An Adaptive Reinforcement-Based Treatment (RBT) Intervention for Pregnant Substance Dependent Women Michelle Tuten, Ph.D. Assistant Professor Department of Psychiatry and Behavioral Sciences and Johns Hopkins University School of Medicine HIV Methods and Intervention Science Meeting September 12-13 th 2013
National Institute on Drug Abuse R01DA14979 Participants Staff at the Center for Addiction and Pregnancy Hendree Jones (original study PI: slides adapted from 2011 CPDD workshop) Collaborators Susan Murphy, Pierre Alexander, Margaret Chisolm (current PI) Acknowledgements 2
Outline 1. Complexities of the Problem 2. Clinical Setting 3. Role of Behavioral Treatment 4. HOME II Study 3
National Survey on Drug Use and Health 2008/9 Past Month Use 1. Complexities of the Problem Alcohol and tobacco substances are the most commonly abused during pregnancy Minority but significant number of pregnant women (also use) illicit drugs during pregnancy Drug addiction begins before pregnancy and is the result of complex past and current factors 4
TREATMENT 1. Complexities of the problem Pregnancy is a window of opportunity for behavior change 5
1. Complexities of the problem Issues Facing Pregnant Drug Users Exposure to violence, trauma Generational drug use Lack of formal education Lack of work history or skills Legal involvement Multiple drug exposure Limited parenting skills History of abuse/neglect Psychiatric issues Unstable housing Lack of social support Medical issues Nutrition deficits 6
2. Clinical setting Center for Addiction and Pregnancy mission statement: Address Barriers to Care Improve maternal and infant outcomes Conduct clinical research 7
2. Clinical setting Comprehensive Care Interdisciplinary approach Substance abuse counseling Psychiatry Obstetrics Pediatrics Nursing Multiple modalities Medically-assisted withdrawal Methadone stabilization 8
2. Clinical setting Outcome research: cost effectiveness Clinical measure: CAP (n=100) No Treatment (n=46) Prenatal visits 8 4 EGA (mean week) 39 35 % positive at delivery 37 63 Infant birth weight (grams) 2934 2539 Apgar scores (1 minute) 8 7 % NICU use 10 26 Length of hospital stay (days) 7 39 Investing in CAP drug treatment services resulted in net savings of $4,644 in NICU costs. Although CAP is costeffective, many women continue to drop out of treatment prematurely and/or do not respond to treatment All group comparisons are significant: p.05 (Data adapted from Svikis et al., 1997) 9
3. Role of behavioral treatment Guiding Principles Drug addiction occurs in absence of alternative positive reinforcers Drug use makes sense in that it meets certain behavioral functions for the individual Drug use behavior can be modified using operant reinforcement Goal of treatment is for nondrug using behaviors become more reinforcing than drug use 10
Treatment Elements Recreational Activities Personalized Feedback Intensive Outreach Peer Support/ Social Club Patient 3. Role of behavioral treatment Behavior Graphing Vocational Assistance/ Job Club Functional Drug Use Assessment Treatment Elements Functional assessment (FA) Personalized feedback (MI adaptation) Social reinforcement (in the form of weekly Social Club) Behavior graphing Vocational assistance/job club Recreational activities Intensive outreach Tangible reinforcers 11
Treatment Elements 3. Role of behavioral treatment Treatment Objectives Functional Assessment (FA) of drug use Behavior Graphing Treatment attendance Recreational Activities Vocational Activities Medication compliance FA: when, where and why of drug use Constellation of behaviors that compete with drug use Graph these target behaviors 12
Abstinence (yes/no) Example of Close Behavioral Observation: DRUG FREE DAYS 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 1 Good Job! 3 5 7 9 11 Good Job! 13 3. Role of Behavioral Treatment 15 Allow for connections between behaviors (if you do this, drug use occurs, if you do this, drug abstinence occurs) Larger goals are broken into smaller goals Positive reinforcement given for successes Days 13
Number of Applications Example Of Behavior Tracking 3. Role of Behavioral Treatment 16 14 12 Number of Job Applications 10 8 6 Goal= 9 applications/week 4 2 0 week 1 week 2 week 3 week 4 Employment obtained 14
Days Drug Free Examples: Participant Treatment Progress 3. Role of Behavioral Treatment 15
Mean Days % Positive Urine Samples 3. Role of Behavioral Treatment 100 80 60 40 20 0 Opioid and Cocaine Negative * RBT Control 100 80 60 40 20 0 RBT (n=72) Control (n=128) * * * * 1-month 3-month 1-month 3-month Opioids Cocaine 3 randomized trials showing RBT s efficacy in nonpregnant patients (1 month, 6 months, 12 months follow-up) 1RBT adaptation and test in pregnant patients 100 80 60 40 20 0 Opioid and Cocaine Negative * RBT Control 100 80 60 40 20 0 RBT Delivery UC * As with any intervention, not all RBT-treated participants attended treatment or reduced drug use, suggesting room to improve patient response 16 (Gruber et al., 2000; Jones et al., 2005; Jones 2011; Tuten et al., 2012 * All treatment condition comparisons are significant at p.05 16
3. Role of Behavioral Treatment We identified two important questions regarding RBT for pregnant women: Would variants of RBT in terms of their intensity and scope be effective? Would patients who differed in terms of their treatment compliance be differentially responsive to these variants of RBT?
4. A SMART trial at HOME HOME II Eligibility Criteria Inclusion Criteria: Evidence of cocaine and/or opiate use Treatment entry at or before 32 weeks EGA with singleton fetus Completion of the eight-day residential detoxification stay Exclusion Criteria: Age 17 or younger Geographical Constraints Severe medical or psychiatric concomitant condition interfering with treatment or needing hospitalization 18
4. A SMART trial at HOME Primary Outcomes- Maternal: Treatment Completion (Delivery) Heroin Use Cocaine Use Secondary Outcomes- Neonatal : Birth weight Head Circumference Length of Hospital/NICU Stay Urine Toxicology at Delivery Physical Birth Parameters Neonatal Complications 19
Decisions for Designing the Trial: How to choose a design structure to answer questions of interest: Initial randomization conditions? Second randomization to what? Which treatment response characteristics: Was strongly related to outcome? Would allow reasonable tailoring? 4. A SMART trial at HOME Solutions: Initial Randomization: Compare TAU RBT to a potentially more cost effective form of RBT intervention Treatment Compliance as the tailoring variable: Failure to comply with treatment soon after treatment entry the biggest reason for failure to complete treatment Length of time to allow for assessment of the tailoring variable Pregnancy time-limited window Assess compliance during the first two weeks following treatment entry 20
Planned Sample Size 4. A SMART trial at HOME HOME II intervention adjusts in intensity or scope following patients initial treatment compliance or non-compliance. It is expected that RBT in an adaptive intervention format will optimize maternal treatment outcomes for both early compliant and early noncompliant participants by matching treatment to patient needs. 21
4. A SMART trial at HOME All non-methadone CAP Patients trbt: Treatment-as-usual RBT All elements of treatment-as-usual RBT trbt 5X/week: Individual sessions 3X/week: Recreation 3X/week: Job Club Random Assignment rrbt 3X/week: Individual sessions 1X/week: Recreation 1X/week: Job Club rrbt: reduced RBT Key RBT elements are provided at a reduced scope to examine a version of RBT that might be more in line with community practice limitations 22
4. A SMART trial at HOME Tailored Treatment Randomization Two-week window for treatment response Early compliant participants randomized to either the same intensity of treatment or a decreased intensity or scope of RBT treatment Early treatment non-compliance: (a) a missed unexcused treatment day, (b) a positive opioid or cocaine urine specimen, (c) or self-report of use of either drug. Early non-compliant participants randomized to receive either the same treatment or a greater intensity or scope of RBT 23
4. A SMART trial at HOME Increased intensity erbt: Enhanced RBT All trbt elements, plus: Home visits to re-engage participant/deliver therapy Immediate re-admission to the residential unit or community recovery housing as time out from drug use Decreased intensity arbt: Abbreviated RBT Most reduced version of RBT: Graphing drug abstinence Outreach limited to follow-up call Individual sessions once a week Recreation and social club once a month 24
4. A SMART trial at HOME Primary Aim: Relative efficacy of providing continued treatment-as-usual RBT in comparison to continued reduced RBT to both early-compilers and early noncompliers throughout the trial. Addresses the question of whether it is necessary to provide treatment-asusual RBT, or if it is possible to successfully treat patients with reduced RBT, regardless of the patient s level of compliance. 25
26 4. A SMART trial at HOME Secondary Aims: Early non-compliers: Relative efficacy of transitioning to a more intensive level of treatment for the TAU RBT and the reduced RBT conditions. Answers questions regarding the relative importance of the initial level of care in determining the efficacy of transitioning to a higher level of care for early-non compliers.
27 4. A SMART trial at HOME Secondary Aims (cont.): Early-compliers: relative efficacy of transitioning to a less intense level of treatment for within the initial TAU RBT and reduced RBT treatment conditions. Answers questions about the importance of the initial level of RBT treatment in determining success in subsequently reducing the level of RBT treatment in early-compliers.
4. A SMART trial at HOME Secondary Aims (cont.): Relative efficacy for early-complier and early-non-compliers who begin and continue in treatment-as-usual RBT or who begin and continue in reduced RBT, respectively. Answers the question about the relative decrement in treatment success that occurs as a result of an early failure to comply with the initial demands of treatment. 28
Current Enrollment 4. A SMART trial at HOME All non-methadone CAP Patients N= 230 trbt n =114 Random Assignment Tailoring Variable Randomization: Treatment Compliance rrbt n =116 Decrease Intensity Early Compliant n=48 Early Non- Compliant n=61 Increase Scope Increase Intensity Early Non- Compliant n=53 Early Compliant n=56 Decrease Scope rrbt n=24 trbt n=24 trbt n=30 erbt n=31 trbt n=26 rrbt n=27 rrbt n=28 arbt n=28
4. A SMART trial at HOME Challenges in Conducting the Trial Recruitment Participant adherence to assessment protocol Maintenance of distinct treatment conditions Adherence of the clinical staff to the protocol Changing clinical programming New Insights Resulting from the Trial Patient acceptance of the tailoring treatment randomization Better-than-expected distribution of patients into tailoring treatment randomization 30
4. A SMART trial at HOME: SUMMARY What we ve learned from process: Feasible and desirable Well-defined tailoring/response definitions Distinct and theoretically sound conditions Procedures to maintain conditions (and documented deviations) 31