Treatment during transition from prison to community and subsequent illicit drug use

Transcription

1 Journal of Substance Abuse Treatment 28 (2005) Treatment during transition from prison to community and subsequent illicit drug use Clifford A. Butzin, (Ph.D.)T, Steven S. Martin, (M.A.), James A. Inciardi, (Ph.D.) Center for Drug and Alcohol Studies, University of Delaware, Newark, DE 19716, USA Received 15 May 2004; received in revised form 21 February 2005; accepted 25 February 2005 Abstract This study examined the effects of postrelease transitional therapeutic community treatment on the drug use and employment rates of drug involved prisoners in the Delaware corrections system followed for up to 5 years after release. A comparison group received standard postrelease supervision. Abstinence rates were 32.2% in the treatment group and 9.9% in the no-treatment group, and the treatment group had a higher overall proportion of time free of drug use. Time to relapse was a mean of 28.8 months in the treatment group versus 13.2 months in the no-treatment group. Relapse was defined as any use of any drug and was confirmed by urinalysis. Positive effects were seen even for those who did not complete treatment. The treatment group had a significantly higher rate of employment after leaving work release (54.6%) than did the no-treatment group (45.4%). Treatment during the transitional period between prison and community showed substantial and persistent benefits even for a cohort marked with extensive criminal history, low rates of marital bonds, and substantial unemployment. D 2005 Elsevier Inc. All rights reserved. Keywords: Treatment in corrections; Substance use; Treatment outcomes; Employment; Relapse 1. Introduction A major emphasis in both federal and state correctional systems has been to establish treatment programs for inmates with histories of drug use. Although a variety of approaches have been implemented, the one that has been most used and that has received the most attention from researchers is the therapeutic community (TC) modified for the prison environment (Inciardi, Martin, & Surratt, 2001). Evaluations of therapeutic communities within correctional systems have generally found positive outcomes in terms of recidivism at 1 3 years following release (Hiller, Knight, & Simpson, 1999; Inciardi, Martin, Butzin, Hooper, & Harrison, 1997; Knight, Simpson, & Hiller, 1999; Martin, Butzin, Saum, & Inciardi, 1999; Pelissier et al., 2000; Wexler, Melnick, Lowe, & Peters, 1999). These studies have variously defined outcomes from the criminal justice T Corresponding author. Tel.: ; fax: address: butzin@udel.edu (C.A. Butzin). perspective as return to custody or new arrests, with both cross-sectional and longitudinal analyses. Two studies have extended these positive findings 5 years after release in examinations of data from California (Prendergast, Hall, Wexler, Melnick, & Cao, 2004) and Delaware (Inciardi, Martin, & Butzin, 2004). Although it is a plausible assumption that the recidivism benefits are an indication of lessened drug use, illicit drug use has been less often examined as an outcome, most likely because those variables are not amenable to use of official records as outcome measures. Those studies that have examined drug use outcomes (Martin et al., 1999; Inciardi et al., 2004; Prendergast et al., 2004) have done so in crosssectional analyses with drug use as a dichotomous outcome and found somewhat different results. The first 2 studies found a significantly higher proportion of those with TC treatment experience remained drug free both 3 and 5 years after release. Prendergast et al. (2004) found no significant differences in the proportion of those reporting heavy drug use in the fifth year after release. The studies differed in the /05/$ see front matter D 2005 Elsevier Inc. All rights reserved. doi: /j.jsat

2 352 C.A. Butzin et al. / Journal of Substance Abuse Treatment 28 (2005) mix of treatment experience, as well as the drug use outcome measure. Treatment experiences examined in the various studies of correctional TC s have varied between treatment within prisons, treatment during work-release transition from prison, and treatment in the community after prison release. Most of the studies, including Prendergast et al., have had at least a preponderance of participants from treatment programs within prison. In contrast, in the Delaware sample, all treatment respondents participated in work-release treatment, with a sizeable minority also participating in treatment within prison. Thus, although there are a range of reports consistently supporting the benefits of correctional TC treatment with respect to recidivism, the data with respect to drug use is less extensive and consistent. The inconsistency is plausibly because differences in the outcome measures and the treatment mix. Participation in treatment during the period of transition from prison back to the community, as opposed to treatment inside prison, has been shown to be particularly effective (Butzin, Martin, & Inciardi, 2002; Wexler et al., 1999), but other explanations remain. The effects of treatment upon drug use have not been the subject of longitudinal analyses, nor have there been any quantitative measures of drug use. Thus, the main goal of this study is an extension of the analyses of the Delaware data, particularly a longitudinal examination of the impact of work-release treatment upon subsequent use of illicit drugs, measured both qualitatively and quantitatively. A second goal of the study is to place the impact of treatment in the context of other variables associated with recidivism and relapse for those in correctional treatment (Gendreau, Little, & Coggin, 1996), including age and criminal and drug-use histories. Additional contextual variables of interest are suggested by theories of criminal desistance (Laub & Sampson, 2001; Warr, 1998) that stress the importance of social bonds such as employment and marriage. The relationship between unemployment and negative outcomes, specifically crime and drug use, is both theoretically and empirically supported (Laub & Sampson, 2001; O Connell, 2003; Petersilia, 2003; Platt, 1995; Uggen, 2000). Most treatment programs place at least some emphasis upon vocational issues, and often employment status is seen as a plausible outcome in evaluating treatment programs (Fiorentine, 1998). An additional interest is in the subsequent treatment experiences of the participants. Within substance abuse research, the importance of understanding recovery patterns has been advanced by the study of addiction and treatment careers (Hser, Anglin, Grella, Longshore, & Prendergast, 1997). Work on btreatment careersq by Anglin, Hser, and Grella (1997) suggests that an understanding of treatment outcomes also requires an understanding of the treatment history of clients. It is an extension of the common finding of an association between more extensive time in treatment with more favorable outcomes (Hubbard et al., 1989). In particular, the favorable consequences of increased time in treatment can come from not only retention in one program but also from time spent in multiple programs over time. Finally, the study also examines differences in outcomes associated with differential treatment experiences of those who participated in work-release treatment. Because, as noted earlier, some of the Delaware sample also participated in treatment within prison, that participation provides a contextual variable for the evaluation. Also of interest is the impact of a partial treatment course represented by those who failed to complete the work-release program. Finally, a comparison is provided between those who participated in an aftercare program after graduation and those who graduated before such a program was in place. The latter comparison is particularly relevant to the more general question of the value of posttreatment continuing care (McKay, 2001) Delaware work-release treatment program The treatment program during the state s period of standard work release for the last 6 months of prisoners sentences has been the centerpiece of the treatment efforts within the Delaware correctional system (details of the program have been presented by Hooper, Lockwood, & Inciardi, 1993; Inciardi, Lockwood, & Martin, 1994; Lockwood, Inciardi, & Surratt, 1997) and is the central focus of the present study. Work release is a form of partial incarceration whereby inmates who are approaching their release dates are permitted to work for pay in the free community, but must spend their nonworking hours either in the institution or, more commonly, in a community-based work-release facility. The Delaware work-release TC examined here is physically adjacent to the regular workrelease center, and correctional officers provide security for both buildings. Both the regular work release and the workrelease TC are 6-month programs. The transitional workrelease TC is similar to that of the traditional TC, with a bfamily setting,q removed from as many of the external negative influences of the street and inmate cultures as is possible. The clinical regimen is modified to address security concerns and the correctional mandate of work release to prepare clients for employment in the community (De Leon, 1997). During the first 3 months, the TC participants are not employed in the outside community, but for the final 3 months, follow the same regimen as regular work-release participants and are allowed access to the community for employment opportunities. To put the transitional work-release program in context, treatment within the Delaware criminal justice system occurs at three levels: TC treatment within prison, TC treatment during work release, and a subsequent aftercare program. Within prison, TC treatment occurs in a separate pod of the facility for a 12-month period. Many of the participants had previously participated in treatment within the prison, but for most the work-release treatment was their primary treatment within the correctional system. The

3 C.A. Butzin et al. / Journal of Substance Abuse Treatment 28 (2005) aftercare program consists of weekly outpatient group meetings and one day each month at the facility for the first 6 months of subsequent probation. Note that although in the correctional treatment literature, the term aftercare is often used to refer to any treatment after that in prison, here it specifically refers to a continuing care program that is an extension of the work-release program. 2. Methods In the Delaware correctional system, those reaching the final 6 months of their sentence may be classified as approved for work release with a recommendation for drug treatment, based upon criminal history and correctional counselor interviews. The present sample is drawn from those so classified between 1991 and Because the number of prisoners in that category exceeded the capacity of the treatment programs during that period, those eligible were assigned to either treatment or to regular work release, a bno-treatmentq group. Priority for entering the program was given to graduates of treatment programs within the prison and to those with judicial sentencing orders that required treatment participation as a condition for release. The research complied fully with the special protections for prisoners as research subjects. The research protocol included baseline and multiple follow-up interviews, as well as urine testing at each interview. The baseline interview was administered at the point of an inmate s transfer to work release from prison. The first follow-up occurred at 6 months, corresponding with completion of work release. Subsequent interviews were conducted 18, 42, and 60 months after baseline. Interviews at baseline and each subsequent follow-up were lengthy, with participation in the research project compensated with up to $50 at each of the testing intervals $25 for completing the questionnaire and $25 for giving a urine sample. Most relevant for the present analyses are three measures of illegal drug use at each interview. Each of the three types of measures was examined to identify the first use of illicit drugs after release from prison. First, respondents were directly asked whether they had used any illegal drugs since the last interview, and if so, when that use had first occurred, and the frequency of use (scaled from 0, no use, to6,used more than once a day). Secondly, each interview asked respondents to recall where they were living when they were last interviewed, and asked to report on the frequency (again scaled from 0, no use, to6,used more than once a day) of their use of illegal drugs while they were living there. The process was then repeated for the next residence until the complete period was described. Thirdly, at each interview urine samples were requested and tested for the presence of opiates, marijuana, cocaine, barbiturates, phencyclidine, and amphetamines. If there was indication of drug use, the time of initial relapse to drug use was determined by the first occurring indicator of drug use in the 3 measures. When the first report of use was during a residential period, the initial date of that period was used as the relapse date. If each of the measures indicated no use, the respondent was classified as abstinent through the last available interview date. The analyses reported below each initially compared those participants who had any participation in transitional treatment versus those who did not. This represented the main question of the study the longitudinal effects of workrelease treatment. Given significant effects of treatment, subsequent analyses then explore differences among the treated group in terms of completion of the treatment program and participation in aftercare. Additional comparisons of treatment graduates with and without aftercare were possible because the aftercare component was not operational until 1996, whereas the other stages of treatment had been implemented several years earlier. Of the original cohort of 1319 interviewed at their prison release, 1247 (94%) were subsequently interviewed. The primary mode of analysis was survival analyses, which allowed the inclusion of cases that did not relapse to drug use, bcensoredq cases, and the full cohort of 1247 with any follow-up. Statistical analyses of data describing time to relapse used Kaplan Meier methods for the initial test and description of treatment effects, with subsequent multivariate tests using Cox multivariate regression methods. The Cox proportional hazards regression models included, in addition to treatment status, variables representing demographics (sex, race, age, and marital status), employment after release, and criminal, drug use, and drug treatment histories. They were selected to include those often found to be associated with treatment failure and recidivism (Gendreau et al., 1996; Wexler et al., 1999). The indicator for criminal history was number of incarcerations. Previous drug use was measured as the report at baseline of frequency of drug use in the 3 months before incarceration on the 6-point scale described above. Previous participation in drug treatment was measured as a direct question at baseline of whether they had ever been in a drug treatment program before this incarceration. Employment was the report of working at least 30 hours per week after leaving work release. Table 1 Demographics and history for treatment and no-treatment groups Characteristics No work-release treatment Work-release treatment n Male sex (%) African American (%) Age (mean) Ever married (%) Prior arrests (mean) Any treatment before incarceration (%) Any treatment during incarceration (%) Prior drug use frequency a (mean) a Response categories 4 and 5 are bseveral times a weekq and bonce a day,q respectively.

4 354 C.A. Butzin et al. / Journal of Substance Abuse Treatment 28 (2005) Group means and percentages on the baseline covariates are shown in Table 1. The btreatmentq and bno-treatmentq groups differ significantly on the frequency of drug use prior to incarceration, with the treatment group reporting more frequent drug use, as well as a higher rate of participation in treatment within prison. Less than 5% of the cohort reported themselves as Asian, Hispanic, or botherq and were excluded from the sample. 3. Results 3.1. Time to relapse The time from release from prison to the first reported or tested instance of illicit substance use, or the last contact in the case of no substance use, was compared between those respondents with and without work-release treatment using Kaplan Meier survival analyses. Relapse rates were 32.2% in the treatment groups and 9.9% in the no-treatment group. Time to relapse to drug use was significantly longer for those respondents who participated in transitional treatment than those not in treatment (means of 28.8 vs months, respectively, p b.001, log rank test). Thus, those who participated in work-release treatment were over three times more likely to remain drug free, and the time until relapse was about twice as long among the treatment participants. The survival curves from this Kaplan Meier analysis are shown in Fig. 1. Each of the curves shows a precipitous decline in the first 2 years after release from prison, then a much more modest decline through the rest of the 5-year period. There is a clear and consistent separation between the survival curves for those with and without treatment, with the treatment group establishing a substantial higher probability of abstinence after the first year and maintaining that difference through the remainder of the follow-up period. The impact of urinalyses at the interview points is seen in the dips in the curves at 18 and 42 months. Relapse for this analysis was defined as any use of any illegal drug and was measured objectively through urinalysis. As such, it is a very stringent definition, including what may be only an occasionally lapse as a full-blown relapse. Further analysis was conducted with a definition of relapse as heavy drug use, rather than a definition of the event as any positive report or test, defined as a report of at least weekly use. The pattern of significance was the same for this analysis, although, of course, the levels of abstinence were much higher, 50.5% for the treatment group and 28.5% for the no-treatment group. Because that definition precluded the use of urinalysis results, however, all analyses reported use the stringent definition of relapse as any test or report of use of illicit drugs. The effect of treatment participation was also statistically significant in the context of the other covariates in a Cox proportional hazards regression analysis. There were 1122 respondents with complete data included in this analysis. As shown in Table 2, treatment participation was the largest predictor, with age, frequency of prior drug use, and employment after work release as additional significant predictors. Rate of employment was also Treatment Any Treatment Probability of no drug use.6.4 No Treatment Months after release Fig. 1. Survival function for drug use as a function of work-release treatment.

5 C.A. Butzin et al. / Journal of Substance Abuse Treatment 28 (2005) Table 2 Cox proportional hazard regression results for time to lapse Independent variables B SE (B) Significance Exp. (B) Age b Female sex African American Married Prior incarcerations Prior drug use frequency b Any treatment prior to current incarceration Treatment in prison Work-release treatment b Employed after release b examined separately with respect to treatment. The treatment group had a significantly higher rate of employment after leaving work release (54.6%) than did the no-treatment group (45.4%). The results from the Cox regression analysis showed that, even when controlling for the other predictors, participating in the transitional treatment program essentially halved the odds of relapsing. Older participants and those employed were significantly less likely to relapse. None of the other predictors, including participation in treatment programs within prison, and participation in treatment programs before the incarceration were significant. Each second-order interaction between participation in work-release treatment and the other variables was also tested. None was statistically significant. Given the significant effect of work-release treatment, additional analyses were conducted with the respondents divided into four groups: those with no work-release treatment (n = 303), those who failed to complete the treatment program (n = 275), those who completed but did not participate in aftercare (n = 234), and work-release treatment graduates with aftercare participation (n = 310). Cox proportional hazards analysis were conducted with these four levels of the treatment predictor and the same set of other variables, with one exception, as in the analysis of Table 2. The exception was the inclusion of the year of release from prison as a predictor. Because the graduate with aftercare subgroup entered the study later than did the other groups, its effects could have been confounded with changes in the program or cohort over time. The additional predictor of year of study entry was not statistically significant ( p =.08). The same pattern of significance for the other predictors was found with the four-group analysis as with the above two-group analysis. The plot of survival probability from the regression analysis is shown in Fig. 2. The curves are the linear estimates at the mean values of each of the other predictors. The most striking aspect is that all three of the treatment groups curves lie significantly ( p b.01, log rank tests) above that for the no-treatment group. Thus, even the group that participated bunsuccessfullyq in workrelease treatment was less likely to relapse than those who did not participate in treatment. The curve for graduates that participated in aftercare was consistently above that 1.0 Treatment subgroups Probability of no drug use Complete & aftercare Complete, no aftercare No complete No treatment Months after release Fig. 2. Survival function for drug use as a function of work-release treatment subgroups.

6 356 C.A. Butzin et al. / Journal of Substance Abuse Treatment 28 (2005) for those without aftercare, although the difference was not statistically significant Amount of time abstinent Each reported residential period for the last three interviews was examined and categorized with respect to whether drug use was reported during that period. The period during work release was excluded because the treatment group would have been at less risk of substance use. Periods of incarceration were also excluded, because opportunity for drug use should have, at least, been greatly reduced during incarceration. The sample was, thus, restricted to those with at least the second follow-up and whose follow-up period has some period in which they were not incarcerated, resulting in 855 cases. The treatment group again showed significantly less drug use. Over the complete follow-up period, the treatment group had a significantly higher proportion of their time abstinent from drug use (53%) than did the no-treatment group (38%, p b.0001, by F test). Fig. 3 shows that proportion of abstinent time over each of the three latter interview periods, for the no-treatment group and the three treatment subgroups. In both the first two periods after work release, the differences between the four groups were statistically significant ( p b.001, by F tests). For the third interview period, beginning 3 years after work release, the four groups were not significantly different from each other ( p =.08), although the treated group as a whole was significantly different than the no-treatment groups (43% vs. 35%, respectively, p b.03 by F test). Table 3 Ordinary least squares regression results predicting amount of time abstinent from use of illegal drugs Independent variables B SE (B) h t Significance Age Female sex African American Prior incarcerations Prior drug use b.001 frequency Any treatment prior to current incarceration Treatment in prison Work-release treatment b.001 Employed after release b.001 This effect of treatment on the time abstinent was also tested in a multivariate context using ordinary least squares regression with the same set of control variables as the previous survival analyses. As can be seen in Table 3, treatment participation remained significantly related to total time abstinent, even in the context of the other predictor variables. In this analysis, age was no longer a significant predictor. Number of previous incarcerations was associated with less time abstinent and participating in treatment within prison was significantly associated with more time abstinent Subsequent treatment We also examined postaftercare treatment in these samples. The no-treatment, graduate, and graduate with No Treatment Complete no aftercare No complete Complete & aftercare Proportions Year 1 Years 2-3 Years 4-5 Years after work release Fig. 3. Proportion of months not using drugs.

7 C.A. Butzin et al. / Journal of Substance Abuse Treatment 28 (2005) aftercare groups did not differ significantly in subsequent treatment participation (37%, 35%, and 32%, respectively). However, significantly more of the dropout group was in later treatment (49%). Those who had any subsequent treatment experience were significantly less likely to have remained abstinent (13.8%) than those with no subsequent treatment (31.6%, p b.001) and to have less time abstinent (45.7% vs. 50.8%, p =.04). Moreover, a substantial amount of all subsequent treatment occurred as part of an arrest or probation violation. Among this sample, only 34% of those who had any subsequent treatment received that treatment exclusively outside the correctional system. For 53% of those who had subsequent treatment, that treatment was entirely from programs within prison systems. 4. Discussion Participation in a work-release TC during the transitional period between prison and the community had a substantial impact on the timing, incidence, and duration of subsequent drug use. The proportion of those treated who remained abstinent was approximately three times that for those without treatment. For those with treatment, the time to relapse was approximately twice as long as for those without treatment. Finally, cumulative time abstinent for those with treatment was about a third more than for those without treatment. Those who did not complete treatment showed reduced but similar outcomes as those who completed treatment. Treatment participants were free to leave the facility after the first 3 months of the program, which was the most common time for treatment dropouts. As a result, the dropout group had a mean stay of about 3 months, rather than the 6 months required for treatment graduation. Thus, the majority of the dropout group was still exposed to a substantial amount of treatment. In addition, that group was also exposed to more subsequent treatment, largely via brecyclingq through the correctional treatment system. The effect of cumulative treatment experience is intricately confounded by the pairing of treatment with the criminal justice system. Thus, the present data do not provide a rich enough history of treatment experiences before this incarceration to allow a truly relevant test of the cumulative treatment hypothesis. As with several previous studies, treatment within prison, as opposed to during the transition from prison, had a much smaller impact on outcomes. Treatment while in prison had significant effects only upon the overall time of subsequent drug use; but not on the rate of abstinence or the time to relapse. One could speculate that the superior impact of treatment in the transitional period is two-fold: The contextual milieu of that treatment is much more similar to the community context in which the participants must maintain their sobriety and/or the timing of the treatment provides support when risks of returning to previous behaviors are much stronger. Treatment also was related to increased likelihood of subsequent employment, and in turn, employment was also associated with decreased drug use. Practically, this effect is sensible in the context of the realistic employment expectations for this cohort. The group as a whole had a very high rate of preincarceration unemployment (58%). The life-course perspective in criminology has depicted desistance from illegal behavior largely as a product of social ties, most clearly demonstrated by job stability and marital attachment (Laub & Sampson, 2001; Warr, 1998). The extensive criminal history of this cohort is supportive of that notion, given there is only a small minority that have ever been married and who had full-time employment of any sort. The evidence of the effectiveness of a transitional TC is thus particularly impressive, although also consonant with the perspective of De Leon (1997, 2000). That perspective is that the focus of treatment is needed to be upon the whole person, rather than the specifics of drug abuse; that bhabilitationq rather than brehabilitationq is the goal, given the individual s history of negative patterns of behavior. Acknowledgments This research was supported by grants DA06124 and DA06948 from the National Institute on Drug Abuse. The authors thank Gregory Postle for his assistance in preparation of the data and the manuscript. References Anglin, M. D., Hser, Y., & Grella, C. E. (1997). Drug addiction and treatment careers among clients in the Drug Abuse Treatment Outcome Study (DATOS). Psychology of Addictive Behaviors, 11, Butzin, C. A., Martin, S. S., & Inciardi, J. A. (2002). Evaluation component effects of a prison-based treatment continuum. Journal of Substance Abuse Treatment, 22, De Leon, G. (1997). Therapeutic communities for special populations and special settings. Westport, CT7 Greenwood. De Leon, G. (2000). The therapeutic community: Theory, model, and method. New York7 Springer. Fiorentine, R. (1998). Effective drug treatment: Testing the distal needs hypothesis. Journal of Substance Abuse Treatment, 15, Gendreau, P., Little, T., & Goggin, C. (1996). A meta-analysis of the predictors of adult offender recidivism: What works. Criminology, 34, Hiller, M. L., Knight, K., & Simpson, D. D. (1999). Prison-based substance abuse treatment, residential aftercare and recidivism. Addiction, 94, Hooper, R. M., Lockwood, D., & Inciardi, J. A. (1993). Treatment techniques in corrections-based therapeutic communities. Prison Journal, 73, Hser, Y., Anglin, M. D., Grella, C., Longshore, D., & Prendergast, M. L. (1997). Drug treatment careers: A conceptual framework and existing research findings. Journal of Substance Abuse Treatment, 14, Hubbard, R. L., Marsden, M. E., Rachal, J. V., Harwood, H. J., Cavanaugh, E. R., & Ginzburg, H. M. (1989). Drug abuse treatment: A national study of effectiveness. Chapel Hill, NC7 University of North Carolina Press.

8 358 C.A. Butzin et al. / Journal of Substance Abuse Treatment 28 (2005) Inciardi, J. A., Lockwood, D., & Martin, S. S. (1994). Therapeutic communities in prison and work release: Some clinical and policy implications. In F. M. Tims, G. De Leon, & N. Jainchill (Eds.), Therapeutic community. Advances in Research and Application: Research Monograph No. 144 (pp ). Rockville, MD7 National Institute on Drug Abuse. Inciardi, J. A., Martin, S. S., & Butzin, C. A. (2004). Five-year outcomes of therapeutic community treatment of drug-involved offenders after release from prison. Crime & Delinquency, 50, Inciardi, J. A., Martin, S. S., Butzin, C. A., Hooper, R. M., & Harrison, L. D. (1997). An effective model of prison-based treatment for drug-involved offenders. Journal of Drug Issues, 27, Inciardi, J. A., Martin, S. S., & Surratt, H. S. (2001). Therapeutic communities in prisons and work release: Effective modalities for drug-involved offenders. In B. Rawlings, & R. Yates (Eds.), Therapeutic communities for the treatment of drug users (pp ). London7 Jessica Kingsley. Knight, K., Simpson, D. D., & Hiller, M. L. (1999). Three-year reincarceration outcomes for in-prison therapeutic community treatment in Texas. Prison Journal, 79, Laub, J. H., & Sampson, R. J. (2001). Understanding desistance from crime. In M. Tonry (Ed.), Crime and justice (pp. 1 69). Chicago7 University of Chicago Press. Lockwood, D., Inciardi, J. A., & Surratt, H. L. (1997). CREST outreach center: A model for blending treatment and corrections. In F. M. Tims, J. A. Inciardi, B. W. Fletcher, & A. M. Horton (Eds.), The effectiveness of innovative approaches in the treatment of drug abuse (pp ). Westport, CT7 Greenwood. Martin, S. S., Butzin, C. A., Saum, C., & Inciardi, J. A. (1999). Three-year outcomes of therapeutic community treatment for drug-involved offenders in Delaware: From prison to work release to aftercare. Prison Journal, 79, McKay, J. R. (2001). Effectiveness of continuing care interventions for substance abusers. Addiction, 96, O Connell, D. J. (2003). Investigating latent trait and life course theories as predictors of recidivism among an offender sample. Journal of Criminal Justice, 31, Pelissier, B., Rhodes, W., Saylor, W., Gaes, G., Camp, S., Vanyur, S., et al. (2000). TRIAD Drug Treatment Evaluation Project final report of three-year outcomes: Part 1. Washington, DC7 Office of Research and Evaluation, Federal Bureau of Prisons. Petersilia, J. (2003). When prisoners come home: Parole and prisoner reentry. New York7 Oxford University Press. Platt, J. L. (1995). Vocational rehabilitation of drug abusers. Psychological Review, 117, Prendergast, M. L., Hall, E. A., Wexler, H. K., Melnick, G., & Cao, Y. (2004). Amity prison-based therapeutic community: Five-year outcomes. Prison Journal, 84, Uggen, C. (2000). Work as a turning point in the life course of criminals: A duration model of age, employment and recidivism. American Sociological Review, 67, Warr, M. (1998). Life-course transitions and desistance from crime. Criminology, 36, Wexler, H. K., Melnick, G., Lowe, L., & Peters, J. (1999). Three-year reincarceration outcomes for Amity in-prison therapeutic community and aftercare in California. Prison Journal, 79,