FORUM on Corrections Research

Correctional Service of Canada Substance Abuse Programs: OSAPP, ALTO, and Choices

Cathy Delnef¹
Research Branch, Correctional Service of Canada

Beginning in the late 1980’s, the development and introduction of three major substance abuse treatment programs represented an important step in the Correctional Service of Canada’s (CSC) efforts to provide high quality effective treatment for federal offenders in Canada. The design of each program was based on the principles of effective correctional treatment and a model of intervention that linked substance abuse need assessments with empirically based substance abuse programming techniques. These three programs are: the Offender Substance Abuse Pre-Release Program (OSAPP), the Community Correctional Brief Treatment, Relapse Prevention and Maintenance Program (referred to as Choices), and the Programme prélibératoire en toxicomanie (referred to as ALTO). This article describes these programs, and provides characteristics of the participants. It provides also a synopsis of an outcome evaluation of their effectiveness.

Background

The implementation of OSAPP, Choices and ALTO followed a deliberate effort on the part of CSC to overhaul its offender reintegration programming strategy. There was a growing awareness among program planners within the Service that substance abuse represented a major criminogenic need among federal offenders. Evidence regarding the proportion of offenders with substance abuse problems and existing research on the link between substance abuse and criminal activity were major background factors leading to the initiation of the Task Force on the Reduction of Substance Abuse by CSC in 1989.² The final report of the Task Force recommended that a comprehensive approach be introduced to addressing substance abuse within the Service.

Following the tabling of the Task Force report, the Program Development Division of CSC issued a Model for the Provision of Substance Treatment.³ An important tenet of the model was the recognition that substance abuse program was integral to the overall goal of reintegrating offenders as “law-abiding citizen”. Hence, substance abuse interventions were viewed as important tools for reducing recidivism. Accordingly, the model focused treatment attention on offenders for whom substance abuse was regarded as a criminogenic factor. The selection of offenders for interventions was to be based on systematic assessment using technology that would adequately differentiate offenders at admission according to the severity of their substance abuse needs. As the substance abuse treatment delivery system in CSC evolved, offender assessment has continued to develop as a critical driving force. Program developers have argued convincingly that valid self-report substance abuse assessment techniques should be used at the program referral stage and periodically throughtout delivery of treatment to measure offender performance.⁴

Another critical component of the proposed model concerned the techniques used to deliver substance abuse interventions. It was recognized that many of the treatment methods that were supported by empirical evidence of effectiveness were sometimes at variance with other approaches that tended to be more prevalent in the wider substance abuse treatment community. More importantly, there was a relative absence of programs based on techniques which were supported by the correctional treatment literature.

Miller who first reviewed a number of controlled studies that showed promising evidence of effectiveness for these approaches relative to more traditional treatments.⁵ Coincident with these empirical developments was a growing optimism that criminal offenders tended to be very good candidates for substance abuse treatment. It is within this context that OSAPP, ALTO, and Choices were developed and introduced as alternatives that employed state-of-the-art correctional programming techniques for addressing the substance abuse needs of federal offenders.

The three programs are primarily based on a social-learning model of substance abuse intervention. As such, the theoretical models on which the programs were constructed are “in-step” with the dominant model employed by CSC to intervene with offenders in a variety of criminogenic need domains in both institutional and community settings.

OSAPP

This program targets offenders who exhibit moderate to severe substance abuse problems. OSAPP has been widely implemented within CSC’s institutions in the Atlantic, Ontario, Prairies, and Pacific regions (Quebec uses ALTO).

Initially, OSAPP was assessed in terms of its ability to affect offender performance during the program using pre-test/post-test assessments. This first pilot evaluation of the program with federal inmates indicated that participants made significant gains on drug related knowledge and appropriate attitudes toward alcohol and drug use.⁶ The participants also provided positive feedback regarding their satisfaction with the program. Following the pilot, the program was revised to take in account feedback from participants as well as from the group leaders who delivered the program.

The revised version of OSAPP, reflecting the program that is currently in use within CSC, was piloted in 1989 in the Ontario region.⁷ The evaluation involved an examination of pre-test/post-test changes as well as qualitative information about the program process. The participants showed significantly positive changes on a number of measures (attitude toward alcohol and drug use, knowledge and attitudes toward employment, and problem solving). The remaining measures, including knowledge of drugs, communication, assertiveness, showed non-significant trends in the expected direction. As in the first pilot, feedback from participants indicated highly favourable attitudes regarding the program. The study was based on a very small sample of inmates (n = 15) who participated in the program at Joyceville institution, in Ontario.

Another study extended the analysis to include post-release outcome variables as part of the evaluation of OSAPP.8 This study employed a considerably larger sample (n = 283) of participants who completed the program in Bath institution, in Ontario, between 1990 and 1992. An important contribution to this study involved the use of recidivism data in the analysis of pre-test/post-test performance data. The authors were able to demonstrate that OSAPP participants who had the best program performance were also the most successful candidates following release.

Generally, the early research on OSAPP has been highly positive to date. The data suggest that the program participants make positive gains on various program target measures over the course of their participation in OSAPP. In 1999, a study was undertaken by T3 Associates to examine the impact of the CSC substance abuse programs on a number of outcomes.⁹ These included program performance data as measured by client self-reports, granting of release, and post-release outcomes as assessed through official recidivism records. The study also examined how well the program performance data predicted post-release outcomes.

Program participants

The national database of OSAPP participants analyzed by T³ Associates contained 2,731 offender records at the time of the latest evaluation study. This sample was inclusive of all the program participants whose assessment and program performance information was entered in the national CSC database. Of those who were enrolled in the OSAPP program, 2,432 offenders completed the program. In 1994/95 and 1995/96, the Ontario region had the highest OSAPP participant rates with 39.9% and 42.7% respectively, followed by the Atlantic region with 46.7% and 32.2%, the Prairies region with 37.8% and 30.8%, and the Pacific region with 34.9% and 30.8%.

The participation rates were noticeably lower in 1996/97. The Atlantic and the Ontario regions had the lowest participation rate (7.6% and 12.9% respectively). The Pacific region had the highest participation rate (30.0%), followed by the Prairies region (26.9%) (12.9%).

Characteristics of OSAPP participants

The average age of all program participants was close to 32 years with the majority (64.8%) ranging between 20 and 34 years of age. Nearly 17% were of Aboriginal status. Over one-third of participants (36.7%) had achieved education levels of grades 9 or 10, while 41.1% had completed between grades 11 and 13. The average sentence length was 4.2 years and 7.9% were serving life sentences. According to the risk level for recidivism, close to 35.5% were in the low risk category, 39.0% were in the moderate risk category and the remaining 25.5% were identified as high risk.

Offence types

T³ Associates categorized the program participants by the types of offences according to a hierarchy of violent crime, robbery, drug offence, break and enter/theft, and other. For example, an offender convicted of both a violent and robbery offence would be ranked in the more serious offence category of violent. According to this approach of offence rating, results showed that 48.4% were incarcerated for a violent crime (e.g., assault, sexual assault, etc.) while almost 24% were serving a sentence for robbery, 12.1% for break and enter/theft and about 10% for a drug-related offence.

Substance abuse severity

Using the Alcohol Dependence Scale (ADS), the Problems Related to Drinking Scale (PRDS), and the Drug Abuse Screening Test (DAST) scores as indices of substance abuse severity, approximately 40% to 45% of the offenders in the sample reported moderate to severe alcohol problems and 59.4% moderate to severe drug problems. Afive level composite index of substance abuse severity was constructed to classify offenders according to the highest level of substance abuse problem (i.e., alcohol, drugs, or combined alcohol and drugs) for each offender on any of the three screening measures. The results of this composite index were as follow: 12.8% of offenders were identified as having low level of substance abuse problems, 33.8% had moderate problems, 36.2% had substantial problems, and 14.1% were evaluated as having severe substance abuse problems. Interestingly, 3.0% of the offenders in the sample were not assessed as having either an alcohol or drug problem.

Taken together, these findings indicated that approximately 85% of the participants presented with alcohol problems, drug problems, or combined alcohol and drug problems of sufficient severity to warrant participation in OSAPP. The remaining 15% were accepted into the program on the basis of additional information obtained from other sources (e.g., case file information, interviews, etc.) that identified them as appropriate program candidates.

Offender performance

The offenders increased their knowledge of the consequences of alcohol and/or drug use. They enhanced their understanding of how other people affect their use, and gained skills for declining offers to use substances. They increased their ability to communicate with peers about managing their substance abuse problems, and increased problem-solving skills necessary for controlling substance abusing behaviour.

Post-release outcomes

An important set of findings reported by T³ Associates concerned the combined or interactive effects of participating in OSAPPalong with other CSC programs. OSAPP participants who attended other substance abuse programs in the community (i.e., Alcoholics Anonymous (AA), Neurotics Anonymous (NA) had superior outcomes than offenders who did not participate in such programs. OSAPP appeared to combine very efficiently with AA/NA and other community programs to promote lower recidivism.

A total of 1,216 OSAPPparticipants were considered eligible for inclusion in the follow-up sample. Only offenders who had been released at a point in time that allowed a minimum of 12 months of post-release were included in the follow-up sample. A breakdown of post-release outcomes for three categories of OSAPP participation: all participants, only program completers, and only program dropouts. The rates for reconvictions varied considerably when OSAPP completers were compared to dropouts. OSAPPcompleters had a lower proportion of reconviction (15.1%) when compared to dropouts (19.7%). There was also a 53% reduction in new convictions for violent offences.

In terms of differential outcomes across sub-groups of offenders, the largest effects of OSAPP on recidivism were associated with offenders who had the highest levels of substance abuse severity and the lowest level of risk. Violent offenders also showed very good response to the program (9.4% of OSAPP participants reconvicted compared to 16.7% of their matched comparison cases).

Findings

OSAPP seems to provide measurable benefits which can contribute significantly to the offender reintegration efforts of the Correctional Service of Canada. Three sets of important findings are emerging:

• OSAPP targets an appropriate group of offenders.
  
• OSAPP appears to positively affect post-release readmission rates and survival time in the community.
  
• OSAPP appears to affect the release outcomes of those who are most in need of the intervention.

ALTO

The ALTO Program was developed for francophone offender to provide a comparable intervention to the national Offender Substance Abuse Pre-Release Program (OSAPP). It was introduced throughout the Quebec region as the primary institutional substance abuse program for offenders in that region. Both programs target offenders with serious substance abuse problems whose drug and/or alcohol use is linked to their offending. They are of similar duration, and, when possible, target offenders who are within six months of probable release.

Program participants

The national database of ALTO participants contained 1,250 offenders records at the time of the latest evaluation study.¹⁰ This sample was inclusive of all the program participants whose assessment and program performance information was entered in the CSC’s national database.

Characteristics of ALTO participants

The average age of all program participants was close to 31 years with the majority (70.4%) ranging between 20 and 34 years of age. About 2% were of Aboriginal status. The average sentence length was 4.6 years and 6.4% of participants were serving life sentences. According to the risk determination, only 16% were grouped as low risk for recidivism while 40.4% were assessed as moderate and the remaining 43.3% were identified as high risk.

Offence types

The types of offences were categorized the same way as for the OSAPPparticipants. Accordingly, the results showed that 38.2% were incarcerated for a violent crime, while 35.8% were serving a sentence for robbery, 12.4% for break and enter/theft and almost 12% for a drug-related offence.

Substance abuse severity

Examination of the distribution of alcohol severity scores generated by the ADS revealed that 21.5% of offenders in the sample reported no alcohol problems, 54.5% reported low level problems, 14.4% reported moderate problems, 6.7% had substantial problems, and 2.9% had severe alcohol problems. Offenders’ responses to the PRDS generated the following distribution: 27.0% had no alcohol problem, 24.8% had “some” problems, 16.8% had “quite a few” problems, and 31.4% had a “lot” of alcohol problems. Finally, the DAST yielded the following drug severity scores: 6.9% had no drug problem, 15.9% had low level problems, 26.5% had moderate problems, 35.7% had substantial problems, and 15.0 had severe drug problems.

Using the ADS, PRDS and DAST scores as indices of substance abuse severity, it can be concluded that approximately 25% to 30% of the offenders reported moderate to severe alcohol problems and a considerably higher 77.2% moderate to severe drug problems. As for OSAPP, a five level composite index of substance abuse severity was constructed, and the results were as follow: 10.6% of offenders were identified as having low level substance abuse problems, 26.6% had moderate problems, 44.1% had substantial problems, and 16.6% were evaluated as having severe substance abuse problems. Interestingly, 2.1% of the offenders in the sample were not assessed as having either an alcohol or drug problem.

Taken together, these findings indicate that approximately 85% of the participants presented with alcohol problems, drug problems, or combined alcohol and drug problems of sufficient severity to warrant participation in ALTO. The remaining 15% were accepted into the program on the basis of additional information obtained from other sources.

Offender performance

The offenders demonstrated statistically significant improvement on each of the measures in the battery. They increased their knowledge of the effects of alcohol and drug use. They demonstrated a better understanding of substance abuse dependence. They showed better acquisition of relapse prevention skills, and showed more control of their life in regard to substance use and a more positive perception of their ability to change their use.

Post-release outcomes

Atotal of 589 ALTO participants were considered eligible for inclusion in the follow-up sample. Again, cases were included if they had been released at a point in time that allowed for a minimum of 12 months of post-release follow-up.

Approximately half of the ALTO participants (50.9%) were readmitted during the 12 month follow-up period. Among these, over one-fifth (22.2%) were readmitted with a new conviction. ALTO completers had a lower proportion of reconvictions (21.6%) when compared to dropouts (31.6%).

Comparison to OSAPP participants

Given the similarities between the two programs, the characteristics of offenders who participated in either the ALTO or OSAPP program were examined.

• ALTO participants were, on average, about one year younger than offenders in the OSAPP group.
  
• Only about 2% of ALTO participants were Aboriginal compared to almost 17% of offenders in the OSAPP sample.
  
• Anoticeably higher percentage of offenders in the ALTO Program had previous federal admissions.
  
• While approximately one-third of OSAPP offenders were assessed as low risk for recidivism, only 16% of ALTO participants were rated the same. Consequently, close to twice as many of those inALTO were assessed in the high risk category (43.4% versus 25.5%).
  
• Both programs had similar proportions of offenders convicted of either drug, break and enter/theft, or other offence types. When compared to OSAPPoffenders, a considerably higher percentage of participants in the ALTO Program were incarcerated for robbery offences while a lower proportion were convicted of violent crimes.
• Considering the highest level of drug or alcohol abuse assessed by the DAST, ADS, or PRDS, the results clearly show that ALTO participants had more serious substance abuse problems as compared to those in the OSAPPsample. Indeed, while 60.7% of ALTO offenders were in the substantial to severe range of abuse, only 50.3% of OSAPP participants were assessed at the same levels.
  

The results show that ALTO participants were, in general, higher risk and higher need offenders compared to those in the OSAPP group.

Choices program

This program was also developed as a national substance abuse program to be offered in the Atlantic, Ontario, Prairie, and Pacific regions. It targets “low to low moderate” substance abuse problems and is delivered to offenders who are on conditional release in the community. While OSAPP has been the focus of the majority of efforts to assess the effectiveness of CSC national substance abuse programs, the community-base program Choices was the subject of an evaluation study during the initial piloting of the program.11 It is important to point out that the Choices Program consists of two components: an intensive phase delivered to offenders over a one-week, full-day or two-week, half-day period and a maintenance phase offered once a week for twelve weeks. Only those participants who successfully complete the intensive phase are eligible to attend the maintenance sessions.

Program participants

The national database of Choices participants evaluated by T3 Associates contained 724 offender records at the time of the study. This sample was inclusive of all the program participants whose assessment and program performance information was forwarded to the CSC national database.

Characteristics of Choices participants

The average age of all program participants was about 30 years with the majority (70.5%) ranging between 20 and 34 years of age. Close to 11% were of Aboriginal status including 2.9% who self-identified as Metis and the remaining 7.6% as North American Indian. Over one-third (34.5%) of participants had achieved education levels of grades 9 or 10 while 42.9% had completed between grades 11 and 13. The average sentence length was 3.5 years and 1.8% of participants were serving life sentences. According to the risk level for recidivism, 31.3% were in the low risk category, 43.6% were assessed as moderate, and the remaining 25.0% were identified as high risk.

Offence types

The types of offences were categorized in the same manner as for the OSAPPand ALTO programs. The results showed that 35.2% were incarcerated for a violent crime, 26.2% were serving a sentence for robbery, 16.6% for break and enter/theft and about 17% for a drug-related offence.

Substance abuse severity

The distribution of alcohol severity scores generated by the ADS revealed that 33.8% of offenders in the sample reported no alcohol problem, 42.4% reported low level problems, 12.6% reported moderate problems, 7.4% had substantial problems, and 3.8% had severe alcohol problems. Offenders’ responses to the PRDS were as follow: 34.5% had no alcohol problem, 28.4% had “some” problems, 14.8% had “quite a few” problems, and 22.2% had a “lot” of alcohol problems. Finally, the DAST yielded the following drug severity scores: 14.1% had no drug problem, 28.3% had low level problems, 23.8% had moderate problems, 22.1% had substantial problems, and 11.7% had severe drug problems.

Using the ADS, PRDS, and DAST as indices of severity, it can be concluded that approximately 25% to 30% of the offenders reported moderate to severe alcohol problems and a considerably higher 57.6% moderate to severe drug problems. Regarding the five level composite index, the results were as follow: 22.9% of offenders were identified as having low level of substance abuse problems, 27.0% had moderate problems, 32.7% had substantial problems, and 13.6% were evaluated as having severe substance abuse problems.

Taken together, these findings indicate that approximately 95% of the participants presented with alcohol problems, drug problems, or combined alcohol and drug problems of sufficient severity to warrant participation in Choices. The remaining 5% were accepted into the program on the basis of additional information obtained from other sources.

Offender performance

The results suggest that the program was successful in increasing the offenders’ knowledge about the effects and consequences of alcohol and drug use as well as in the development of number of skills that are deemed essential in abstaining or controlling future substance use.

Post-release outcomes

Atotal of 536 Choices participants were eligible for the follow-up sample. Cases were included if they had been released at a point in time that allowed for a minimum of 12 months of post-release follow-up. Approximately 2 out of every 5 Choices participants (42.5%) were readmitted during the 12 month follow-up period. The rate of reconvictions was 14.6%. Choices completers had a considerably lower proportion of readmissions (40.3%) when compared to non-completers (57.1%). Among offenders who completed the maintenance phase of Choices, the readmission rate was 22.7% comparatively at 51.0% for the offenders who had not completed the maintenance phase.

Lightfoot and Boland found that the Choices participants made significant positive changes from pre-test to post-test on 4 of the 6 psychometric instruments they used to assess program performance (Alcohol knowledge, Attitudes toward Substance Use, Problem Solving Skills, and Relapse Prevention Knowledge).

Survival in the community

The results indicated that the matched comparison cases assessed in the none to low category were readmitted at a faster rate during the first eight months of follow-up time. For the maintenance completer group, the trend reversed, showing a greater rate of failure near the end of the follow-up period. Considering offenders assessed with moderate substance abuse problems, the data showed that those in the matched comparison group generally failed sooner than the Choices participants. The survival rates for offenders in the substantial to severe levels demonstrated that Choices participants remained in the community for a longer period of time compared to those in the matched comparison sample.

---

1.   340 Laurier Avenue West, Ottawa, Ontario K1A 0P9.

2.  Correctional Service of Canada (1991). Task Force on the Reduction of Substance Abuse: Final Report. Ottawa, ON: Correctional Service of Canada.

3.  Fabiano, E. A. (1993). Developing a Model for the Provision of Substance Abuse Treatment. Ottawa, ON: Correctional Service of Canada.

4.  Weekes, J. R., Moser, A. E., and Langevin, C. M. (1997). Assessing Subsance Abusing Offenders for Treatment. Ottawa, ON: Correctional Service of Canada.

5.  Miller, W. R. (1992). The effectiveness of treatment for substance abuse: Reasons for optimism. Journal of Substance Abuse Treatment, 9, 93-102.

6.  Lightfoot, L. O., and Barker, J. (1989). A field test of the revised Substance Abuse Pre-Release Program: Joyceville Institution. Ottawa, ON: Correctional Service Canada.

7.  Barker, J. (1990). An Evaluation of the Substance Abuse Pre-Release Program at Joyceville Institution. Unpublished master’s thesis.

8.  Millson, W. A., Weekes, J. R., and Lightfoot, L. O. (1995). The Offender Substance Abuse Pre-Release Program: Analysis of Intermediate and Post-Release Outcomes. Research Report R-40. Ottawa, ON: Correctional Service of Canada.

9.  T3 Associates (1999). An outcome Evaluation of CSC Substance Abuse Programs; OSAPP, ALTO, and Choices, Final Report. Ottawa, ON: Correctional Service of Canada.

10.  T3 Associates (1999).

11.  Lightfoot, L. O., and Boland, F. J. (1994). Developing an Evaluation Framework for Choices. Kingston, ON: Unpublished document.