Towards a national model for managing impaired driving offendersadd_

REVIEW doi:10.1111/j.1360-0443.2010.03339.x Towards a national model for managing impaired driving offendersadd_3339 1221..1227 Robert B. Voas 1, Robert L. DuPont 2, Stephen K. Talpins 3 & Corinne L. Shea 2 Pacific Institute for Research and Evaluation, Calverton, MD, USA, 1 The Institute for Behavior and Health, Inc., Rockville, MD, USA 2 and National Partnership on Alcohol Misuse and Crime, Alexandria,VA, USA 3 ABSTRACT Aims To describe a proposed national model for controlling the risk presented by offenders convicted of driving while impaired (DWI) and promoting behavioral change to reduce future recidivism. Setting Traditional methods of controlling the risk they present to the driving public are not adequate, as indicated by the fact that approximately 1000 people are killed each year-in alcohol-related crashes involving drivers convicted of DWI in the previous three years. However, stimulated by the success of special drug courts for substance abusers and new technological methods for monitoring drug and alcohol use, new criminal justice programs for managing impaired driving offenders are emerging. Intervention A national model for a comprehensive system applicable to both drug and alcohol impaired drivers is proposed. The program focuses on monitoring offender drinking or the offender driving employing vehicle interlocks with swift, sure but moderate penalties for non-compliance in which the ultimate sanction is based on offender performance in meeting monitoring requirements. Findings Several new court programs, such as the 24/7 Sobriety Project in South Dakota and North Dakota and the Hawaii s Opportunity Probation with Enforcement (HOPE) Project, which feature alcohol/drug consumption monitoring, have produced evidence that indicates even dependent drinkers can conform to abstinence monitoring requirements and avoid the short-term jail consequence for failure. Conclusions Based on the apparent success of emerging court monitoring systems, it appears that the cost of incarcerating driving-while-impaired offenders can be minimized by employing low-cost community correction programs paid for by the offender. Keywords Alcohol-related crashes, drugged drivers, DUI, jail, monitoring drinking, recidivism. Correspondence to: Robert B. Voas, Pacific Institute for Research and Evaluation, 11720 Beltsville Drive, Suite 900, Calverton, MD 20705, USA. E-mail: voas@pire.org Submitted 13 August 2010; initial review completed 13 October 2010; final version accepted 20 December 2010 INTRODUCTION In the US criminal justice system, the attention given to the handling of driving-while-impaired (DWI) offenders has been sporadic and often parochial, with no comprehensive vision articulated for their management that both protects the public and produces behavioral change. The current system for handling DWI offenders has emerged over half a century from state legislation, court case law and court-sponsored treatment programs. Thus, unlike the basic per se impaired-driving laws [such as the 0.08 BAC (blood alcohol concentration) limit] that apply across nearly all states and produce a relatively standardized enforcement system throughout the country, DWI sanctions, offender monitoring systems and treatment programs vary widely among and within states. Given that drivers convicted of DWI within the last 3 years are responsible for 1000 highway deaths each year [1], it is time to give more attention to why judicial programs designed to prevent DWI offender recidivism are not more effective. Current developments in the methods by which the US criminal justice system is handling DWI offenders suggest that a unique opportunity is developing to substantially restructure the DWI sanction/treatment system: (i) The economy dictates that incarceration must be used more efficiently to control costs so that the dwindling jail space can be used for more serious offenders. (ii) The effectiveness of license suspension, the traditional system for controlling DWI offenders, is compromised by high-traffic volumes and limited police resources. (iii) States are passing laws that apply penalties reserved formerly for

1222 Robert B. Voas et al. Table 1 Special features of current emerging driving-whileimpaired (DWI) offender management systems that can be incorporated in a national model. 1. Emphasis is placed on replacement of jail with low-cost monitoring programs paid by the offender 2. Sanction alternatives are maximized to increase flexibility in meeting offender needs 3. Behavioral triage is used to create performance-based sanctions 4. Control of consumption is emphasized to impact all substance abuse problems 5. Treatment is focused on offender needs in meeting monitoring requirements 6. Monitoring data is available to enhance offender screening and assessment 7. Rapid development of new technologies will increase the requirement for program evaluation multiple offenders to high-bac (0.15 + g/dl) first offenders. (iv) Programs that use monitoring, such as drug courts, are apparently effective in promoting rehabilitation. (v) The success of alcohol monitoring and interlock programs, limited initially to multiple offenders, is leading to their extension to some first offenders. (vi) Current monitoring programs demonstrate that even many users dependent on drugs and alcohol can comply with abstinence requirements. (vii) Advances in technology are providing effective, low-cost monitoring methods. Based on these trends, the elements of an emerging US National Model for Managing Impaired Driving Offenders, shown in Table 1, can be identified. A national model that informs state legislatures and courts should increase uniformity in handling both alcohol- and drug-impaired driver programs. It also should stimulate research by highlighting current limitations in offender screening, monitoring and treatment programs. By presenting a clear set of objectives, a national model can encourage the development of new programs and provide an improved basis for evaluating their effectiveness. The concept of a national model for impaired-driving laws is not new. The National Committee on Uniform Traffic Laws and Ordinances, which develops and maintains a model traffic law program, has been in existence for more than 50 years. In addition, the National Highway Traffic Safety Administration (NHTSA) has issued a report on model court programs for DUI offenders [2]. What follows is one vision of a traffic law system that deals with sanctioning and rehabilitating impaired drivers. It is intended to promote discussion of the possibility of establishing guidelines to ensure maximum effectiveness of new technologies and programs to stimulate progress in reducing the crash involvements of drivers impaired by both alcohol and drugs who enter the criminal justice system. BACKGROUND For the last several decades, the enforcement of drug abuse laws has emphasized incarceration. Twenty-five years ago, the cost of the growing prison population from drug abuse cases motivated the development of drug courts that focus on coerced treatment and abstinence monitoring [3] as an alternative to incarceration. Offenders avoid jail terms by participating in the program but are subject to short-term jail sentences for failure to maintain abstinence or attend treatment. The success of drug courts [4 6] has led to the expansion of this close monitoring of DWI offenders in other settings with varied but generally positive results [7 10]. Although DWI courts are less costly than lengthy jail sentences for repeat offenders, they require multiple hearings before the judge involving the prosecutor, the defense attorney, the treatment professional and the probation officer, increasing the cost of managing offenders over that of traditional probation [3]. Thus, some judicial policy makers have been developing less expensive community corrections programs, based on the low-cost alcohol- and drug-monitoring systems that are paid by the offender. The best documented of these is the South Dakota 24/7 Sobriety Project [8,9], which has now spread to North Dakota [7], and Hawaii s Opportunity Probation with Enforcement (HOPE) Program [10]. The 24/7 Sobriety Project, which focuses on alcohol offenses, requires offenders to visit a sheriff s department or police station to be tested for alcohol twice a day or to wear a continuous alcohol monitoring bracelet. Offenders are also subject to random urinalysis or must wear sweat patches that detect drug use. Failures to appear or positive tests are met with immediate short-term periods in jail. Unlike DWI/drug courts, treatment is not a formal part of the intensive-monitoring program although, as in most traditional programs, it is available at the offender s expense as one element of the overall sanctions program. This trend towards greater use of alcohol- and drugmonitoring has been stimulated by technological developments in alcohol and drug sensing, such as transdermal sensors that can be attached to the ankle or wrist to record BAC indirectly from the surface of the skin [11]. One such unit, the Secure, Continuous, Remote Alcohol Measuring (SCRAM ) system (Alcohol Monitoring Systems, Inc.; URL: http://www.alcoholmonitoring.com/index/scram/ product-overview; archived at http://www.webcitation. org/5uuoupkk7 on 24 November 2010) is used widely in courts as well as probation and parole offices across the country, including in the 24/7 program [9]. Other devices e.g. the IN-HOM unit (SmartStart, Inc.; http:/ /www.smartstartinc.com/ index.php/ products /; archived at http://www.webcitation.org/5uuoozyoe

Monitoring convicted impaired drivers 1223 on 24 November 2010) that measure the BAC through portable breath-test devices equipped with photo capability that identifies the individual providing the sample and transmits the information to the court are being introduced, as are skin patches that can detect illicit drug use [9]. All these devices lend themselves to monitoring offenders for extended periods at low cost, which is paid by the offender. Alcohol monitoring in the form of vehicle interlocks has also been extended to the principal historic sanction for impaired-driving: license suspension. Suspension has been losing its effectiveness because of the increasingly heavy traffic on the roads and its dependence on local police deterring illicit driving. The annual number of unlicensed drivers in fatal crashes has averaged 6934 over the past 10 years and has increased from 11 to 14% of all drivers in fatal crashes over that period [12]. In any case, deterrence is so low that 75% of suspended drivers admit to driving illicitly [13], a figure confirmed by undercover observations [14]. Although license suspension is beneficial relative to no licensing action, vehicle alcohol ignition interlock programs have demonstrated the advantage of monitoring driving, as they reduce recidivism by 64% when compared to suspended drivers [15]. Forty-eight states now have laws providing for interlock programs, and 210 000 units are currently in use, which is equivalent to 15% of the 1.4 million drivers arrested for DWI each year ([16]; see slide 3). To date, the use of intensive BAC monitoring systems and vehicle interlock systems have tended to be limited to multiple offenders, but states are increasingly passing laws creating a category of high-bac first offenders who are subject to penalties similar to multiple offenders [17]. Thus, the growth of low-cost drinking and drinkingdriving monitoring programs can potentially be scaled up to include first DWI offenders, which would transform the current methods for handling DWI offenders. Although not eliminating the need for the two principal traditional control systems, incarceration and license suspension, monitoring could produce a lower-cost, more effective management system. Whether monitoring abstinence will have an advantage over traditional license suspension, probation and incarceration remains to be demonstrated fully, but initial evidence for its effectiveness is beginning to accumulate [10,18]. PROPOSED MODEL FOR A NATIONAL PROGRAM Based on the trends described above and listed in Table 1, the elements of an emerging National Model for Managing Impaired Driving Offenders can be outlined. The justification for including each of the seven elements is described briefly below: 1 Emphasis is placed on replacement of jail with low-cost monitoring programs paid by the offender. Until recently, incarceration was the only sanction that ensured complete incapacitation for repeating the DWI offense. The development of electronic monitored home confinement, alcohol consumption monitors, such as the SCRAM and IN-HOM devices and vehicle alcohol interlocks, make it possible to incapacitate non-violent substance abusers at much lower cost and for longer periods, as the offender retains the capability to work and thereby to finance the control program. Further, these community correction programs are applied within the normal living environment where the targeted consumption adjustment must occur and where both professional treatment assistance and family support is most available. Thus, this process of employing monitoring as an alternative to jail promises not only to save money for communities but also to provide a better road to offender rehabilitation. This, however, remains to be proven. 2 Sanction alternatives are maximized to increase flexibility in meeting offender needs. Compared to the past, when options for controlling an offender s impaireddriving behavior were constrained for the most part to three sanctions with limited effectiveness probation [10], incarceration [19,20] or license suspension [12,13] the new control programs based on BAC measurement and interlock technology have greatly increased the options available to courts for managing impaired drivers. An important purpose of a national model would be to identify, evaluate and distribute information on emerging new technologies. In the program model shown in Fig. 1, the primary current options are arranged along the left margin of a matrix according to their punitive characteristics based on the extent of their intrusion into the offender s life. Moving from bottom to top, they define a sanctioning continuum of increasing severity as represented by higher-intensity monitoring/moreintrusive control programs culminating in incarceration. Adding a time-frame running from days to years creates a matrix that, conceptually, can accommodate drivers convicted of both alcohol- and drugged-driving offenses of varying levels of seriousness, as indicated by the sanction severity arrow in Fig. 1. 3 Adaptive programming and behavioral triage are used to create performance-based sanctions. Offenders would enter the model shown in Fig. 1, based on current sanctioning practices that consider the offense, the prior record and treatment needs of the offender. Instead of specifying a single sanction with a set length, with the offender remaining in a single cell of the Fig. 1 matrix until the sentence has been served, this model would encourage the use of more than one

1224 Robert B. Voas et al. Figure 1 National model for managing driving while impaired (DWI) offenders. TD Monitor Transdermal Sensor; Secure, Continuous, Remote, Alcohol Monitoring (SCRAM) unit produced by Alcohol Countermeasures Systems a leg bracelet worn 24/7 that monitors BAC; 24/7 a program that requires reporting to police for twice-a-day breath testing. Home Monitor IN-HOM, a portable automatic breath-test device with photographic verification of the user produced by Smart Start, Inc. RandomTesting offenders call in daily to determine whether they must come in for a breath test type of sanction over a variable period based on offender performance as measured by monitoring data. This would enable a behavioral triage program that responds both positively and negatively to offender performance [21]. Low-cost monitoring systems would create a continuous record of key criterion measures of alcohol and/or drug use or impaired driving (from interlock data). Based on the close monitoring of such measures, conformity to sanction requirements could be rewarded by reduction of time under supervision or by movement from greater- to less-intrusive control measures as indicated by the performance and responsibility arrows in Fig. 1. Alternatively, failures to conform would be penalized by increases in sanction severity through extensions of the time under supervision and/or by imposition of tighter control measures. Within this framework, multiple DWI offenders with the most serious records (and evidence of addiction) could avoid long jail terms by participating in DWI/drug court programs. Second offenders would participate in programs, such as 24/7 or HOPE, or programs employing new technologies, such as SCRAM or IN-HOM, that provide close monitoring with immediate but moderate consequences (1 or 2 days jail or extension of monitoring period) for test failure. High-BAC first offenders could be placed in alcohol- or drug-monitoring programs similar to those applied to second offenders. Vehicle alcohol interlocks may provide an appropriate monitoring system for other first DWI offenders, allowing them to use their interlocked vehicle rather than being suspended. When suspended they tend to drive illicitly, bringing on more serious penalties such as vehicle impoundment and jail [13]. However, many first offenders avoid interlocks by claiming not to have a vehicle. BAC monitoring might be an appropriate sanction for those offenders [22]. 4 Control of consumption is emphasized to impact all substance abuse problems. While the target of DWI management programs is to reduce recidivism and crash involvement and not necessarily abstinence, the use of sanctions directed at the control of substance use offers the possibility of impacting non-driving alcohol-related injuries and fatalities, as well as family conflicts and unprotected sex. To date, there have been no studies that specifically consider non-traffic-related benefits, other than continued abstinence, of DWI court-mandated treatment programs, but the increasing use of alcohol and drug monitoring should create data from which to evaluate this expected serendipitous value of monitoring programs. 5 Treatment is focused on offender needs in meeting monitoring requirements. Historically in the United States, mandated treatment programs for DWI offenders have operated mainly outside the other traditional criminal sanctioning elements incarceration, fines and license suspension with the court becoming involved only when offenders fail to attend the treatment program. DWI/drug courts pioneered an exception to this practice. Moving to a court system that relies more on monitoring drinking than on incarceration or preventing driving through license suspension promises to transform the role of treatment in managing non-violent substance-using offenders. Implementation of alcohol-monitoring programs will require that, beyond the current standard treatment modalities (in-patient, out-patient group therapy and alcohol education), individualized counseling programs be offered to offenders who fail to meet specific courtmonitoring requirements. Treatment providers will need to have access to the alcohol- and drug-use data collected by the monitoring program and will need to establish intervention procedures responsive to the offender s needs in meeting the requirements of the monitoring process. More intensive monitoring focused on enforcing abstinence should increase the relevance of court treatment programs for offenders and probation departments. Most offenders have little or no intrinsic motivation to achieve recovery, particularly when they often do not recognize they have a problem (readiness to change) [23,24]. However, offenders faced with close monitoring of their drinking with significant consequences for failure to conform should be more likely to embrace the support

Monitoring convicted impaired drivers 1225 offered by court-sponsored treatment. The Florida interlock program exploits that opportunity by placing interlock program offenders who continue to have high-bac tests into a treatment program where termination of the interlock requirement is dependent on clinical judgment as well as interlock performance [25]. Contingent treatment program planning will be needed for offenders who cannot meet required standards, even with maximum support, to avoid excessive sanction levels. 6 Monitoring data are available to enhance offender screening and assessment. Determining the alcoholand drug-use status of the offenders is essential to determining treatment needs. Assessment would become even more significant in a future system supplemented by alcohol- and drug-use monitoring for early identification of those likely to have trouble meeting monitoring requirements requiring their assignment to supportive services. Currently, the level of treatment to which an impaired-driving offender is assigned is determined by (i) prior offenses, (ii) arrest BAC and (iii) screening tests/interviews. Each of these has limitations for determining the offender s substance use status. The official number of prior offenses is determined by the look-back period, which differs across states. Arrest BAC correlates only marginally with future recidivism [26,27]. Screening tests depend on self-reported information that may be relatively accurate in clinical settings but is problematic in criminal justice settings, where the offenders may fear unpleasant consequences for being candid [28]. This limited information would be augmented substantially by data flowing from an emerging offender control system that monitors drinking and drug use or from vehicle interlock breath-test data. Both methods produce a large amount of relatively objective behavioral data (BAC) that can be used to assess the offender s drinking status and treatment progress [29,30]. Although these data are not available when sentenced, unless monitoring occurs as a condition of bond they accumulate during probation and can be used to augment the original assessment. These data should influence the final diagnosis of the offender as well as the ultimate severity of the sentence based on the demonstrated ability of offenders to control their drinking and drug behavior. Adding information on drinking and drug behavior from monitoring measures to the traditional assessment information should provide a method for measuring progress in treatment. As noted, interlock breath-test records, as well as biomarker data from blood-test programs (measuring carbohydrate deficient transferrin (CDT), gamma glutamyltransferase (GGT) and phosphatidylethanol (PETH), have been shown to be predictive of future recidivism [30,31]. It is expected that data from the programs that monitor drinking can have similar predictive value; after the standard treatment program has been completed, such data can be used to detect relapse. It also should provide an objective method of determining when offenders can be released from probation supervision. 7 Rapid development of new technologies will increase the requirement for program evaluation. A national model program will highlight weak links in the current offender control system by identifying unsatisfied needs. This should stimulate innovation and research. It also should increase the adoption of new evidencebased programs featuring more effective monitoring, assessment and treatment methods. Many of the new technologies and programs described here have not yet been evaluated adequately; however, it is impossible to delay innovation until that happens. Because of the pressure to reduce costs while increasing control over offenders, new programs and technological devices will spread through the court system rapidly. That places a responsibility on funding agencies and researchers to implement evaluations of new programs as soon as possible after their introduction. COMMENTARY The threat of certain but moderate sanctions, such as brief periods in jail, is apparently the key to successful enforcement of abstinence in both drug courts and in the 24/7 and HOPE programs. Although offenders with alcohol- or drug-use disorders would be expected to have difficulty maintaining abstinence, experience in those programs indicates that even third and fourth offenders who would clearly be classified as drug- or alcoholdependent can usually meet those requirements, at least while in a monitoring program [8 10,32] and, in the case of drug courts, in the period following release from the program [3]. The therapeutic benefit of enforced abstinence remains to be demonstrated. There is emerging but limited evidence that it may reduce recidivism [6 9,31], but the extent to which it reduces substance use has not been determined. To date, the application of the 24/7 or SCRAM-type monitoring programs have been relatively short (60 180 days) [7,8] compared to typical lengths of court probation periods, license suspension or interlock installation mandated typically for multiple DWI offenders. Whether substance-monitoring procedures can be maintained over similar periods to detect relapse remains to be determined. The use of alcohol and drug monitoring to control impaired-driving offenders is not new. Sweden has focused on treatment programs for impaired drivers with a requirement for objective measures of recovery, such as blood tests and clinical determinations, for more than 40 years. Recently, a requirement to meet biomarker criteria

1226 Robert B. Voas et al. for license reinstatement has been built into the Swedish vehicle interlock program [33]. That program has reduced hospital costs and recidivism [34 36]. There is strong evidence that, in the United States, adding a biomarker-monitoring to current interlock programs would increase the ability to predict recidivism [31,37]. What we have proposed is transformational in its methods for managing impaired-driving offenders but not radical. It actually depends on concepts such as control of substance use and treatment that have been a feature of criminal justice programs for years. The proposed behavioral modification procedures grow out of classical learning theory, and the principal feature certain and swift sanctions is the key feature of deterrence theory ([38], chapter 1). The focus on the agent substance misuse in the traditional public health trilogy (agent, host and environment) moves the problem more centrally into the health field and away from a purely criminal justice setting. Declarations of interest Steven Talpins is a consultant to Alcohol Monitoring Systems, Inc., the distributer of the SCRAM device mentioned in this paper. The other authors have no conflicts. Acknowledgements The authors wish to acknowledge the contribution to this paper made by the following members of the Institute for Behavior and Health DWI Committee who agreed to read and comment on this paper. Their participation in the review process does not necessarily imply agreement with the findings and recommendations included in the paper: Larry Gentilello MD, Chairman, Division of Burns, Trauma & Critical Care, University of Texas Southwestern Medical School; Rear Admiral James E. McPherson JAGC, USN (Ret.), former Judge Advocate General of the Navy; Marilyn Huestis PhD, Chief, Chemistry and Drug Metabolism, Intramural Research Program, National Institute on Drug Abuse, National Institutes of Health; and Jane Maxwell PhD, Senior Research Professor, The University of Texas at Austin School of Social Work. References 1. National Highway Traffic Safety Administration. Fatality Analysis Reporting System (FARS): U.S. Department of Transportation, National Highway Traffic Safety Administration; 2009. 2010. Available at: http://ftp://ftp.nhtsa.dot.gov/ fars/ (accessed 30 March 2010). 2. National Highway Traffic Safety Administration. A Guide to Sentencing DUI Offenders. Report no.: DOT HS 808 365. Washington, DC: National Highway Traffic Safety Administration. 2005. 3. Marlowe D. B., Festinger D. S., Arabia P. L., Croft J. R., Patapis N. S., Dugosh K. L. A systematic review of DWI court program evaluations. In: Marlowe D. B., editor. Drug Court Review. Alexandria: National Drug Court Institute; 2009, p. 1 52. 4. Fuller B., Carey S. M., Kissick K. Michigan DUI Courts Outcome Evaluation. October Report no.: Final Report. Portland: NPC Research; 2007. 5. Rempel M., Fox-Kralstein D., Cissner A., Cohen R., Labriola M., Farole D. et al. The New York State Adult Drug Court Evaluation: Policies, Participants and Impacts. New York: Center for Court Innovation; 2003. 6. Belenko S. Research on Drug Courts: A Critical Review. New York: The National Center on Addiction and Substance Abuse, Columbia University; 2001. 7. DuPont R. L., Talpins S. K., Voas R. B. Leveraging the Criminal Justice System to reduce drug-related crime: a review of three promising and innovative model programs. Prosecutor 2010; 44: 38. 8. Long L. The 24/7 sobriety project. Public Lawyer 2009; 17: 2 5. 9. Bachand P. South Dakota s 24/7 Sobriety Program. Between the Lines, National Traffic Law Center 2008. 17(3). p. 2 5. 10. Hawken A., Kleinman M. Managing Drug Involved Probationers with Swift and Certain Sanctions: Evaluating Hawaii s HOPE. Report no.: 229023. Washington, DC: US Department of Justice; 2009. 11. Marques P., McKnight A. S. Field and laboratory alcohol detection with two types of transdermal devices. Alcohol Clin Exp Res 2009; 33: 703 11. 12. National Highway Traffic Safety Administration. Traffic Safety Facts: Trends in Fatal Crashes Among Drivers with Invalid Licenses. Report no.: DOT HS 811 229. Washington, DC: National Center for Statistics and Analysis, National Highway Traffic Safety Administration; 2009. 13. Ross H., Gonzales P. The effect of license revocation on drunk-driving offenders. Accid Anal Prev 1988; 20: 379 91. 14. McCartt A. T., Geary L. L., Nissen W. J. Observational Study of the Extent of Driving While Suspended for Alcohol-Impaired Driving. Report no.: DOT HS 809 491. Washington, DC: US Department of Transportation, National Highway Safety Administration; 2002. 15. Willis C., Lybrand S., Bellamy N. Alcohol ignition interlock programmes for reducing drink-driving recidivism. Cochrane Database Syst Rev 2005; 4: CD004168. 16. Roth R. Interlocked DWI Offenders have lower cumulative recidivism for six years after installation. A Presentation of Impact DWI and PIRE at the 2010 RSA Conference. 2010. Available at: http://www.rothinterlock.org/presentations. htm; archived at http://www.webcitation.org/5uumaghrh (accessed 24 November 2010). 17. McCartt A. T. Evaluation of Enhanced Sanctions for Higher Bacs: Summary of States Laws. Report no.: DOT HS 809 215. Washington, DC: National Highway Traffic Safety Administration; 2001. 18. Flango V., Cheeseman F. When should judges use alcohol monitoring as a sentencing option in DWI cases? Court Rev Am Judges Assoc 2009; 44: 102. 19. Wagenaar A. C., Maldonado-Molina M. M., Erickson D. J., Ma L., Tobler A. L., Komro K. A. General deterrence effects of U.S. statutory DUI fine and jail penalties: long-term follow-up in 32 states. Accid Anal Prev 2007; 39: 982 94. 20. Yu J. Punishment and alcohol problems: recidivism among drinking-driving offenders. J Crim Justice 2000; 28: 261 70.

Monitoring convicted impaired drivers 1227 21. Paternoster R., Brame R., Bachman R., Sherman L. W. Do fair procedures matter? The effect of procedural justice on spouse assault. Law Soc Rev 1997; 31: 163 204. 22. Voas R. B., Marques P. R. Commentary: barriers to interlock implementation. Traffic Inj Prev 2003; 4: 183 7. 23. Prochaska J., DiClemente C. Toward a comprehensive model of change. In: Miller W., Heather N., editors. Treating Addictive Behaviors, Processes of Change. New York: Plenum Press; 1986, p. 3 27. 24. Rider R., Kelley-Baker T., Voas R., Murphy B., McKnight A. J., Levings C. The impact of a novel educational curriculum for first-time DUI offenders on intermediate outcomes relevant to DUI recidivism. Accid Anal Prev 2006; 38: 482 9. 25. Voas R. B., Tippetts A. S., Fisher D. A., Grosz M. Requiring suspended drunk drivers to install alcohol interlocks to reinstate their licenses: effective? Addiction 2010; 105: 1422 8. 26. Marowitz L., DeYoung D., Yu J. Drunk driving recidivism: predicting factors from arrest context and case disposition. J Stud Alcohol 1996; 57: 679 80. 27. Wieczorek W. F., Miller B. A., Nochajski T. H. The limited utility of BAC for identifying alcohol-related problems among DWI offenders. J Stud Alcohol 1992; 53: 415 9. 28. Lapham S. Screening and brief intervention in the criminal justice system. Alcohol Res Health 2004/2005; 28: 85 93. 29. Marques P. R., Voas R. B., Tippetts A. S. Behavioral measures of drinking: patterns in the interlock record. Addiction 2003; 98: 13 9. 30. Marques P. R., Tippetts A. S., Voas R. B., Beirness D. J. Predicting repeat DUI offenses with the alcohol interlock recorder. Accid Anal Prev 2001; 33: 609 19. 31. Marques P., Tippetts S., Allen J., Javors M., Alling C., Yegles M. et al. Estimating driver risk using alcohol biomarkers, interlock BAC tests, and psychometric assessments: initial descriptives. Addiction 2010; 105: 226 39. 32. Hawken A. The message from Hawaii: HOPE for probation. Perspectives 2010; Summer: 36 43. 33. Bjerre B. Primary and secondary prevention of drinkdriving by the use of alcolock device and program. Swedish experiences. Accid Anal Prev 2005; 37: 1145 52. 34. Bjerre B. An evaluation of the Swedish interlock program. Traffic Inj Prev 2003; 4: 98 104. 35. Bjerre B., Marques P., Selen J., Thorsson U. A Swedish alcohol ignition interlock programme for drink-drivers: effect on hospital care utilization and sick leave. Addiction 2007; 102: 560 70. 36. Bjerre B., Kostela J., Selén J. Positive health-care effects of an alcohol ignition interlock programme among driving while impaired (DWI) offenders. Addiction 2007; 102: 1771 181. 37. Marques P. R. The alcohol ignition interlock and other technologies for the prediction and control of impaired drivers. In: Verster J. C., Pandi-Perumal S. R., Ramaekers J. G., de Gier J. J., editors. Drugs, Driving and Traffic Safety. Basel: Birkhäuser Verlag AG; 2009, p. 457 76. 38. Ross H. L. Deterring the Drinking Driver: Legal Policy and Social Control, 2nd edn. Lexington: DC Heath and Company; 1982.