware of what avenues are open to them in the event that they violate (e.g., appeals, second urinalysis tests). > Work with female drunk drivers to achieve a good balance in how their time is scheduled and filled. Female offenders need a sense of purpose and to be kept busy in order to increase feelings of self-worth, to occupy 76 Recommendations FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

327 free time in a constructive manner, and to maintain a certain level of accountability. However, balance is needed as they also must address their issues and not be permitted to be so busy that they can run from life or their problems. Practitioners are integral to help them achieve this balance through active listening and practical suggestions about the scheduling of various responsibilities. > Recognize the individual value and accomplishments and/or progress of offenders. As one woman noted I don t want to be just another case or docket number. Of importance, this recognition builds their selfesteem and motivates them to try harder and be successful. Many of these women have not felt good about themselves in a very long time which is often a contributing factor to their substance use. One of the biggest motivators for these women is to have their progress recognized and for them to know that they are making gains and getting their life back on track. This sense of accomplishment can be fostered through demonstrations of trust, respect, and encouragement on the part of probation officers when the women are able to meet conditions of their supervision such as gaining employment. > Be honest, clear, and follow through in your interactions with female drunk drivers. For example, if appointments are scheduled with offenders, it is important that appointments take place as scheduled, particularly if the offender is trying to maintain employment and manage childcare arrangements, and that follow-up on inquiries is completed. > Consider increased flexibility in the scheduling of probation appointments, testing, and treatment for female offenders as appropriate. Many female drunk drivers find it challenging to meet all of these requirements and still maintain employment as scheduling often occurs during work hours. Few employers are sympathetic to frequent requests for time off, especially on a weekly basis. While it may not always be possible to accommodate offenders schedules, probation officers are encouraged to work with them to develop solutions that allow them to maintain employment and still meet all conditions of probation. Otherwise, the women may feel as though they are being set up to fail. > Identify the most appropriate and best suited treatment intervention for each offender as this is a key to their success in recovery. If women are forced to attend treatment services that they feel lack value or that FEMALE DRUNK DRIVERS A QUALITATIVE STUDY 77 Recommendations

328 they are not deriving benefit from, it is unlikely that it will assist them in the maintenance of sobriety (e.g., mixed gender groups that consist of participants with varying degrees and types of addiction issues). Practitioners are encouraged to find a balance between programs that are available and affordable and those that address the individual risks and needs of female offenders. Practitioners may consider working collaboratively with the women they supervise to find the right treatment program so they are comfortable to disclose, share experiences, and address their issues. Also, in an effort to further establish rapport and respect, it is recommended that probation officers listen to women when they say that a treatment program is not working. At a minimum, this may warrant a discussion about what aspects of the intervention the women perceive as problematic or of little value. > Minimize stress and discomfort in situations when a female drunk driver is assigned to a new probation officer. It is not uncommon for female offenders to switch probation officers over the course of their period of supervision (which is often a function of caseload/workload of different officers). This may be a difficult transition, particularly if a positive relationship and good rapport had been established with the current officer or case manager. In order to make the transition as smooth as possible, practitioners might consider making the change gradually or scheduling a meeting with both officers and the offender to ensure that everyone is brought up to speed about the case. > Minimize situations in which female drunk drivers must constantly relive their story and review the circumstances which led to their offense. This is painful and depressing and reinforces their low self-image. While they recognize that it is important to learn from past mistakes so as not to repeat behavior, at a certain point this becomes counterproductive. Women are more likely to benefit from an increased focus on the present and recognition of the progress that they have made and future challenges. This will allow them to move forward instead of being stuck in the past. > Find constructive strategies to deal with emotions and emotional situations. It is at these times that women are most in need of support and encouragement. While this may pose challenges, it can be useful to listen to concerns instead of perceiving the situation as a power struggle and/ 78 Recommendations FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

329 or dismiss it as a petty issue. Helping females work through emotional situations can be important to teach them how to problem solve and not overreact when solutions are available to them. > Manage perceptions around the inconsistent application of sanctions across individual offenders by making clear why sanctions are applied and what factors are taken into consideration. Perceived differences in sanctions across female drunk drivers is another source of frustration to them. It is important to note that many of these women are in group therapy together and share their experiences and stories. Often, they learn what others with similar histories or offenses have received, and what they perceive to be more lenient sanctions or conditions can lead to perceptions of a lack of fairness or inequality. For this reason, it might be beneficial for probation officers to explain to the offenders they supervise why they received a particular sentence and/or conditions. It is also recommended that probation officers take into consideration compliance history and alter conditions as appropriate to recognize progress. > Increase the availability of education for criminal justice practitioners about substance abuse and dependence. This can help practitioners not only understand the psychology of addiction but also understand about relapse and the likelihood of its occurrence, and the importance of matching offenders to appropriate treatment interventions. The ability to identify triggers and red flags among probationers could help prevent relapse and also to make appropriate treatment and intervention recommendations. > Increase education for criminal justice practitioners about female offender issues and about alcohol monitoring technologies such as ignition interlocks. The more information available to practitioners, the better equipped they are to apply appropriate sanctions, and effectively monitor offenders, and adress concerns Recommendations from practitioners > Work to talk, listen, connect, and develop rapport with the women that are supervised. Female offenders take a longer time to open up than their male counterparts and they need to feel as though they are in a safe environment in order to feel comfortable disclosing information. It might FEMALE DRUNK DRIVERS A QUALITATIVE STUDY 79 Recommendations

330 take women longer to ask for help as many have never learned to ask for assistance and have historically relied on themselves to do everything. > Seek to build the self-esteem and the trust of the women that are supervised. In this regard, it is important to know the women as individual persons as opposed to case files. The establishment of trust between the two parties can also facilitate information-sharing and motivate female offenders to comply with their conditions. > Set boundaries. While it is important to be a source of support for female offenders, probation officers cannot let women view them as a friend as that is not their role; they must be able to hold probationers accountable for their actions and avoid being manipulated. > Learn more about social work and increase the availability of social services, educational and vocational services for offenders generally to the extent possible. Practitioners recommend better communication and the development of relationships with other community and social services (e.g., employment, education). Increased follow-up with these services is also important to ensure that offenders have access and are able to complete the requirements/conditions of their supervision. > Develop a different protocol for home visits for female probationers. Factors to take into consideration include: the presence of children in the home who may be frightened; the embarrassment associated with probation officers announcing their presence, particularly in an apartment or communal living setting (e.g., showing up with jackets that say Probation or with weapons visible); the invasion of personal space that some women might feel if a male goes through their home and possessions; and the impact that visits to work for random testing can have on employment. It is recommended that a female probation officer be present during home visits or at least until a level of trust and comfort is established. 5.3 Treatment system recommendations Female drunk drivers also provided recommendations to improve their experiences within the treatment system. The recommendations highlight strategies that practitioners may wish to utilize when treating or making treatment and intervention referrals for this offender population. Suggestions 80 Recommendations FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

331 are also put forward as to the most promising and well-received treatment approaches as identified by both offenders and experienced practitioners Recommendations from offenders > Increase the availability of better and more extensive assessments (for both substance use and mental health) at the time of first offense along with referrals to appropriate and intensive treatment interventions for those who require them. Often, there is a lack of consequences for a first offense; some women perceive mandated education sessions as a joke because these classes are viewed as a place to meet people to drink with. Many of the women in this study reported that if there had been more serious assessment or if their issues (e.g., the underlying cause of their offending behavior) had been identified following their first offense, a second offense might not have occurred. As one woman noted, the first offense is really more about process than consequences. > Provide more affordable and higher quality treatment services. Not everyone has insurance and women often end up paying for the cheapest treatment options because they are unable to afford more intensive interventions or the therapy that is best suited to their needs. This frequently results in women paying for and attending treatment that is not relevant or beneficial, and this is a source of frustration and can result in unsuccessful treatment outcomes. > Offer the option of women-only treatment programs and services where possible. Attendance in a treatment program or therapy that is tailored to women is perceived by female offenders as the most preferable option. Not only do women prefer female-only groups but they also prefer groups that are comprised of women with similar issues (e.g., comparable levels of dependence) and who are in the same age category (e.g., older women have different experiences, issues, and responsibilities than younger women). When women are with peers who they feel they can relate to, they are more likely to share experiences and be receptive to feedback. > Provide women with opportunities to integrate real life responsibilities into their treatment. In other words, they want to learn how to cope with pressure and how to organize and manage their life in conjunction with treatment. They also want to apply the skills that they learn in therapy as FEMALE DRUNK DRIVERS A QUALITATIVE STUDY 81 Recommendations

332 they cannot learn skills in a vacuum and then be expected to know how to apply them outside of treatment in the real world in the absence of guidance and support. As one woman stated, with 30 days you get the information and knowledge to understand but then you are out the door and not yet equipped with the tools and skills. Female offenders need to learn how to deal with life differently and in a better way and, as such, they also require opportunities to practice the application of skills in an effort to change behavior and achieve better outcomes. > Encourage counselors to refrain from sharing their personal stories as treatment should be focused on the offender, not the counselor. Continual discussion of use may trigger some women and prompt relapse. One woman stated that the role of the counselor is not to teach you how to use, but how to live a sober life. > Do not require participation in treatment for women who do not have alcohol abuse or dependence issues (as determined by screening and/or assessment). They are unlikely to benefit from mandated participation in treatment and, in these instances, women tend to become resentful and view the treatment process as a waste of both time and money Recommendations from practitioners > Make available to practitioners more research about which treatment interventions and strategies work best with a female drunk driver offender population. Currently, there is a lack of research in this area and knowing what works would be beneficial. > Increase the availability of treatment services and higher quality services; this is not specific to DWI offenders, but rather for all offenders in general. There is also a need for more gender-sensitive treatment strategies and gender-specific therapy groups. > Increase funding and resources to help women who lack insurance or are of low socioeconomic status enter into evidence-based treatment programs that can adequately identify and address their individual risks and needs. > Make available more training and resources to improve the quality of treatment. Practitioners recognize that most available counselors are pretty well-equipped to deliver programming, however this varies across jurisdictions. There is also a need to increase recognition that women 82 Recommendations FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

333 are often more complex clients than men and it is difficult for them to manage all of their priorities and responsibilities (e.g., childcare and employment) which could affect their success in treatment. > Provide more training for clinicians and strengthen state certification protocols as appropriate. In some jurisdictions, little training is required to deliver therapy or treatment interventions and this is not an ideal scenario. It is often assumed that if a clinician is licensed that they are qualified, but that is not always the case. Subsequently, practitioners note the importance of the creation of standards in education and standardized testing for clinicians in order to be licensed. The implementation of audit protocols of treatment providers to verify that appropriate services are being provided is one potential option. Clinicians would also benefit from backgrounds or education in the areas of trauma and relationships, particularly when working with female clientele. > Use alcohol-intake instruments that acknowledge and identify a history of trauma as this can have implications for treatment (e.g., it could assist practitioners in making better referrals to more appropriate services). There is also a need for more trauma services, trauma-informed clinicians, and one-on-one counseling services for trauma. > Increase screening for co-occurring disorders. There is a need for medical services that are acceptable and affordable that includes psychiatric, medical, and physical health services. In order for treatment of substance use to be successful, any co-occurring mental disorders must also be identified and treated. > Provide more outpatient services outside of regular business hours (e.g., in the evening and on weekends). This would make it easier for women who work during the day or care for children to attend treatment. > Increase skills among alcohol education counselors in relation to the moderation of group sessions. It is important that counselors are able to shut down difficult clients and ensure that the group is functional and constructive and that all participants are able to share their experiences. > Increase the availability of aftercare which is important to success following the completion of treatment and the maintenance of longterm sobriety. It is essential to offer these services which enable women to return for follow-up if they successfully completed treatment. The FEMALE DRUNK DRIVERS A QUALITATIVE STUDY 83 Recommendations

334 provision of follow-up services can also help women maintain abstinence and prevent relapse. Melissa came from an affluent family and was never in trouble as a teenager. By the time she was in her mid-20s, she had completed her undergraduate degree and was employed full-time as a high school teacher and looking to complete a Masters of Education. Then she was arrested for two DWIs within a six-month period. On both occasions, she had been celebrating accomplishments and had not considered herself to be impaired or unable to drive because she had been drinking. Her convictions meant she was forced to resign from her job and she entered an inpatient program for substance abuse. She found that the intense media coverage of her offenses made it more difficult for her to cope with her already challenging situation. There was tremendous stigmatization because she lived in a small community and was a well-known teacher. She says she found it difficult to manage after her convictions and could not have done it without the support of family and friends. Now re-located, she cautions that the significant costs and demands resulting from a DWI can make it difficult to succeed, even though she had means and was very motivated. I had to work multiple jobs to cover all of the costs and fees associated with my DWIs and to be compliant. Getting a job once you have a criminal record is really challenging. If I had known the consequences, I would never have driven. 84 Recommendations FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

335 6. CONCLUSIONS As evidenced by TIRF s 2011 review of the literature, much of the available research about female drunk drivers is outdated and several gaps in knowledge exist. What is known is that there are important differences between female drunk drivers and their male counterparts. In particular, they are often older at arrest, are more likely to be single, separated or divorced, have more education, and are often the primary caregivers of children. Mental health issues and prescription drug use is typically more pronounced, and histories of abuse, trauma, and health problems are common. In contrast, what is not well understood are the pathways to this behavior among women, the factors that contribute to it or that compound it, their experiences postconviction, or what strategies can best serve this population. This study adds to this body of knowledge and demonstrates that, in addition to issues identified above, offending behavior among a large majority of female drunk drivers appears to often be associated with a significant emotional event or trigger (e.g., intense pressure to succeed, financial problems, the end of a relationship, the illness of a child, death of a parent). A majority of women struggle with low self-esteem or depression and turn to alcohol use as a coping mechanism without understanding its effects. They further report that the stigma associated with a drinking problem not only compounds these problems but frequently discourages them from admitting it or seeking treatment. This is equally true in relation to other issues such as trauma and abuse. And, they are immensely attuned to how the acknowledgement of such problems will negatively influence the perceptions of family, the custody of children, and their professional status. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Conclusions 85

336 This study also identified three distinct and unique profiles of female drunk drivers that begin to shed light on the confluence of factors that play a role in their offending. Of importance, such profiles illustrate that women may be subject to different risks and needs, and that drinking and driving behavior may not emerge until later in life. This knowledge has important implications for prevention initiatives as well an interventions targeted towards this population. Once convicted of drunk driving, it appears that female offenders experience greater financial pressures as the sole breadwinner and caregiver of children, and they are often forced to choose between an unhealthy relationship and sobriety. They are frequently overwhelmed and unequipped to manage the demands of supervision in combination with ongoing parental, employment, and community responsibilities. They are more often reliant on emotional support from family members and friends, and require assistance with childcare, housing, and transportation. Moreover, women with fewer financial resources express significant safety concerns in relation to their limited transportation options. Perhaps most notably, these women frequently define their experiences in the criminal justice and treatment systems in terms of emotional consequences; shame, anxiety, frustration, anger, fear, and uncertainty. This study also revealed that there are important gaps in existing criminal justice and treatment systems that can make it more challenging for female drunk drivers to successfully complete their sentence and comply with the conditions imposed upon them. Practitioners report that they often lack knowledge about the risk/need dynamic of this specific population, and have limited understanding of addiction issues. They also note that the tools they frequently utilize are not designed to identify mental health issues or histories of trauma, and it can be challenging to gather sufficient information about the personal history of the women they supervise in order to best serve them. Existing supervision and scheduling protocols are generally less appropriate for females because of the competing responsibilities that they face. It is also apparent from this study that women often receive conflicting or confusing information from the different systems they are involved with (i.e., criminal justice, treatment, and licensing) regarding expectations, requirements, and consequences. This is perhaps most poignantly illustrated by the fact that many women resign themselves to not reinstating their 86 Conclusions FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

337 driving privileges until they have completed their supervision because they become so frustrated in trying to determine when they are in fact eligible and the process that they must complete. There are also important gaps in the interventions and services that are available to women. There is a greater demand for more specialized treatment services that are capable of addressing the complexity of issues that contribute to their drinking, however, commonly available treatment programs often do not meet their diverse needs. The integration of these interventions with relevant community and social services is also lacking and practitioners working in criminal justice and treatment systems are often less prepared to make referrals to outside agencies to help these women address mental health, counseling, housing, childcare, and employment issues. In summary, the results of this study clearly demonstrate the importance of focusing efforts to begin to better understand this problem and to develop more effective strategies both to prevent and manage it. Subsequently, more research about the characteristics of these women, what works with female drunk drivers in relation to the effectiveness of traditional sanctions, and how such strategies can be best implemented in existing criminal justice and treatment systems is needed. In particular, research to increase understanding of what supervision strategies lead to successful completion of probation and what components of treatment produce better outcomes can guide efforts to address this problem in the form of policies, programs, and individual interventions. In recognition of this, some agencies have already recognized the need to improve the delivery of services and to develop different approaches and strategies to address the needs of this specialized population. The results from this study can bolster such initiatives. As a first step in this regard, the results obtained from the interview focus groups, in-depth interviews, survey, and key informant interviews will be available to the APPA to inform the development of a training module on female drunk drivers for probation and court practitioners. Findings will be also be shared with the NCDC to help identify opportunities within the DWI Court model to better address the specific risks and needs of female drunk drivers in an effort to reduce recidivism. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY 87 Conclusions

338 Also of importance, this study provides insight that can inform future research initiatives. In particular, the data collected suggests opportunities for inquiry and further exploration into issues that pertain to the supervision and treatment of female drunk drivers. As the purpose of this study was to generate hypotheses as opposed to test them, potential areas for future research include: > Examination of which types of treatment interventions produce the best outcomes among female drunk drivers. > Comparison of whether outcomes are better among this population using gender-sensitive supervision and treatment approaches opposed to traditional approaches. > Examination of what impact the use of screening and assessment of first offenders followed by appropriate referrals to treatment services has on future recidivism. > Examination of the benefits of early intervention among young women and first DWI offenders in regards to alcohol consumption. > Further investigation of the profiles described in this report, and identification of what supervision strategies and treatment interventions produce the best outcomes among each of them. > Examination of the benefits of the inclusion of victimization and trauma training for probation officers who supervise female offender caseloads. > Determination of the magnitude of the prevalence of mental health and trauma issues among female drunk drivers and the identification of best practices to address these problems. > Measurement of the impact and effectiveness of Leandra s Law in New York on women in relation to outcomes. > Examination of the financial impact that drunk driving offenses have on female drunk drivers in comparison to males. 88 Conclusions FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

339 7. REFERENCES Argeriou, M., McCarty, D., Potter, D., & Holt, L. (1986). Characteristics of men and women arrested for driving under the influence of liquor. Alcoholism Treatment Quarterly, 3, Bloom, B., Owen, B., & Covington, S. (2003). Gender-Responsive Strategies: Research, Practice and Guiding Principles for Women Offenders. Washington, D.C.: National Institute of Corrections. Brady, K. & Randall, C. (1999). Gender differences in substance use disorders. Psychiatric Clinics of North America, 22, Caldwell-Aden, L., Kaczowka, M., & Balis, N. (2009). Preventing First-Time DWI Offenses. First-Time DWI Offenders in California, New York and Florida: An Analysis of Past Criminality and Associated Criminal Justice Interventions. DOT HS Washington, D.C.: U.S. Department of Transportation. California Department of Alcohol & Drug Programs. (2012). Fact Sheet: Driving Under the Influence Statistics Sacramento: Author. Carey, S., Allen, T., & Einspruch, E. (2012). San Joaquin DUI Monitoring Court Process and Outcome Evaluation. Available online: sites/default/files/reports/california-evaluation-1.pdf Center for Disease Control and Prevention (2013). Binge drinking: A serious, underrecognized problem among women and girls. CDC Vitalsigns, January Chalmers, D., Olenick, N.L., & Stein, W. (1993). Dispositional traits as risk in problem drinking. Journal of Substance Abuse, 5, Chang, I., Lapham, S.C., & Barton, K.J. (1996). Drinking environment and sociodemographic factors among DWI offenders. Journal of Studies on Alcohol, 57, Creswell, J.W. (2013). Qualitative Inquiry and Research Design. Choosing Among Five Approaches. Third Edition. Los Angeles: Sage. DeYoung, D. (2002). An Evaluation of the Implementation of Ignition Interlock in California. Sacramento: California Department of Motor Vehicles. DeYoung, D., Tashima, H., & Masten, S. (2005). An Evaluation of the Effectiveness of Ignition Interlock in California: Technical Report. Sacramento: California Department of Motor Vehicles Dowling, A. (2013). Alcohol-related Crashes and Fatalities, Impaired Driving Arrests, DWI Recidivism Rates. Presented at the New York State Fourth Annual Ignition Interlock Qualified Manufacturers Conference, Albany, New York, June 13, Drew, L., Royal, D., Moulton, B., Peterson, A., & Haddix, D. (2010). National Survey on Drinking and Driving Attitudes and Behaviors: Washington, D.C.: U.S. Department of Transportation. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY References 89

340 Federal Bureau of Investigation (FBI). (2012). Crime in the United States, 2011: Table 69, Arrests by State. Accessible online: Freeman, J., Maxwell, J.C., & Davey, J. (2011). Unraveling the complexity of driving while intoxicated: A study into the prevalence of psychiatric and substance abuse comorbidity. Accident Analysis and Prevention, 43(1), Gray, P., Williamson, J., Karp, D., & Dalphin, J. (2007). The Research Imagination. Cambridge: Cambridge University Press. Green, C.A. (2006). Gender and use of substance abuse treatment services. Alcohol Research & Health, 29(1), Greenfield, S., Brooks, A.J., Gordon, S.M., Green, C., Kropp, F., McHugh, K., Lincoln, M., Hien, D., & Miele, G.M. (2007). Substance abuse treatment entry, retention, and outcome in women: A review of the literature. Drug and Alcohol Dependence, 86(1), Grella, C.E., & Greenwell, L. (2004). Substance abuse treatment for women: Changes in settings where women received treatment and types of services provided, Journal of Behavioral Health Services and Research, 31(4), Gudrais, E. (2011). Women and Alcohol. Harvard Magazine, July-August, Jones, R.K., & Lacey, J.H. (2001). Alcohol and Highway Safety 2001: A Review of the State of Knowledge. DOT HS Washington, D.C.: U.S. Department of Transportation. Khan, M. (2011). California s Interlock Program. Presented at the 12th International Alcohol Interlock Symposium, Palm Springs, California, October 18th, Lapham, S.C., Skipper, B.J., Hunt, W.C., & Chang, I. (2000). Do risk factors for rearrest differ from female and male drunk-driving offenders? Alcoholism: Clinical and Experimental Research, 24(11), Laplante, D.A., Nelson, S.E., Odegaard, S.S., LaBrie, R.A., & Shaffer, H.J. (2008). Substance and psychiatric disorders among men and women repeat driving under the influence offenders who accepts a treatment-sentencing option. Journal of Studies on Alcohol and Drugs, 69(2), Lynskey, M.T., Bucholz, K.K., Madden, P.A.F., & Heath, A.C. (2007). Early-onset alcohol-use behaviors and subsequent alcohol-related driving risks in young women: A twin study. Journal of Studies on Alcohol and Drugs, 68(6), Maxfield, G., & Babbie, E. (2006). Basics of Research Methods for Criminal Justice and Criminology. Toronto: Thomson Wadsworth. Maxwell, J.C., & Freeman, J. (2007). Gender differences in DUI offenders in treatment in Texas. Traffic Injury Prevention, 8, References FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

341 Mayhew, D.R., Ferguson, S.A., Desmond, K.J., & Simpson, H.M. (2003). Trends in fatal crashes involving female drivers, Accident Analysis and Prevention, 35(3), McDavid, J., & Hawthorn, L. (2006). Program Evaluation and Performance Measurement: An Introduction to Practice. Thousand Oaks: Sage Publications. McMurran, K., Riesman, R., Manning, N., Misso, K., & Kleijnen, J. (2011). Interventions for alcohol-related offending by women: A systematic review. Clinical Psychology Review, 31, Michigan Department of State Police. (2013) Michigan Annual Drunk Driving Audit. Available online: updated_052013_425487_7.pdf National Highway Traffic Safety Administration (NHTSA). (2009). Alcohol-Impaired Drivers Involved in Fatal Crashes, by Gender and State, Traffic Safety Facts DOT HS Washington, D.C.: U.S. Department of Transportation. National Highway Traffic Safety Administration (NHTSA). (2012). Traffic Safety Facts 2011 Data: Alcohol-Impaired Driving. DOT HS Washington, D.C.: U.S. Department of Transportation. New York State Division of Criminal Justice Services (DCJS). (2013). Aggravated DWI with a Child Cases. Provided by DCJS with data through May 21, Peck, R.C., Gebers, M.A., Voas, R.B., & Romano, E. (2008). The relationship between blood alcohol concentration (BAC), age, and crash risk. Journal of Safety Research, 39, Popkin, C.L. (1991). Drinking and driving by young females. Accident Analysis and Prevention, 23(1), Rauch, W.J., Zador, P.L., Ahlin, E.M., Howard, J.M., Frissell, K.C., & Duncan, G.D. (2010). Risk of alcohol-impaired driving recidivism among first offenders and multiple offenders. American Journal of Public Health, 100(5), Robertson, R.D., Vanlaar, W, Simpson, H., Boase, P. (2009). Results from a national survey of Crown prosecutors and defense counsel on impaired driving in Canada: A System Improvements perspective. Journal of Safety Research, 40, Robertson, R.D., & Simpson, H.M. (2002a). DWI System Improvements for Dealing with Hard Core Drinking Drivers: Prosecution. Ottawa: Traffic Injury Research Foundation. Robertson, R.D., & Simpson, H.M. (2002b). DWI System Improvements for Dealing with Hard Core Drinking Drivers: Adjudication and Sanctioning. Ottawa: Traffic Injury Research Foundation. Robertson, R.D., & Simpson, H.M. (2003a). DWI System Improvements for Dealing with Hard Core Drinking Drivers: Monitoring. Ottawa: Traffic Injury Research Foundation. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY References 91

342 Robertson, R.D., & Simpson, H.M. (2003b). DWI System Improvements: Stopping the Revolving Door. Ottawa: Traffic Injury Research Foundation. Royal, D. (2003). National Survey on Drinking and Driving Attitudes and Behaviors: Washington, D.C.: U.S. Department of Transportation. Schwartz, J. & Steffensmeier, D. (2007). The Nature of Female Offending: Patterns and Explanation. In: R. Zaplin (Ed.), Female Offenders: Critical Perspective and Effective Interventions. (pp ). Boston: Jones & Bartlett. Schwartz, J., & Rookey, B. D. (2008). The narrowing gender gap in arrests: Assessing competing explanations using self-report, traffic fatality, and official data on drunk driving, Criminology, 46(3), Shaffer, H.J., Nelson, S.E., LaPlante, D.A., LaBrie, R.A., & Albanese, M. (2007). The epidemiology of psychiatric disorders among repeat DUI offenders accepting a treatment-sentencing option. Journal of Consulting and Clinical Psychology, 75(5), Shore, E.R., & McCoy, M.L. (1987). Recidivism among female DUI offenders in a Midwestern American city. Journal of Criminal Justice, 15(5), Simpson, H. M., Beirness, D. J., Robertson, R. D., & Hedlund, J. H. (2004). Hard core drinking drivers. Traffic Injury Prevention, 5(3), Substance Abuse and Mental Health Services Administration (SAMHSA) (2005). Substance abuse treatment for adults in the justice system: A treatment improvement protocol TIP 44. U.S. Department of Health and Human Services. Center for Substance Abuse Treatment. Sun, A-P. (2006). Program factors related to women s substance abuse treatment retention and other outcomes: A review and critique. Journal of Substance Abuse Treatment, 30, Tsai, V.W., Anderson, C.L., & Vaca, F.E. (2010). Alcohol involvement among young female drivers in US fatal crashes: Unfavorable trends. Injury Prevention, 16, U.S. Census Bureau (2012). State and County QuickFacts. Retrieved from: quickfacts.census.gov/qfd/index.html Webster, J.M. Pimentel J.H., Harp, K.L.H., Clark, D.B., & Staton-Tindall, M. (2009). Substance abuse problem severity among rutral and urban female DUI offenders. American Journal of Drug & Alcohol Abuse, 35(1), Wells-Parker, E., Pang, M.G., Anderson, B.J., McMillen, D.L., & Miller, D.I. (1991). Female DUI offenders: A comparison to male counterparts and an examination of the effects of intervention on women s recidivism rates. Journal of Studies on Alcohol, 52(2), White, W. & Hennessey, M. (2006). Evaluating, Treating and Monitoring the Female DUI Offender. 92 References FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

343 Williams, A. F., McCartt, A. T., & Ferguson, S. A. (2007). Hardcore Drinking Drivers and Other Contributors to the Alcohol-impaired Driving Problem: Need for a Comprehensive Approach. Traffic Injury Prevention, 8(1), Wilsnack, R.W., Wilsnack, S.C., & Klassen, A.D. (1984). Women s drinking and drinking problems: Patterns from a 1981 national survey. American Journal of Public Health, 74(11), Zador, P.L., Krawchuck, S.A., & Voas, R.B. (2000). Alcohol-related relative risk of driver fatalities and driver involvement in fatal crashes in relation to driver age and gender. Journal of Studies on Alcohol, 61, FEMALE DRUNK DRIVERS A QUALITATIVE STUDY References 93

344 8. APPENDIX - CASE STUDIES 8.1 Jurisdictional profile: California DWI statistics > Progress has been made in California in reducing alcohol-impaired fatalities. Between 2009 and 2010, the number of fatalities decreased by 14.4% from 924 to 791 and continued to decrease in Of the five states with the greatest number of total traffic fatalities, California has the best alcohol-impaired driving fatality rate. In 2011, there were 774 fatalities in crashes involving a driver with a BAC of.08 or higher in the state of California. > In 2010, there were 195,879 DUI arrests in the state. Females comprised 22.4% of these arrests. The proportion of females among convicted DUI offenders has risen consistently every year since 1989 (California Department of Alcohol & Drug Programs 2012). > California maintains a high DUI conviction rate of approximately 79%. State impaired driving laws Driving Under the Influence (DUI). A person commits the crime of DUI if they are under the influence of any alcoholic beverage or drug, or under the combined influence of any alcoholic beverage and drug, and drives a motor vehicle. > 1 st offense: fine of not less than $390 and not more than $1,000; imprisonment in county jail of no less than 96 hours (at least 48 of which must be continuous) and not in excess of six months. > 2 nd offense: fine of not less than $390 and not more than $1,000; imprisonment in county jail of no less than 90 days and not in excess of one year. > 3 rd offense: fine of not less than $390 and not more than $1,000; imprisonment in county jail of no less than 120 days and not in excess of one year. > Habitual offenders must be convicted of three or more DUIs and the designation is applied for a period of three years. 94 Appendix FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

345 Driving with prohibited blood alcohol concentration. It is unlawful for any person who has.08 percent or more, by weight, of alcohol in his or her blood to drive a vehicle. Reckless driving involving alcohol. DUIs in California are often reduced to the lesser charge of reckless driving involving alcohol, commonly referred to as wet reckless. In order to be eligible for the reduction, offenders must be willing to plead guilty and not have any prior DUI convictions. Penalties include a $1,000 fine and mandatory completion of a DUI education program. Programs Supervision. Female DUI offenders can be processed through either traditional courts or DWI/Drug Courts. California has 99 Adult Drug Courts and 9 DWI Courts. If a first offender is granted probation they will be granted summary probation for a period of three years, which is generally the maximum period that can be imposed. While there are various terms and conditions attached to a sentence of summary probation (including participation in a DUI education program), there is a lack of active supervision by either a probation officer or the court. This generally holds true for second and even third time offenders, because under California law, unless the DUI driver causes injury to another person, a non-injury DUI cannot be charged as a felony until and unless offenders have sustained three prior DUI convictions within a ten year period before the pending (fourth) arrest. Once offenders are convicted of a felony offense, they must be placed on formal probation, because by statute, summary probation is not available as a sentencing option for felony offenders. It is at this point that they will be actively supervised by a probation officer and/or the court. Interlock program. California has a hybrid interlock program (judicial and administrative components) that has been in place since 1986 and was implemented statewide in Recently, the program has undergone a number of legislative changes culminating in a pilot initiative in four counties for first offenders launched in The courts are responsible for ordering the interlock device as part of sentencing for impaired driving offenders. The interlock legislation is mandatory for repeat offenders and discretionary for first offenders. The court has the general authority (as per Sec of the Vehicle Code) to FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Appendix 95

346 order the installation of an interlock on any DUI conviction and is to give heightened consideration in cases of high BACs, test refusals, or to offenders with two or more prior traffic violations. This statute is not mandatory per se and relies heavily upon judicial discretion to order the device. There are mandatory interlock provisions for offenders who are caught driving while suspended or revoked for DUI. For example, offenders convicted for a 2 nd DUI serve a two-year suspension but can get an interlock-restricted license after 90 days permitted that they meet all eligibility criteria (e.g., proof of DUI education, financial responsibility, and interlock installation). The interlock is ordered infrequently and not uniformly applied as a sanction across the state; judges in some counties tend to order the device with greater consistency than others. The monitoring of interlock offenders is the responsibility of probation officers but only if those individuals are being actively supervised (many offenders are on paper probation and do not regularly meet with a probation officer). Several evaluations of California s interlock program have found that more first offenders participate in the program than repeat offenders (DeYoung 2002; DeYoung et al. 2005). Reinstatement rates after three to four years are also higher for first offenders than repeat offenders (75% vs. 50%). In an effort to further address the impaired driving problem in the state, California recently introduced a pilot program (see Sec of the Vehicle Code) in four counties (Alameda, Los Angeles, Sacramento, and Tulare) for first offenders. As of July 1st, 2010, interlocks are required for all DUI offenses both first-time and repeat. First offenders are required to have the device installed for a period of five months. Repeat offenders are required to have the device installed for periods of 12, 24, or 36 months depending on the number of prior DUI convictions. The results of this pilot are due to be presented to the legislature in At present, the participation rate for the program is 20-25% (Khan 2011). Treatment. A common condition of probation for DUI offenders is the enrollment in and completion of a driving under the influence program that is approved by the State Department of Alcohol and Drug Programs (ADP). The ADP currently licenses 472 programs that fall into four different categories: 96 Appendix FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

347 > Wet Reckless Program. This program is designed for offenders convicted of reckless driving with a measurable amount of alcohol in their blood and involves 12 hours of alcohol education. > First Offender Program. This program is designed for offenders convicted of their first DUI offense; they must complete a state-licensed threemonth or nine-month program, depending on their BAC (.2 or higher requires participation in the nine-month program). These programs involve alcohol and drug education and counseling. > 18-Month Program. This program is designed for offenders convicted of their second DUI. This program involves 52 hours of group counseling, 12 hours of alcohol and drug education, six hours of community re-entry monitoring, and bi-weekly individual interviews. > 30-Month Program. This program is designed for offenders convicted of their 3 rd and subsequent DUI. This program is only available in select counties and involves 78 hours of group counseling, 12 hours of alcohol and drug education, hours of community service, and regular individual interviews. These programs are designed to enable participants to consider attitudes and behavior, support positive lifestyle changes, and reduce or eliminate the use of alcohol and/or drugs. Program components include education, group counseling, individual interview sessions, and treatment. Other treatment components may be required as a condition of probation. Offenders could be required to complete inpatient or residential treatment, the quality and availability of which would be dictated on a county-bycounty basis. Female DUI offenders also routinely participate in Alcoholics Anonymous; gender-specific meetings are available but again, this depends on the jurisdiction. County selected The county chosen as a site for interview focus groups in California was San Joaquin County. According to the 2010 U.S. Census, the county has a population of 685,306 (480/mile 2 ). In 2008, San Joaquin County implemented a system change whereby all repeat DUI offenders in the largest judicial district (City of Stockton) were required to participate in a DUI Monitoring Court program. The program is comprised of two different tracks that identify varying levels of risk among this offender population: FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Appendix 97

348 > Track 1 is a monitoring track in which offenders are required to come to court infrequently and report on progress (e.g., completion of terms of probation, re-licensing). > Track 2 is designed for offenders who are unable to comply with Track 1 requirements and are assessed as needing substance abuse treatment. Track 2 adheres to a traditional Drug Court model. Participation in the program lasts a minimum of 12 months and participants must be sober for a period of 120 days before being permitted to exit. As part of the program, probation officers assess and re-assess offenders every six months to identify any changes in risk and needs. The assessment instrument that practitioners rely upon is the Static Risk and Offender Needs Guide (STRONG) which is a comprehensive, fourth generation risk and needs assessment tool as well as an automated case planning system. 8 A 2011 evaluation by NPC Research found the program to be effective. Participants were less likely than a comparison group of DUI offenders on traditional probation to be re-arrested after 18 months (9% vs. 12%). To access a copy of the evaluation, please refer to: org/sites/default/files/reports/california-evaluation-1.pdf Alternative Work Program (AWP). The AWP is a community corrections program in San Joaquin County that allows individuals who are sentenced to jail to serve their time living at home and working within the community. Participants are assigned to work 8-10 hours daily on public works or for non-profit organizations. Each day worked in the community counts as two days served in jail. The program is viewed as beneficial because it allows offenders to maintain employment and family ties, avoid incarceration, and contribute to society. Profile of offenders > There were 30 female DUI offenders that participated in the California interview focus groups. Approximately three-quarters of these women were repeat offenders and one-quarter were first offenders. > The majority of the women who participated in the interviewfocus groups were estimated to be between the ages of 20 and For more information about STRONG, please refer to: assessments_documentation/strong%20fact%20sheet.pdf 98 Appendix FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

349 > In comparison to the other states involved in the study, California offenders tended to be of lower socioeconomic status. > There was greater cultural diversity among female DUI offenders in California as compared to other jurisdictions. In particular, there was a much larger Hispanic population represented. > Many of the female offenders were placed on continuous alcohol monitoring (i.e., SCRAM) as a condition of their probation. > Many of the women had the following issues present in their history as confirmed by the participants themselves and through the observations of practitioners:» broken/dysfunctional household;» mental health symptoms and/or diagnoses;» polysubstance use;» sexual abuse/assault;» domestic violence/abuse; and,» difficulty sleeping. Profile of practitioners > There were five probation officers, three alcohol education program managers, two treatment providers, two defense attorneys, and one judge that participated in practitioner focus groups in California. A majority of these practitioners had considerable experience in their respective professions, although not all of them. Alcohol education and treatment professionals had considerable experience > The DUI Monitoring Court Program has a specialized felony DUI unit that allows for intensive supervised probation of offenders. The unit is funded through a grant from the California Office of Traffic Safety. > Probation officers who supervised female DUI offenders were male and female and many of them had mixed caseloads. Practitioners noted that there were many more females working in the criminal justice system as probation officers, prosecutors, and defense attorneys. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Appendix 99

350 > Some of the officers reported using an alternating schedule in which offenders come to the office for appointments for a period of two weeks followed by two weeks of home visits. All offenders are required to submit to random breath testing and urinalysis during both office appointments and home visits. > There are no special protocols in place for the supervision of female DUI offenders compared to males. > Probation officers attempt to accommodate offenders who work and some of them keep flexible hours to allow probationers to come for appointments and/or testing in the evening or on weekends. > Practitioners do make an effort to refer women to appropriate services and programming whenever they have the opportunity to do so. For example, pregnant women are referred to the Healthy Connections Program. 8.2 Jurisdictional profile: Missouri DWI statistics > In 2011, there were 258 fatalities in crashes involving a driver with a BAC of.08 or higher in the state of Missouri (NHTSA 2012). > In 2011, there were 29,447 impaired driving arrests in the state (FBI 2012). State impaired driving laws Driving While Intoxicated (DWI). A person commits the crime of DWI if they operate a motor vehicle while in an intoxicated or drugged condition. The statute is not limited to impairment by alcohol and includes prescription drugs. > 1 st offense: Class B misdemeanor; up to $500 fine; up to six months in jail. > 2 nd offense (within five years): Class A misdemeanor; up to $1,000 fine; up to one year in jail. > 3 rd offense: Class D felony; up to $5,000 fine; up to four years in prison. > 4 th offense: Class C felony; $5,000 fine; up to seven years in prison. 100 Appendix FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

351 > 5 th or subsequent offense: Class B felony; minimum $5,000 fine; prison sentence of not less than five years to a maximum of 15 years. Driving with excessive blood alcohol content (BAC). A person commits the crime of driving with excessive blood alcohol content if they operate a motor vehicle with a BAC of.08 or higher. The offense, also known as BAC, is commonly thought of as a lesser offense than DWI however, a reduction of a DWI to a BAC will generally not lessen the impact of a conviction or sanctions. The court can require, as a condition of probation, that any offender convicted of an intoxication-related traffic offense abstain from consuming or using alcohol as demonstrated by continuous alcohol monitoring or by verifiable breath alcohol testing performed a minimum of four times per day as scheduled by the court for a minimum period of 90 days. Aggravated, chronic, persistent, and prior offenders. Missouri law further identifies different categories of repeat impaired driving offenders and the associated penalties for each category are summarized below. Courts are not permitted to suspend the imposition of sentences for aggravated, chronic, persistent, or prior offenders. These offenders are also not eligible to pay fines in lieu of imprisonment. > Aggravated offender. A person who: 1) has pled guilty or has been found guilty of three or more intoxication-related traffic offenses; or, 2) has pled guilty or has been found guilty of intoxicated driving leading to death or bodily injury. Aggravated offenders are not eligible for parole or probation until they serve a minimum of 60 days imprisonment. > Chronic offender. A person who: 1) has pled guilty or has been found guilty of four or more intoxication-related traffic offenses; or, 2) has pled guilty or has been found guilty on two or more separate occasions of intoxicated driving leading to death or bodily injury; or, 3) has pled guilty or has been found guilty of two or more intoxication-related offenses and, in addition, has been found guilty of intoxicated driving leading to death or bodily injury. Chronic offenders are not eligible for parole or probation until they serve a minimum of two years imprisonment. > Persistent offender. A person who: 1) has pled guilty or been found guilty of two or more intoxication-related traffic offenses; or, 2) has pled guilty or been found guilty of involuntary manslaughter, assault in the FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Appendix 101

352 second degree, or assault of a law enforcement officer in the second degree related to an intoxication traffic offense. Persistent offenders are not eligible for parole or probation until they serve a minimum of 30 days imprisonment unless they perform at least 60 days (480 hours) of community service or participate in and successfully complete a courtordered treatment program under the supervision of the court. > Prior offender. A person who has pled guilty to or has been found guilty of one intoxication-related traffic offense, where such prior offense occurred within five years of the occurrence of the intoxication-related traffic offense for which the person is charged. Prior offenders are not eligible for parole or probation until they serve a minimum of 10 days imprisonment unless they perform at least 30 days (240 hours) of community service or participate in and successfully complete a courtordered treatment program under the supervision of the court. Programs Supervision. Female DWI offenders can be processed through either traditional courts of DWI/Drug Courts. Missouri has a total of 38 hybrid DWI/ Drug Courts, 37 Adult Drug Courts, and 20 DWI Courts. In Missouri, a DWI offense is classified as a misdemeanor which is a non-supervisable offense. Only felony offenders (three or more offenses) are actively monitored on probation. Interlock program. Missouri s interlock program was originally a court-based program but has since evolved to become a hybrid program. The administrative component was recently added when completion of the interlock program became a condition of license reinstatement. Program participation rates have steadily increased as a result of this structural change. In 2012, there were 7,500 interlocks installed in the state. First-time offenders are not eligible to enter into the interlock program. There are rare cases where judges will mandate a first offender into the interlock program but these are usually reserved for instances where offenders were either underage or had a very high BAC. For repeat DWI offenders, participation in the interlock program is mandatory. Judges are required to order their participation (although this is not always done) and it is also a condition of license reinstatement. The courts have the ability to respond to violations and can schedule a hearing and possibly revoke probation 102 Appendix FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

353 and/or apply sanctions such as a weekend in jail. The court also has the authority to extend interlock program participation for those offenders who continually demonstrate an inability to comply with program rules. However, court-based usage of interlocks is still relatively low although there has been a concerted effort to deliver education to criminal justice practitioners to improve program participation. Treatment. In Missouri, successful completion of the Substance Abuse Traffic Offender Program (SATOP) is a condition of full license reinstatement. This education and treatment program is overseen by the Division of Drug and Alcohol Abuse in the Department of Mental Health (DMH). The program serves more than 30,000 DWI offenders annually who are referred as a result of an administrative license suspension or revocation, court order, condition of probation, or plea bargain. All SATOP clients undergo an assessment and may receive a referral to different levels of treatment as a result. Interventions range from 10 hour education courses to 50 hours of outpatient counseling and residential treatment interventions. Similar to other jurisdictions, female DWI offenders are required to complete some form of treatment as a condition of their probation which could include either intensive inpatient or outpatient programming. Services are limited in rural jurisdictions and, as a result, some women have few options available. Attendance at support group meetings such as Alcoholics Anonymous is common and encouraged. County selected The county chosen as a site for interview focus groups in Missouri was Greene County. It is the fourth most populous county in the state and according to the 2010 U.S. Census, the population is 275,174 (408/mile 2 ). Greene County has an Academy DWI Court, meaning it is a model for other courts in the nation. The Greene County DWI Court was started in 2004, and is located in Springfield, Missouri. This DWI Court only accepts felony DWI cases with a program capacity of 115 individuals with over 125 graduates. The court uses a team approach whereby all members work together collaboratively in an effort to hold offenders accountable and protect public safety while also promoting rehabilitation. The mission of the Greene County DWI Court is to promote public safety by expediting the time interval to get felony DWI offenders into accountability and treatment quickly and to keep the FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Appendix 103

354 felony DWI offender engaged in treatment long enough to receive treatment benefits. Agencies represented as part of the DWI Court team include the judge, prosecutor s office, members of the defense bar, probation and parole, Sigma House 9 (treatment provider), and private contractors who provide mental health and case management services. In addition, other specialized treatment and female-only groups used to be offered more frequently in Greene County but they are no longer available to the same degree that they were previously. There are four phases to participation. Upon successful completion of each of these phases, offenders are eligible to graduate from the court. > Phase 1 includes weekly court attendance for a minimum of three months; 10:30pm curfew; completion of a naltrexone screening; must have 30 consecutive days of sobriety immediately prior to advancing to the next phase. > Phase 2 includes court attendance every two weeks for a minimum of four months and during that time the offender identifies family issues; maintenance of employment; completion of the pre-test for the GED; must have 60 consecutive days of sobriety immediately prior to moving to the next phase. > Phase 3 includes court attendance every four weeks for a minimum of five months; development of an aftercare plan; begin community service; must have 90 consecutive days of sobriety prior to advancing to the next phase. > Phase 4 includes court attendance every six weeks for a minimum of six months; development of a relapse prevention plan; maintenance of stable housing and employment; completion of 60 hours of community service; must maintain six months of sobriety in order to graduate. 9 Sigma House, also referred to as Clarity Recovery & Wellness is a non-profit corporation that provides full-spectrum drug and alcohol rehabilitation for male and female clients over age 18 at four locations in the state; both residential and outpatient services are offered and the length of treatment is based on client needs. The treatment programs are abstinence-based and incorporate the 12-Step model, individual counseling, group therapy, 12 Step meetings, and activities designed to lead to physical, emotional, social, and spiritual wellness. Weekly family groups are also offered to help loved ones understand the impact of addiction and the treatment process. The cost for both residential and outpatient treatment is affordable, and those without insurance coverage are eligible for sliding fee scale programs. 104 Appendix FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

355 At all times during the phases, court participants are expected to comply with court orders, demonstrate consistency in attending treatment, keep all appointments, and submit to random drug testing from three times a week in Phase 1 to once a week in Phase 4 or as directed. Profile of offenders > There were 33 female DWI offenders that participated in the Missouri interview focus groups. Approximately three-quarters were repeat offenders and one-quarter were first offenders. > It is estimated that the women ranged in age from 22 to 59. > Double digit BACs of.2 or higher were common and some of the women who had higher BACs were involved in crashes with most of these involving property damage only; a minority involved injury and/or death. > There was a wide cross-section of socioeconomic status and education levels represented among interview focus group participants. > Reported illicit drug use, particularly methamphetamines and marijuana, was higher among women in Missouri as compared to other jurisdictions in the study. > Transportation (i.e., finding alternative transportation options) was a significant problem in Missouri. Several women reported that they were forced to move from rural to more urban areas in order to comply with supervision requirements (e.g., attendance of court, treatment, testing, and other appointments) because they had no driver s license. These women reported spending considerable sums of money on taxi cab fares. > Many of the women had the following issues present in their history as confirmed by the participants themselves and through the observations of practitioners:» broken/dysfunctional household;» mental health symptoms and/or diagnoses of depression, anxiety or other disorders;» prescription drug use and/or polysubstance use;» sexual abuse/assault;» domestic violence/abuse; FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Appendix 105

356 » difficulty sleeping;» unhealthy relationships and/or a partner with a substance use problem; and,» other criminal offenses (e.g., manufacturing/distribution of methamphetamines and fraud/writing bad checks/shoplifting). Profile of practitioners > There were four probation officers who participated in a Missouri focus group. The level of experience of these probation officers varied as one had less than two years experience whereas another had eight years experience. One of the officers had previously worked as a social worker for 20 years. > Probation officers were both male and female and had general mixed caseloads and there were no specialized caseloads. > There is no specific or gender-sensitive training available to practitioners who supervise female DWI offenders although practitioners acknowledged that this would be beneficial. > The scheduling of team meetings is considered important within the context of the DWI Court. Practitioners meet regularly to review cases, share information and determine appropriate action plans, courses of treatment, and sanctions when necessary. 8.3 Jurisdictional profile: New York DWI statistics > In 2011, there were 315 fatalities in crashes involving a driver with a BAC of.08 or higher in the state of New York (NHTSA 2012). > Over a five year span ( ), the percentage of female drivers involved in alcohol-related crashes in New York ranged from 21-24% (Dowling 2013). > In 2011, there were 35,541 impaired driving arrests in the state (FBI 2012). > Over a five year span ( ), the percentage of females arrested for impaired driving in New York increased by 1% each year from 18% in 2007 to 22% in In the last 13 years, the percentage of female 106 Appendix FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

357 DWI recidivist drivers has also increased with women accounting for 12% of recidivists in 1999, 16% in 2009 and 18% in 2012 (Dowling 2013). > In 2009, 23,150 individuals were sentenced for felony and misdemeanor DWI convictions. Most offenders (13,140) either paid a fine, paid a fine and had their license suspended, or paid a fine with a conditional discharge. 5,072 offenders received a probation sentence, and another 2,095 received a combination of jail and probation. 2,128 offenders went to jail, and 514 went to prison (NYS DCJS 2010). Periods of probation supervision for a misdemeanor DWI are three years and five years for a felony DWI. Not all convictions result in a term of probation. > Since Leandra s Law took effect on December 18 th, 2009, a total of 2,932 individuals have been arrested (as of May 21 st, 2013). In this same period, 1,042 individuals were convicted of Aggravated DWI with a Child under Leandra s Law (resulting in 1,939 total sentences). Of those arrested under Leandra s Law, 1,081 were female. Aggravated DWI with a child cases 2009* ** TOTAL January February March April May June July August September October November December TOTAL Note: Includes cases where an Aggravated DWI With a Child charge does not appear as a charge at arrest. * Law became effective on December 18 th, ** Through May 21 st, Source: DCJS. Computerized Criminal History system (as of 05/21/2013). FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Appendix 107

358 > Between August 10 th, 2010 and December 31 st, 2012, a total of 12,055 interlocks have been installed in New York which translates to an installation rate of 29.3%. State impaired driving laws Driving While Ability Impaired (DWAI). A BAC more than.05 up to.07; first offense is a traffic infraction as is a second offense within five years. A third or subsequent offense within a ten year period is a misdemeanor. > 1 st offense: $300-$500 fine; up to 15 days in jail; 90-day license suspension. > 2 nd offense: $500-$750 fine; up to 30 days in jail; minimum six month license revocation. > 3 rd and subsequent offense: $750-$1,500 fine; up to 180 days in jail; minimum six month license revocation. Driving While Impaired (DWI). A BAC of.08 and higher; first offense is a misdemeanor and repeat offenses within a ten year period are a felony. > 1 st offense: fine of $500-$1,000; jail sentence of up to one year; minimum six month license revocation. > 2 nd offense: fine of $1,000-$5,000; jail sentence of up to four years; minimum one year license revocation. > 3 rd and subsequent offense: fine of $2,000-$10,000; jail sentence of up to seven years; minimum one year license revocation. Penalties are the same for Driving while impaired for drug (DWAI-Drug) and Driving while impaired by combined alcohol and drug (DWAI-Combination). Aggravated Driving While Impaired (AGG-DWI). A BAC of.18 and higher; first offense is a misdemeanor and repeat offenses within a ten year period are a felony. Penalties are as follows: > 1 st offense: fine of $1,000-$2,500; jail sentence of up to one year; minimum one year license revocation. > 2 nd offense: fine of $1,000-$5,000; jail sentence of up to four years; minimum 18 month license revocation. > 3 rd and subsequent offense: fine of $2,000-$10,000; jail sentence of up to seven years; minimum 18 month license revocation. 108 Appendix FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

359 Leandra s Law. This recently passed legislation (2009) amended the New York Vehicle and Traffic Law and the Penal Law to establish a new Class E felony to DWI when a child (under age 16) is a passenger in the vehicle at the time of arrest. The law also requires that all offenders convicted of both misdemeanor and felony DWI install an interlock for a minimum of six months. Programs Supervision. Female DWI offenders can be processed through either traditional courts or DWI/Drug Courts. New York has 75 hybrid DWI/Drug Courts. Interlock program. New York s interlock program is court-based and the Department of Criminal Justice Services is the designated program authority. Interlock legislation was first passed in 1992 and in 2007 a statewide multiyear pilot interlock program involving seven counties with post-revocation interlock installations was expanded. In 2009, Leandra s Law was implemented which made interlocks mandatory for all individuals convicted of DWI. The law went into effect on August 15th, 2010 and has resulted in significant growth in the state s interlock participation numbers. Prior to Leandra s Law, 10% of DWI convictions resulted in an interlock sanction and now they are a mandated condition of sentencing although court compliance varies across counties. Interlocks are now required for first and repeat offenders. It is mandatory on all vehicles owned or operated by persons convicted of misdemeanor and felony DWI offenses for a minimum period of six months. Treatment. Most female DWI offenders are required to complete some form of treatment as a condition of probation. The availability of services depends on the jurisdiction as rural counties afford offenders fewer options. Variety in treatment options is also dictated by whether the offender has insurance. If they do not have insurance, they typically opt for the cheapest available option. Available services run the gamut from intensive inpatient programs to outpatient therapy. Treatment programs often have lengthy wait times for admission, particularly for inpatient or residential programs. Treatment programming can take the form of individual counseling, group therapy, or a combination of both. Mixed gender group therapy is most commonly available whereas female-only group therapy is offered in fewer locations. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Appendix 109

360 Many female offenders also attend Alcoholics Anonymous meetings either on a voluntary basis or as a condition of probation. Counties selected The counties chosen as focus group sites varied in terms of population/ population density. The following statistics are derived from 2010 U.S. Census data: > Dutchess County (Poughkeepsie) - 297,488 (371 people/mile 2 ) > Warren County (Lake George) - 65,707 (75 people/mile 2 ) > Westchester County (White Plains) - 949,113 (2,193 people/mile 2 ) In general, services such as treatment and specialized programs were much more limited in rural jurisdictions such as Dutchess and Warren Counties as compared to metropolitan areas such as Westchester County. The Dutchess County Office of Probation and Community Corrections offers traditional probation supervision for female DWI offenders. There is no specialized programming or DWI/Drug Court available to this offender population. In Dutchess County, the installation rate for ignition interlocks is 13% (probation as monitor) and 51% (Drinking Driver Program as monitor). All of the female offenders who participated in the interview focus groups in Warren County were Drug Court probationers. This program involves specialized intensive supervision of high-risk felony offenders who have alcohol or drug abuse issues. Participation in Drug Court is offered in lieu of local jail or state prison. In Warren County, the installation rate for ignition interlocks is 15% (probation as monitor) and 29% (District Attorney as monitor). Gender-specific AA groups are available in Warren County. The Westchester County Department of Probation oversees the DWI Enforcement Program. In Westchester County, 20-25% of all offenders placed on probation are supervised for a DWI or DWI-related offense. As a result, three units (which consist of 20 specialized probation officers and three supervisors each), were created to supervise all DWI cases in the county. Offenders in the program are subject to strict supervision and are required to complete treatment and participate in Victim Impact Panels as a condition of their probation. According to an evaluation of intensive supervision programs for DWI offenders conducted by NHTSA, probationers in the Westchester County DWI Enforcement Program had an 18.1% lower recidivism rate 110 Appendix FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

361 compared to offenders not in the program. In Westchester County, the installation rate for ignition interlocks is 37% and probation is the program monitor. Gender-specific AA groups are available in Westchester County. Profile of offenders > There were 91 female DWI offenders that participated in the New York interview focus groups. An estimated one-third of these were first offenders and two-thirds were repeat offenders. > It is estimated that the women ranged in age from 19 to 65. > Several participants had felony DWI convictions, some resulting from having a minor in the vehicle at the time of arrest. > Double digit BACs of.16 or higher were common. > In comparison to the other states involved in the study, New York offenders tended to be older and have more DWIs in their history. > There was a wider cross-section of socioeconomic status and education levels represented among interview focus group participants with more women holding white collar jobs or being financially independent. > Many of the women had the following issues present in their history as confirmed by the participants themselves and through the observations of practitioners:» broken/dysfunctional household;» mental health symptoms and/or diagnoses;» sexual abuse/assault;» domestic violence/abuse;» difficulty sleeping;» lost parent to disability or death as a child;» lost sibling due to violent death, accident, or illness; and,» estranged from parent who abandoned the household. Profile of practitioners > 13 probation officers (both frontline practitioners and supervisors) participated in the New York focus groups. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Appendix 111

362 > Probation officers who participated in focus groups had varying levels of experience (some had in excess of fifteen years and others had less than three years of experience). > In Westchester County and Dutchess County, some probation officers had a specialized caseload and only supervised female offenders. The majority, however, supervised mixed caseloads that were not specific to DWI offenders or either sex. > Probation officers who worked with female DWI offenders were male and female although an effort was made in Dutchess County to have a female probation officer supervise most female DWI offenders. > In Westchester County, it is a common practice to review offender case files and determine which probation officer will best suit the offender s needs (i.e., background and specialized training). This is easily facilitated due to the large size of the probation department. > Probation officers recognize the importance of having training to deal with mental health issues and a history of trauma as it is recognized that these are prevalent amongst the female DWI offender population. > Practitioners commonly move on from supervising a female-only caseload because of the potential for burnout due to the emotional demands of female probationers. The Westchester County Forward Motion Program A growing number of female DWI offenders in combination with several high profile fatal collisions involving female drunk drivers in New York sparked the creation of the Forward Motion Program. The program incorporates treatment, educational and other social services. Features include close monitoring, immediate sanctions for non-compliance, mentorship to assist offenders in the development of short and long-term goals, and assistance to help them identify and utilize community resources to achieve their goals. The program seeks to stabilize addictions, improve education, provide vocational assistance, improve social skills, provide targeted interventions, increase positive reinforcement, and enhance intrinsic motivation in order to improve overall functioning. Program participants must be between the ages of 21-45, be a Level I or Level II offender, have needs that outweigh risks (as determined by the COMPAS Risk and Needs Assessment System), and be stable in their sobriety. The focus of the program is on the achievement of at least one of three goals that the women identify during the course of participation (e.g., completion of GED, attainment of stable employment, home ownership). For more information see: Appendix FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

363 8.4 Jurisdictional profile: Michigan DWI statistics > In 2011, there were 255 fatalities in crashes involving a driver with a BAC of.08 or higher in the state of Michigan (NHTSA 2012). > In 2011, there were 29,443 impaired driving arrests in the state (FBI 2012). > In Ottawa County in 2012, 236 women were arrested on OWI-related charges (Michigan Department of State Police 2013). Women accounted for 25% of OWI-related arrests in the county. State impaired driving laws Operating While Visibly Impaired (OWVI). A person commits the crime of OWVI if their ability to operate a motor vehicle is visibly impaired because of alcohol or other drugs. > 1 st offense: fine up to $300; up to 93 days in jail; up to 360 hours of community service; Driver Responsibility Fee of $500 for two consecutive years. > 2 nd offense: fine of $200-$1,000; five days to one year in jail; 30 to 90 days of community service; Driver Responsibility Fee of $500 for two consecutive years. > 3 rd offense: felony; fine of $500-5,000; one to five years imprisonment or probation with 30 days to one year in jail required; 60 to 180 days of community service; Driver Responsibility Fee of $500 for two consecutive years. Operating While Intoxicated (OWI). There are three different types of violations that fall under the auspices of OWI. These include: 1) alcohol or drugs present in the body that substantially affect an individual s ability to operate a motor vehicle safely; 2) a BAC at or above 0.08; and, 3) a high BAC of.17 or above. > 1 st offense: fine of $100-$500; up to 93 days in jail; up to 360 hours of community service; Driver Responsibility Fee of $1,000 for two consecutive years. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Appendix 113

364 > 1 st offense (high BAC): fine of $200-$700; up to 180 days in jail; up to 360 hours of community service; Driver Responsibility Fee of $1,000 for two consecutive years. > 2 nd offense: fine of $200-$1,000; five days to one year in jail; 30 to 90 days of community service; Driver Responsibility Fee of $1,000 for two consecutive years. > 3 rd offense: felony; fine of $500-5,000; one to five years imprisonment or probation with 30 days to one year in jail required; 60 to 180 days of community service; Driver Responsibility Fee of $1,000 for two consecutive years. Programs Supervision. Female DWI offenders can be processed through either traditional courts of DWI/Drug Courts. Michigan has a total of 49 hybrid DWI/Drug Courts, 6 Adult Drug Courts, and 12 DWI Courts. Interlock program. Michigan has a judicial interlock program that is mandatory for repeat and high BAC (.17 >) first offenders. Under Michigan s Repeat Offender Laws, habitual offenders (defined as having two or more convictions within seven years OR three or more convictions within ten years) are required to have an interlock. These offenders can apply for a restricted license after serving a minimum period of license revocation. Hearing officers must require that habitual offenders install an interlock on any vehicle they own or operate for at least one year. As of 2010, any individual with a restricted license that requires an interlock must continue to drive with the device until the Secretary of State authorizes its removal. A pilot project began in 2009 and was administered through the Eaton County DWI Court. All high BAC first offenders were required to install the device for a period of one year. Due to a compliance rate of approximately 88%, the Michigan Legislature passed a law the following year that expanded the pilot program to every DWI Court in the state. The program was further expanded in 2011 to include all repeat DWI offenders. Previously, these offenders were not eligible to obtain a restricted license without serving a lengthy hard suspension. The law was changed to permit these offenders to receive a restricted license after 45 days permitted that they are making progress in their DWI Court program and have installed an interlock. 114 Appendix FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

365 For more information on the Michigan pilot programs, please refer to: pdf Treatment. Most offenders convicted of alcohol-related offenses in Michigan are mandated by the court to undergo an assessment. For OWI/OWVI offenders, participation in an Alcohol Highway Safety Education program is mandatory. Based on the results of the assessment, offenders are required to complete different levels of programming. For example, one licensed provider offers three levels for offenders with varying degrees of risk: > Alcohol 1 is designed for first offenders with a focus on education and prevention. The program consists of one six-hour session. > Alcohol 2 is designed for more serious offenders with a focus on recognizing the seriousness of substance abuse. The program consists of several sessions for a total of eight hours. > Alcohol 3 is designed for the highest risk offenders and DWI/Sobriety Court participants. Participants in this program will already have been involved in treatment previously. The program consists of small group sessions that focus on cognitive skill-building as well as the development of an action plan for change. It spans multiple sessions and lasts a total of 20 hours. County selected The county chosen as a site for practitioner interviews in Michigan was Ottawa County. According to the 2010 U.S. Census, the population is 263, 801 (466/mile 2 ). In operation since May 2004, the 58th District DWI Court aims to promote community safety and reduce alcohol and drug abuse through a coordinated program involving intensive supervision, judicial interaction, treatment, incentives, sanctions and accountability. This Academy DWI Court is a postplea treatment court that is targeted toward a population that does not have a record of serious violent behavior or ongoing mental illness but does have a serious substance abuse pattern. In order for a DWI offender to be eligible to participate, they must be facing a second or subsequent charge in Ottawa County. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Appendix 115

366 Profile of offenders Note that the jurisdictional profile for Michigan is limited due to a lack of focus groups with female DWI offenders. While originally scheduled, these focus groups did not take place due to funding. However, interviews with two offenders and six practitioners were conducted. The following profile of offenders is based on the information provided to the researchers by the interviewed practitioners. > Practitioners report that there is a growing number of young female impaired drivers in courts and on probation caseloads (late teens and early 20s). > Women represent a wide cross-section of socioeconomic status and education levels; female offenders tend to work menial jobs or lowpaying jobs in relation to their male counterparts and most have only a high school education or some college education. > Most female offenders have children and are often single mothers. Payment of child support among fathers of these children is an issue. > Reported illicit drug use, particularly marijuana which is used to manage anxiety. > Transportation (i.e., finding alternative transportation options) was a significant problem in Michigan, particularly during the winter months. > Female offenders tend to struggle more financially than males. They also tend to violate probation more frequently due to missed appointments (which is often as a result of being unable to find childcare). > Many of the women had the following issues present in their history as confirmed through the observations of practitioners:» broken/dysfunctional household;» low self-esteem;» co-occurring disorders, particularly depression, bi-polar disorder, anxiety, and/or PTSD combined with alcohol dependence;» early onset of drinking;» prescription drug use and/or polysubstance use; 116 Appendix FEMALE DRUNK DRIVERS A QUALITATIVE STUDY

367 » history of trauma and grief which often served as a trigger for alcohol consumption;» domestic violence/abuse;» lack of support networks;» general health problems in addition to problems sleeping;» unhealthy relationships and/or a partner with a substance use problem; and,» other criminal offenses (e.g., drug offenses, retail fraud, assault, and driving on a suspended/revoked license). Profile of practitioners > Six practitioners participated in interviews (two judges, two treatment providers, one probation officer, and one probation officer/treatment counselor). > The experience level of practitioners ranged from 5 to 30 years in the field. > Treatment practitioners utilize assessment tools at intake that include identification of past trauma as it is recognized that incidences of abuse are common in the history of offenders. > The Substance Abuse Subtle Screening Inventory (SASSI) is another assessment tool used during the intake process for both male and female offenders. > The intake process utilizes a holistic approach and considers information about the offender, their background information, family, community environment, and any systemic issues that may be present. > Practitioners in these counties tend to be sensitive to lesbian-gay-bisexualtransgender (LGBT) issues. > Practitioners do not receive any gender-specific training. > Cases are rarely transferred from one probation officer to another without cause. Judges in these jurisdictions prefer there to be continuity in supervision and monitoring. > Caseloads involved low-risk and drug offenders. FEMALE DRUNK DRIVERS A QUALITATIVE STUDY Appendix 117

368 NOTES

369 T R A F F I C I N J U R Y R E S E A R C H F O U N D A T I O N Traffic Injury Research Foundation (TIRF) 171 Nepean Street, Suite 200 Ottawa, Ontario Canada K2P 0B4 Toll Free: Fax: Registered Charity No RR0001