Focus Area: Substance Abuse

For people transitioning from incarceration back to their communities, substance abuse is often closely related to their difficulties with housing, employment, and mental health. Research on the effectiveness of various in-prison and community-based substance abuse intervention programs in improving reentry outcomes has begun to show what approaches reduce recidivism, promote public safety, and improve the lives of the formerly incarcerated and their families. However, more research is needed to develop a better understanding of what works, for whom, and under what circumstances.

This section provides an overview and examination of key research on the relationship between substance abuse program participation and recidivism, substance use, employment, and mental health outcomes of those returning from incarceration.

A total of thirty-nine publications met eligibility criteria, including a number of publications that evaluated therapeutic community treatment approaches. Below is general description of the trends and themes that emerged from the research review.

What the Research Says about Substance Abuse Programs

Therapeutic Community Programs: Based on a systematic review of literature published through 2010, 17 evaluations of in-prison therapeutic community (TC) programs met criteria for methodological rigor. Across these 17 studies, most found that program participation had a positive impact on both recidivism and substance abuse. Many of the evaluations of TC programs included analyses examining the effects of participation in aftercare programs. Overall, the findings suggest that the presence of a formal aftercare may be effective at supporting improved outcomes, but potential selection biases may muddle these findings. Among the six studies reviewed that assessed the effectiveness of TC programs with female inmates, a strong pattern emerged in which gender-responsive treatment was more likely to yield positive outcomes than traditional TC treatment.

Other Institutional Substance Abuse Treatment Programs: In addition to in-prison TC programs, five studies of other institutional substance abuse treatment programs met eligibility criteria for inclusion. Three evaluated the impact of methadone treatment as an alternative to detoxification on recidivism and substance abuse outcomes. Findings suggested that none of the programs impacted recidivism but the one in Baltimore did demonstrate a modest beneficial impact on substance use. Participants in the program were less likely to test positive for opioids and cocaine than those in the comparison group. The two remaining studies evaluated the Offender Substance Abuse Pre-Release Program implemented by the Correctional Service of Canada to meet the needs of persons housed in federal facilities and five in-jail substance abuse treatment programs in California and New York. Both found a beneficial impact of treatment on recidivism outcomes.

Driving While Intoxicated Programs: Four studies of two Driving While Intoxicated (DWI) programs—the San Juan County DWI Program and the Turning Point Multiple Driving Under the Influence (DUI) Treatment Program—met eligibility criteria. The San Juan County DWI Program was designed to address alcohol-related motor vehicle crashes by targeting first-time DWI offenders through inpatient counseling. The Turning Point Program is a residential treatment intervention designed to provide substance abuse treatment counseling to individuals with multiple DUI convictions through the development of individualized treatment plans and the provision of services such as AA/NA meetings, and educational programs. Findings across the two studies of the San Juan County DWI Program were mixed, but largely supported the effectiveness of the program in reducing alcohol consumption and arrests for DWI. A study of the Turning Point program found modest evidence of a beneficial effect of the program on recidivism outcomes, with stronger program effects on those who had spent more time in jail for the initial offense and who had three or more previous DUI convictions. One study explored a particular component of the DUI program—victim impact panels (VIP), which were designed to emphasize for participants the dangers of drinking and driving on the lives of others—but did not find evidence of a beneficial effect of VIPs on recidivism or substance abuse outcomes.

Community-Based Substance Abuse Treatment Programs: The remaining thirteen studies evaluated various transitional or community-based substance abuse treatment programs. Findings across these evaluations were mixed which is not surprising given the wide range of program models included. Five found no impact on recidivism and/or substance abuse outcomes – the transitional, strengths-based case management program, one of two evaluations of the Treatment Alternatives to Street Crime (TASC) program based in Colorado, a peer counseling/nurse case management program in California, the Recovery Management Checkup model implemented in Cook County, Illinois, and a multimodal community based prison reentry program targeting men at high risk of recidivating.

The remaining eight studies found some evidence of a beneficial impact of program participation on recidivism and/or substance use outcomes though interestingly none of them found positive impacts on both types of outcomes. Evaluations of four programs - a study of the Drug Offender Sentencing Alternative (DOSA) policy in Washington state, a separate evaluation of the TASC program in Colorado, a social support program for parolees in Maryland, and an intensive outpatient program in the northeastern United States - found that participants were less likely to recidivate than those in the comparison groups. The other four studies – a study of five outpatient treatment aftercare programs in Baltimore, Maryland, an evaluation of a motivational interviewing program for women in the Rhode Island Department of Correction facility, a study of a parole intervention modelled after Hawaii’s Opportunity Probation and Enforcement program, and an evaluation of Step’n Out, a collaborative behavioral management program for parolees – found that participants were less likely to engage in illegal substance use than those in comparison groups.

Future Areas of Research

Few studies sought to examine the effects of substance abuse programming on outcomes other than recidivism or substance use, such as employment, housing, or mental health outcomes. Future research should seek to examine how substance abuse programs impact other reentry outcomes, particularly those that have a strong impact on determining whether formerly incarcerated individuals with substance abuse histories successfully reenter society.

Only one of the TC studies examined the effectiveness of programming by individuals’ risk to reoffend. Future studies should seek to examine whether TC programs are more effective with individuals of certain risk levels, as well as other characteristics.

The impact of aftercare needs additional focus, especially for TC programs. Future studies of TC programs should randomly assign in-prison TC participants to aftercare or no aftercare if possible. Most analyses of the effectiveness of aftercare for TC program participants compare individuals who graduate from in-prison TCs and choose to continue in aftercare to individuals who either (1) do not graduate from the in-prison TC program or (2) do graduate from the in-prison program but choose not to continue in aftercare upon release. In both cases, these analyses are subject to selection bias.

Finally, each of the programs studied would benefit from replication studies or reevaluation, in particular, TC programs that have aftercare components and other non-TC, non-DWI substance abuse programs.

Summary of Evaluations and Outcomes

Rigor Evaluation Recidivism Employment Substance Abuse
High Rigor Anglin et al., 2002; Zhang, Roberts, & McCollister, 2011 (not evaluated)
Basic Rigor Aos, Drake, et al. 2005 (not evaluated) (not evaluated)
Basic Rigor Daley et al. 2004 (not evaluated) (not evaluated)
High Rigor Duwe 2010 (not evaluated) (not evaluated)
Basic Rigor Farrell 2000 (not evaluated)
Basic Rigor Gordon et al., 2008; Kinlock et al., 2009; Gordon et al., 2012; Kinlock et al., 2013 (not evaluated) (not evaluated)
High Rigor Grommon et al., 2013 (not evaluated)
High Rigor Grommon, Davidson II, & Bynum, 2013 (not evaluated)
High Rigor Hanlon, Nurco, et al. 1999
High Rigor Inciardi et al. 2009 (not evaluated)
Basic Rigor Kunitz et al., 2002; Woodall et al., 2004; Delaney et al., 2005 (not evaluated) (not evaluated)
Basic Rigor Langworthy, Latessa, Pratt, & Holsinger 1993 (not evaluated) (not evaluated)
Basic Rigor Magura, Rosenblum, Lewis, & Joseph, 1993; Magura, Rosenblum, & Joseph, 1992 (not evaluated)
Basic Rigor McMillan, Lapham, & Lackey, 2008 (not evaluated) (not evaluated)
Basic Rigor Messina et al. 2010; Saxena et al. 2014 (not evaluated)
Basic Rigor Messina, Burdon, & Prendergast, 2006 (not evaluated) (not evaluated)
Basic Rigor Miller, Miller, & Barnes, 2015 (not evaluated) (not evaluated)
High Rigor Nyamathi et al. 2015
Basic Rigor Owens et al. 1997 (1) (not evaluated) (not evaluated)
Basic Rigor Owens et al. 1997 (2) (not evaluated) (not evaluated)
Basic Rigor Pelissier et al. 2000-2003 (1)
Basic Rigor Pelissier et al. 2000-2003 (2)
Basic Rigor Porporino et al., 2002 (not evaluated) (not evaluated)
Basic Rigor Prendergast et al. 2011 (not evaluated)
Basic Rigor Prendergast, Hall, et al. 2004
High Rigor Sacks et al. 2012 (not evaluated)
Basic Rigor Sacks et al., 2012 (not evaluated) (not evaluated)
Basic Rigor Scott & Dennis 2012 (not evaluated)
Basic Rigor Stein et al. 2010 (not evaluated) (not evaluated)
Basic Rigor Tunis, 1994; Tunis, 1995; Tunis et al., 1996 (not evaluated) (not evaluated)
Basic Rigor Welsh 2002- 2009; Welsh & Zajac 2013 (not evaluated)
Basic Rigor Wexler et al 2005 (not evaluated) (not evaluated) (not evaluated)
High Rigor Wexler, Melnick, and Cao 2004
Basic Rigor Woodall et al. 2007 (not evaluated)
Basic Rigor Zanis et al. 2003 (not evaluated) (not evaluated)