Assessing the Impact of Methadone Maintenance Treatment (MMT) on Post-release Recidivism among Male Federal Correctional Inmates in Canada

Abstract

Methadone maintenance treatment (MMT) can benefit post-release outcomes for correctional populations with opioid dependence, yet few outcome data exist for Canada. This retrospective study examined return to custody (RTC) outcomes following correctional release among three samples of male federal offenders with problematic opioid use (n = 856): offenders who continued MMT post-release (MMT-C) and offenders who discontinued MMT post-release (MMT-T) after being initiated on correctional MMT, and a non-MMT treated control group (MMT-N). MMT status was determined by community-based urinalysis; administrative data were used for outcomes. While the rate of continued MMT was lower, the MMT-C group had a 36% lower risk of RTC than the MMT-N group; RTC risk was not significantly different between the MMT-T and the MMT-N groups. Continuous MMT for correctional offenders with opioid dependence appears to be related to decreased post-release recidivism, and thus to social reintegration outcomes. Better understanding and improvement of corrections-to-community transition dynamics concerning MMT are required.

Keywords

opioid dependence methadone maintenance treatment criminal recidivism correctional populations,Canada

It is well-documented that substance-use disorders are disproportionately prevalent in correctional populations. The recent large-scale reviews of the studies of correctional samples have found that up to 48% of male inmates and up to 60% of female inmates have at least one substance-use disorder (Fazel, Bains, & Doll, 2006). Data from Canada confirm these high rates of substance-use problems. A survey of federal correctional inmates’ substance abuse in the 3 months prior to incarceration found that 77% reported cannabis, 68% cocaine, and 15% heroin use (Plourde & Brochu, 2002). A 2007 federal inmate survey found that 57% of the male and 60% of the female federal inmates used illicit drugs just prior to incarceration, with about half continuing such use when imprisoned (Zakaria, Thompson, Jarvis, & Borgatta, 2010). Heroin users, in comparison with users of stimulant drugs such as cocaine and amphetamines, have been found to be more likely to continue their drug use while in prison; importantly, ongoing substance abuse while incarcerated often occurs in conjunction with use patterns (e.g., injecting, needle sharing) posing great risks for health and safety (Calzavara et al., 2003; Strang et al., 2006).

A substantive proportion of correctional inmates have a history of illicit opioid use. In the federal inmate survey, 11% of male and 16% of female inmates reported a pre-incarceration history of opioid injection with 8% and 6% reporting continued opioid injection use while incarcerated, respectively (Zakaria et al., 2010). In Canada, illicit opioid use is common among high-risk drug user (e.g., street user) populations, although there has been a substantive recent shift from predominant heroin to prescription opioid use in many locales (Fischer, Rehm, Patra, & Firestone Cruz, 2006; Public Health Agency of Canada, 2006). A large literature of community samples documents that illicit opioid use is associated with a number of key risks and harms: namely, injection drug use and infections with blood borne viruses (BBV, e.g., HIV and HCV), physical and mental health problems, and criminal activity (e.g., property or drug acquisition crime; Amari, Rehm, Goldner, & Fischer, 2011; Bruneau, Roy, Arruda, Zang, & Jutras-Aswad, 2012; Connock et al., 2007; Fischer et al., 2005; Marsch, 1998).

Introduced some 40 years ago and acknowledging the nature of opioid dependence as a chronic brain disease, opioid maintenance treatment (OMT)—that is, either MMT or, increasingly, treatment using buprenorphine—has evolved as the primary treatment response to opioid dependence in North America, with both substances included in the WHO’s list of essential medicines (Hyman & Malenka, 2001; McLellan, Lewis, O’Brien, & Kleber, 2000; Nicholls, Bragaw, & Ruetsch, 2010; World Health Organization, 2009). A large body of evaluation literature documents for those illicit opioid users effectively retained and recruited into OMT: reductions in illicit opioid and other drug use, reductions in BBV risks, improvement of health status and lowered mortality risk, and reductions in criminal activity (Connock et al., 2007; Marsch, 1998; Mattick, Breen, Kimber, & Davoli, 2009; Ward, Hall, & Mattick, 1999). In Canada—especially since OMT guidelines were liberalized in the mid-1990s and treatment became increasingly delivered in community-based settings—the number of individuals enrolled in OMT has majorly increased from about 10,000 in the year 2000 to about 50,000 to 60,000 in 2011 (Fischer, 2000; Fischer & Argento, 2012; Strike, Urbanoski, Fischer, Marsh, & Millson, 2005).

While OMT was originally promoted as primarily a crime prevention measure (Courtwright, 1982; Dole et al., 1969; Massing, 1998), and despite high levels of opioid-use problems in correctional populations, OMT has only been gradually and selectively implemented in correctional settings (Dolan & Wodak, 1996; Hedrich et al., 2012; Luyt, 2007; Stallwitz & Stover, 2007; Stover, Casselman, & Hennebel, 2006). A small body of controlled, English-language studies (primarily from the United States and Australia) on OMT outcomes in correctional populations illustrates OMT’s overall positive effects in correctional populations, mainly by reduced illicit heroin use and reduced criminal activity or recidivism (Dole et al., 1969; Dolan et al., 2003; Dolan et al., 2005; Dolan, Wodak, & Hall, 1998; Gordon, Kinlock, Schwartz, & O’Grady, 2008; Johnson, van de Ven, & Grant, 2001; Kinlock et al., 2007; Kinlock, Gordon, Schwartz, Fitzgerald, & O’Grady, 2009; Kinlock, Gordon, Schwartz, & O’Grady, 2008; Larney, Toson, Burns, & Dolan, 2012; Magura, Rosenblum, Lewis, & Joseph, 1993). In what is commonly an intensively criminogenic population, these effects are of evident importance for correctional populations in the contexts of community release and reintegration.

Specifically, Dole and colleagues (1969) showed that a random sample of inmates on MMT had lesser rates of re-incarceration than untreated controls within the first year of release, and while all untreated controls resumed regular heroin use shortly after release, no inmates who received MMT relapsed to daily heroin use. Magura and colleagues (Magura et al., 1993) found that a sample of correctional MMT participants was more likely than a control group undergoing heroin detoxification to be in a drug-abuse treatment program at a six and a half month post-release follow up, which was associated with lower rates of opioid and cocaine use and property crime, as well as less illegal income post-release. A series of studies at various time points following release from prison demonstrated superior post-release outcomes in terms of reductions in illicit drug use as well as in criminal behavior for an intervention group receiving continuous MMT in corrections and post-release, compared with controls who received community-based MMT or counseling only (Gordon et al., 2008; Kinlock et al., 2007; Kinlock et al., 2008; Kinlock et al., 2009). Dolan and colleagues (1998) compared methadone maintained patients with participants who received standard drug treatment, or time-limited methadone, and found a lower prevalence of injection heroin use among offenders maintained on methadone. In a subsequent research based on a randomized waitlist control design, Dolan et al. (2003, 2005) found significantly lower heroin use and re-incarceration rates at follow-up among offenders in MMT compared with the waitlist group; the strength of these effects were strongly correlated with the duration of MMT.

However, some studies have not found clear effects of MMT on key post-release outcomes in correctional samples. For example, Kinlock et al. (2009) and Gordon et al. (2008) did not find significant differences in re-arrest outcomes between the study groups. Other studies have also not found associations between correctional MMT participation and recidivism, although these studies feature important methodological limitations—mainly, the inability to determine the continuity of MMT following release from prison, which has been found to be associated with a significantly decreased risk of re-incarceration (Larney et al., 2012; Marzo et al., 2009; McMillan, Lapham, & Lackey, 2008). In general, it is well-evidenced that the transition of corrections- to community-based health care following offenders’ release is a major challenge; this is particularly relevant for MMT, as due to the nature of opioid dependence—including high vulnerability to stress and drug-use-related cues in the context of release—the continuity of MMT can easily be interrupted, and relapse can occur (Hammett, Gaiter, & Crawford, 1998; Hammett, Roberts, & Kennedy, 2001; Hyman & Malenka, 2001; McLellan et al., 2000; Osher, Steadman, & Barr, 2003).

The Correctional Service of Canada (CSC), that is, the Canadian federal correctional system for offenders receiving a carceral sentence of 2 years or more, implemented its MMT program in 1998. The CSC’s MMT program was initially available only for offenders to continue community-based MMT on correctional admission, yet was expanded to also initiate offenders on MMT who met the criteria for opioid addiction in 2002 (Johnson et al., 2001; Sibbald, 2002). A 2007 survey of federal inmates found that 7% (approximately 962 men and women offenders) of the CSC inmate population were enrolled in MMT, with an even higher rate (9%) of incarcerated offenders (approximately 1,237 offenders) interested in receiving such treatment (Zakaria et al., 2010). A couple of studies have examined correctional MMT outcomes in Canada. Johnson et al. (2001) compared a sample of initial enrollees in CSC’s MMT program with an untreated group of illicit opioid users and found a trend toward lower reoffending and readmission rates in the MMT group, but similarities in terms of time to and number of new offenses. Other studies were small and without controls (Motiuk, Dowden, & Nafekh, 1999; Stones, 1999).

In the context of CSC’s systematic capacity for MMT initiation, as well as the large number of inmates who have been admitted to MMT in CSC, this study seized the opportunity to evaluate the impact of CSC’s MMT program on inmates’ recidivism. Specifically, the main objective of this study was to assess the impact—and identify key covariates—of MMT on post-release criminal reoffending and RTC in a sample of male inmates admitted to CSC’s MMT between 2006 and 2008.

Method

Sample

The retrospective cohort consisted of male federal offenders who were initiated on MMT while incarcerated in a federal correctional facility in Canada between January 1, 2006, and December 31, 2008 (methadone was the only OMT medication approved by CSC during the study period, so the exclusive focus of this study is on MMT). These MMT participants (N = 642) were further classified into two groups, based on the results of community-based urinalysis testing following their first release from prison during the study period, undertaken as part of monitoring while they were under CSC supervision in the community: offenders continuing MMT involvement in the community (MMT-C group, N = 161), as evidenced by at least one confirmed positive legitimate methadone urinalysis test, and offenders who terminated MMT involvement on release (MMT-T group, N = 481), as indicated by absence of methadone-positive community urinalysis test results during the follow-up period, or methadone-positive results for illegitimate (i.e., nonprescribed) use. Once released into the community, offenders are not required to provide information about ongoing MMT to CSC; therefore, the proxy measure based on community-based urinalysis testing—which is a condition of community supervision after an offender’s release—relied on the best available data to determine post-release MMT status.

In addition to the above samples, a comparison group for this study was created comprising a sample of offenders admitted to federal prison during the same time period, who were assessed at correctional intake as having a moderate to severe substance-abuse problem, indicated that opioids were their most commonly used drug prior to arrest, did not participate in MMT while incarcerated, and were released from prison during the follow-up period (MMT-N group, N = 214). Offenders identified for the MMT-N group who had evidence of community-based MMT participation following release (n = 9), based on urinalysis results, were excluded from the analysis. Comparisons on the covariates examined for RTC showed no statistically significant differences between the MMT-N group and the excluded cases.

Data Sources

Sociodemographic, sentencing, criminogenic risk and need, and readmission data on the study sample were extracted from the CSC’s Offender Management System (OMS), an electronic administrative database used to manage offender records throughout their sentence. Two additional data sources were used for the identification and classification of study participants: CSC’s MMT research database (comprising administrative health records identifying CSC’s MMT participants), and the Computerized Assessment of Substance Abuse (CASA; an automated assessment of substance-abuse problems, which was used to identify the MMT-N group; Kunic & Grant, 2006). This study relied exclusively on administrative data routinely collected by CSC from inmates based on signed consent and de-identified for analysis purposes, as governed by CSC protocols to protect inmates’ safety, privacy, and data confidentiality.

Measures

Sociodemographic and sentence information, including age at correctional admission (continuous, in years), Aboriginal ancestry (comprising offenders with Inuit, Metis, and First Nations background vs. all other ethnic groups), number of federal sentences served (categorized as 1, 2, or 3 or more), current sentence length (2-4 years, 4-6 years, or more than 6 years), criminogenic risk and need (categorized as low/moderate vs. high), release type (discretionary vs. statutory), and RTC occurrence and reason leading to RTC (see below) were extracted from the OMS. Criminogenic risk and need are global measures evaluated during the offender intake assessment, with the assessment of risk considering historical factors including criminal history, offense severity, and sex-offense history. Correspondingly, the assessment of need considers information from seven domains: employment/education, marital/family, associates, substance abuse, community functioning, personal/emotional orientation, and attitude. The type of release an offender receives is the result of a risk assessment for “public safety” conducted by the Parole Board of Canada. Specifically, offenders who are perceived as presenting low risk receive discretionary release (i.e., day or full parole prior to serving two thirds of their sentence) and offenders with higher risk receive statutory release (i.e., release mandated by law typically after an offender has served two thirds of a determinate sentence).

The main outcome of interest was RTC in the three samples following release from prison, while under community supervision (i.e., parole or statutory release), as well as possible covariates of readmission. Offenders may be returned to custody for three reasons: the commission of a new offense; a technical revocation, which is a violation of a condition of their release (e.g., breaking an abstinence condition, failing to report to their parole officer, etc.); or conviction for an outstanding charge (i.e., an offense that occurred previous to their most recent period of incarceration). Duration of follow-up was defined as the number of days between an offender’s release and their RTC or the end of the follow-up period, whichever came first.

Statistical Analyses

Time to RTC was examined using the Cox Proportional Hazards Model (Dohoo, Martin, & Stryhn, 2009). The event of interest was RTC due to a violation of a release condition, or due to committing a new offense. The RTC analysis was limited to individual offenders’ first RTC, and hence is based on a sample of unique cases (i.e., does not include multiple RTCs by the same individual). Offenders were observed for a maximum of 3 years following release from prison. Offenders who reached their warrant expiry date (i.e., the date CSC supervision ends), or the end of the follow-up period and who were not re-incarcerated were censored. In addition, offenders with an RTC for an outstanding charge were censored, as their return to custody was not related to their behavior following release.

The covariates of criminogenic risk, need, Aboriginal ancestry, number of sentences, age at correctional admission, and release type, along with post-release MMT status, were tested for independent association with the outcome variable of RTC. Covariates significant at the p < .25 level were entered into the model. Forward, backward, and stepwise model selection were used, all of which resulted in the same final model. Covariates were dropped from or retained in the model at the p < .25 level. Adjusted hazard ratios (AHR; adjusted for other covariates in the model), 95% confidence intervals (CI), and significance levels were reported for all covariates retained in the final model.

Results

Sample Characteristics

Sociodemographic, sentence, and criminogenic risk and need characteristics are presented by the study group in Table 1. Offenders in the three groups did not significantly differ in terms of age at correctional admission or Aboriginal ancestry. Offenders in the MMT-N group had fewer federal sentences, shorter current sentences, and were assessed as having lower criminogenic risk and need ratings compared with offenders in the MMT-C and MMT-T groups. In addition, offenders in the MMT-T group were significantly more likely to receive statutory rather than discretionary release compared with the MMT-C and MMT-N groups; offenders in the MMT-C group were also significantly more likely to receive statutory rather than discretionary release than offenders in the MMT-N group (see Table 1).

Table 1:

Sociodemographic and Criminal Justice Characteristics of the Sample, by Study Group

Characteristic	MMT-T (N = 481)	MMT-C (N = 161)	MMT-N (N = 214)	Significance
Age at admission M (SD)	34.3 (8.1)	35.3 (8.6)	34.6 (8.3)	F(2, 856) = 0.90, p = .41
Aboriginal ancestry % (n)	15.0 (72)	16.8 (27)	15.4 (33)	χ²(2, n = 856) = 0.30, p = .86
Sentence number % (n)				χ²(4, n = 856) = 14.45, p = .006, v = 0.09^{+ ~}
1	46.6 (224)	37.9 (61)	57.0 (122)
2	26.0 (125)	31.7 (51)	22.9 (49)
3 or higher	27.4 (132)	30.4 (49)	20.1 (43)
Sentence length % (n)				χ²(4, n = 855) = 38.56, p < .0001, v = 0.15^{+ ~}
2-4 years	64.2 (308)	63.4 (102)	80.4 (172)
4-6 years	18.8 (90)	15.5 (25)	17.8 (38)
Over 6 years	17.1 (82)	21.1 (34)	1.9 (4)
Risk				χ² (2, n = 841) = 8.47, p = .01, v = 0.10^{+ ~}
Low/moderate	45.5 (216)	41.4 (65)	55.5 (116)
High	54.5 (259)	58.6 (92)	44.5 (93)
Need				χ²(2, n = 841) = 8.07, p = .02, v = 0.10^{+ ~}
Low/moderate	25.1 (119)	19.8 (31)	32.5 (68)
High	75.0 (356)	80.3 (126)	67.5 (141)
Release type				χ²(2, n = 856) = 34.78, p < .0001, v = 0.20
Discretionary	20.0 (96)	31.1 (50)	41.1 (88)
Statutory	80.0 (385)	68.9 (111)	58.9 (126)

Note. +Significant (p < .05) differences between MMT-T and MMT-N groups. ^{^}Significant (p < .05) differences between MMT-T and MMT-C groups. ^~Significant (p < .05) differences between MMT-C and MMT-N groups. ⁺⁺Significant (p < .05) differences between all three groups.

Post-Release Outcomes

While 52.4% of the MMT-T reported a RTC, this rate was 40.4% for the MMT-C group and 47.7% for the MMT-N group (see Table 2); on this basis, the MMT-C group was significantly less likely to indicate a RTC than the MMT-T group; however, there were no differences between the MMT-N group and the other two study groups. In addition, the three study groups did not significantly differ on the type of RTC among those where this event occurred. The duration of time spent in the community under supervision, however, was significantly different across the three groups. Offenders in the MMT-C group remained under community supervision for longer periods of time than those in the MMT-T or the MMT-N groups, while offenders in the MMT-T group remained under community supervision for the shortest period of time.

Table 2:

Proportions and Details of Return to Custody, by Study Group

Variable	MMT-T (N = 481)	MMT-C (N = 161)	MMT-N (N = 214)	Significance
Return to custody				χ²(2, n = 856) = 7.16, p = .0279, v = 0.09
No	47.6	59.6	52.3
Yes	52.4	40.4	47.7
Type of return to custody (subset)				χ²(2, n = 419) = 1.64, p = .4406, v = 0.06
New offence	16.3 (41)	15.4 (10)	21.6 (22)
Technical revocation	83.7 (211)	84.6 (55)	78.4 (80)
Duration of follow-up (days)	274.2 (190.8)	358.8 (231.7)	325.3 (205.4)	F(2, 856) = 12.26, p < .0001
M (SD)	274.2 (190.8)	358.8 (231.7)	325.3 (205.4)	F(2, 856) = 12.26, p < .0001

Results of the Cox Proportional Hazards Model for the risk of RTC and possible covariates are presented in Table 3. All covariates tested were retained in the final model, with the exception of criminogenic risk. The MMT-C group was found to have a 36% lower risk of RTC than the MMT-N group (AHR = 0.64, CI = 0.47, 0.88); however, the risk of RTC was not significantly different between the MMT-T and the MMT-N groups (AHR = 1.15, CI = 0.91, 1.46; see Figure 1). As presented in Table 3, offenders with a high level of criminogenic need were found to have a 46% greater risk of having RTC compared with those with low or moderate need levels. The number of previous sentences was also found to be a significant predictor of RTC, with offenders serving their second federal sentence having a 31% greater risk and offenders serving their third or subsequent federal sentence having a 56% greater risk of RTC than those serving their first federal sentence. Offenders who received statutory release had a 44% greater risk of having RTC compared with those who received discretionary release. In addition, with each 1-year increase in age, the risk of RTC decreased by 2%. Aboriginal offenders were found to have a 20% greater risk of RTC than non-Aboriginal offenders; however, this difference was not statistically significant.

Table 3:

Adjusted Cox Proportional Hazards Model of Risk of Return to Custody

Covariate	Wald χ²	p value	Estimate (β)	SE	AHR	95% CI
Treatment status	17.07	.0002
MMT-N			Ref	Ref	Ref	Ref
MMT-T			0.14	0.12	1.15	[0.91, 1.46]
MMT-C			−0.45**	0.16	0.64	[0.47, 0.88]
Need	9.75	.0018
Low/moderate			Ref	Ref	Ref	Ref
High			0.38**	0.12	1.46	[1.15, 1.85]
Sentence number	11.23	.0036
First			Ref	Ref	Ref	Ref
Second			0.27*	0.13	1.31	[1.02, 1.67]
Third or higher			0.44**	0.14	1.56	[1.20, 2.03]
Aboriginal ancestry	1.81	.1787
Non-Aboriginal			Ref	Ref	Ref	Ref
Aboriginal			0.18	0.13	1.20	[0.92, 1.55]
Age at admission	7.78	.0053	0.02**	0.01	0.98	[0.97, 0.99]
Release type	8.92	.0028
Discretionary			Ref	Ref	Ref	Ref
Statutory			0.37**	0.12	1.44	[1.14, 1.84]

Note. Parameter estimates are adjusted for other covariates in the model. AHR = Adjusted hazard ratio; CI = confidence interval.

p < .05. **p < .01.

Figure 1:

Adjusted Survival Curve on Return to Custody, by Study Group

Discussion

This study examined RTC outcomes and covariates following correctional release among three distinct samples of federal male offenders within CSC with problematic opioid use, two of which had been initiated on MMT during incarceration, and a non-MMT control group.

First, our results showed that about half of the total combined study sample was returned to correctional custody, following an average of less than one year of post-release community supervision. This is a high RTC rate in comparison with the general CSC population (where approximately one third of offenders were returned to custody prior to reaching their warrant expiry date during the same time period; Public Safety Canada, 2011), which, in different possible ways, is likely influenced by this particular population’s commonly complex and severe substance-use issues. While opioid dependence is considered a chronic disease (Kritz et al., 2009)—that is, commonly brings the risk of relapse even for individuals in treatment such as MMT—many opioid misusers commonly have multiple other substance-use problems (e.g., alcohol or cocaine) that are associated with criminal activity (Fischer, Cruz, & Rehm, 2006; Johnson, MacDonald, Cheverie, Myrick, & Fischer, 2012; Larney et al., 2012; Leri, Bruneau, & Stewart, 2003). These ongoing dynamics may easily lead to criminal behaviors or violations of release conditions resulting in an RTC. These risks further corroborate the fact that the offender population under study is overall considered a group with high criminogenic risks and needs, including a high proclivity for RTC (Bellin, Tomasino, Nolan, Glick, & Oquendo, 1999). These dynamics are also expressed in the fact that those offenders who rank “high” on the criminogenic needs assessment, as well as those with previous sentences—that is, those with more elaborate criminal histories that may or may not be drug-use-related—were found to be at a substantially elevated risk of RTC independent of MMT status.

A finding of key importance is that the MMT-C group—that is, those offenders successfully initiated on MMT while incarcerated, and transitioned and involved in community-based MMT—featured a significantly lower likelihood of RTC than the two other groups under study (MMT-T and MMT-N). This overall confirms findings from numerous general MMT evaluation studies concluding that individuals with opioid dependence actively treated in MMT feature superior drug use, health, social, and crime outcomes compared with those not in treatment (Amato et al., 2005; Marsch, 1998; Mattick et al., 2009; Pearson & Lipton, 1999). More specifically, these findings also confirm outcomes found by several studies on correctional populations initiated on MMT—that longer periods of retention in MMT, including the continuation of treatment in the community following release, are associated with more positive treatment outcomes (Dolan et al., 2005; Larney et al., 2012; Magura et al., 1993). As such, the effective initiation of and transition to community-based MMT following release can be considered an intervention measure contributing to more effective crime prevention and rehabilitation in federal correctional offenders with opioid dependence. Given the high degree of active illicit opioid use or histories, there may be need and room to further expand MMT provision in CSC with the likely benefit of further reduction in reoffending and RTC among eligible offenders (Johnson et al., 2012; Kunic & Grant, 2006; Zakaria et al., 2010).

The results shown for the comparison groups—specifically, the MMT-T group—should not be interpreted as MMT initiation in corrections with those who do not successfully transition to or remain in community-based MMT post-release to be a failure. First, even though there were no significant differences in RTC rates between the MMT-T and the MMT-N groups, the MMT-T group may have experienced substantive benefits—including improvements in institutional behavior and/or health (e.g., due to decreased opioid and/or injection drug use)—from the MMT received while in corrections (Dolan et al., 2003; Dolan et al., 2005; Hedrich et al., 2012; Johnson et al., 2001; Stallwitz & Stover, 2007). In addition, it has been demonstrated that individuals who participate in prison-based MMT are retained in treatment longer than those who commence treatment in the community post-release, which in turn, as mentioned, is associated with more positive treatment outcomes, including reductions in criminal activity (Dolan et al., 2005; Kinlock et al., 2009; Larney et al., 2012; Magura et al., 1993). Prison-based MMT may also help to facilitate community-based MMT participation. Approximately 25% of the offenders in our sample who had been initiated on CSC’s MMT program continued treatment in the community, whereas only 4% of the offenders with an identified opioid problem who did not receive MMT while in corrections participated in community-based MMT (internal calculations).

As mentioned, MMT occurs with relapse episodes, or involves multiple-treatment exit and reentry episodes, based on the nature of opioid dependence as a chronic disease in many instances, and a key challenge will be to reconnect those who terminated MMT with appropriate treatment services in cases of persistent need (Connock et al., 2007; Merrill, Alterman, Cacciola, & Rutherford, 1999). A crucial task for further research and interventions here is to better understand the reasons for post-release termination or non-access of MMT, specifically as the majority (75%) of the study sample originally initiated on correctional MMT did not successfully transition to community-based MMT (MMT-T), based on this study’s indicators. There could be numerous reasons for MMT-initiated offenders to not successfully transition into community-based MMT following release. In general, the corrections-to-community transition is the “Achilles’ heel” or difficult point of challenge for many key health and social services or interventions provided to and existentially needed by correctional offenders, as seamless or successful continuity often does not occur (Chandler, Fletcher, & Volkow, 2009; Marzo et al., 2009; Prendergast, 2009; Rich et al., 2001). This could be influenced by many extrinsic reasons—for example, nonavailability of services needed (also depending on location of release, e.g., urban or nonurban), information breakdown or system complexity, task overload by offender or probation officer, and so on—which should be investigated to allow for improved practice. In the specific case of MMT, while such treatment is fairly available in most Canadian urban environments, its availability is much more limited and sporadic in nonurban or rural environments (Deck & Carlson, 2004; Luce & Strike, 2011; Reist, 2010). The interruption or noncontinuity of MMT following an opioid-dependent offender’s correctional release by just 24 to 48 hr could critically precipitate conditions (e.g., opioid withdrawal) for acute relapse and/or discontinuation of treatment (World Health Organization, 2009). In many instances, successful transitioning to continued MMT treatment following correctional release may require intensive individual case management to avoid such risks.

This study has some important limitations. The sample was restricted to offenders for which community urinalysis data was available, as information concerning continuing MMT involvement in the community was not obtainable using any other method (i.e., offenders do not have to share MMT program participation information with CSC once released). Previous studies that did not find an association between prison-based MMT participation and recidivism cite the inability to examine continuity of MMT services in the community as a key limitation (Marzo et al., 2009; McMillan et al., 2008). Determined by the parameters of the data available, the study was based on a convenience sample, and results, therefore, cannot be generalized to other correctional or MMT populations. Given the use of urinalysis data as a proxy for MMT group status, the continuity or exact duration of post-release MMT was not known for study participants as this information was not available. As treatment duration is an important predictor of successful outcomes, this information would have been valuable for inclusion in the analyses (Dolan et al., 2005; Lind, Chen, Weatherburn, & Mattick, 2005; Oliver et al., 2010; Zhang, Friedmann, & Gerstein, 2003).

In sum, our study has found lowered risk for RTC among federal correctional offenders initiated and retained in community-based MMT following release compared with controls, underscoring the possible positive contribution of a successful transition from correctional to community-based MMT as a community integration and crime prevention measure for opioid-dependent individuals in Canadian corrections. In conclusion, OMT—MMT as well as available alternatives as appropriate—should be offered to all correctional inmates in ascertained clinical need of such treatment, and barriers to successful transition to community-based MMT should be investigated and addressed to further maximize benefits of these treatment interventions to the offender and society.

Footnotes

Acknowledgements

The authors thank Sara Johnson, Mindy White, Kelsey Robertson, and Chantal Burnett for their contributions to this research project. Benedikt Fischer acknowledges funding support from a CIHR/PHAC Research Chair in Applied Public Health.

Mary-Ann MacSwain holds a BA in psychology from the University of Prince Edward Island, Canada. Until recently, she worked as a research officer at the Correctional Service Canada’s Addictions Research Centre. She is currently a data analyst for Prince Edward Island’s Chief Public Health Office and works primarily in the area of chronic disease surveillance.

Shanna Farrell-MacDonald holds a BA in sociology from McGill University, Quebec, Canada. She is a senior research officer with Correctional Service Canada. Her main research interests include substance-abuse assessment and treatment issues among correctional inmates, and database creation and management.

Madelon Cheverie holds an MSc from the University of Waterloo, Ontario, Canada and a BSc (honors) from the University of Prince Edward Island, Canada. She currently works as a research officer with the Correctional Service of Canada.

Benedikt Fischer, PhD, is professor and CIHR/PHAC Chair in applied public health in the Faculty of Health Sciences and the School of Criminology, and director, Centre for Applied Research in Mental Health and Addictions (CARMHA) at Simon Fraser University, Vancouver; he is also a senior scientist in social and epidemiological research, Centre for Addiction and Mental Health (CAMH) and professor (status) in the Department of Psychiatry, University of Toronto, Canada. His research focuses on illicit and nonmedical psychoactive substance use, key co-morbidities, and interventions, primarily in high-risk or marginalized (including correctional) populations.

References

Amari

Rehm

Goldner

Fischer

(2011). Nonmedical prescription opioid use and mental health and pain comorbidities: A narrative review. Canadian Journal of Psychiatry, 56, 495–502.

Amato

Davoli

Perucci

C. A.

Ferri

Faggiano

Mattick

R. P.

(2005). An overview of systematic reviews of the effectiveness of opiate maintenance therapies: Available evidence to inform clinical practice and research. Journal of Substance Abuse Treatment, 28, 321–329.

Bellin

Tomasino

Nolan

Glick

Oquendo

(1999). High dose methadone reduces criminal recidivism in opiate addicts. Addiction Research, 7, 19–29.

Bruneau

Roy

Arruda

Zang

Jutras-Aswad

(2012). The rising prevalence of prescription opioid injection and its association with hepatitis C incidence among street-drug users. Addiction, 107, 1318–1327.

Calzavara

Burchell

Schlossberg

Myers

Escobar

Wallace

Millson

(2003). Prior opiate injection and incarceration history predict injection drug use among inmates. Addiction, 98, 1257–1265.

Chandler

R. K.

Fletcher

B. W.

Volkow

N. D.

(2009). Treating drug abuse and addiction in the criminal justice system: Improving public health and safety. Journal of the American Medical Association, 301, 183–190.

Connock

Juarez-Garcia

Jowett

Frew

Liu

Taylor

R. J.

Taylor

R. S.

(2007). Methadone and buprenorphine for the management of opioid dependence: A systematic review and economic evaluation. Health Technology Assessment, 11, 1–171.

Courtwright

. (1982). Dark paradise: Opiate addiction in America before 1940. Cambridge, MA: Harvard University Press.

Deck

Carlson

M. J.

(2004). Access to publicly funded methadone maintenance treatment in two western states. Journal of Behavioral Health Services & Research, 31, 164–177.

10.

Dohoo

Martin

Stryhn

(2009). Veterinary epidemiologic research (2nd ed.). Charlottetown, PEI: VER.

11.

Dolan

Shearer

MacDonald

Mattick

R. P.

Hall

Wodak

A. D.

(2003). A randomised controlled trial of methadone maintenance treatment versus wait list control in an Australian prison system. Drug and Alcohol Dependence, 72, 59–65.

12.

Dolan

Shearer

White

Zhou

Kaldor

Wodak

A. D.

(2005). Four-year follow-up of imprisoned male heroin users and methadone treatment: Mortality, re-incarceration and hepatitis C infection. Addiction, 100, 820–828.

13.

Dolan

Wodak

(1996). An international review of methadone provision in prisons. Addiction Research, 4, 85-97.

14.

Dolan

Wodak

Hall

(1998). Methadone maintenance treatment reduces heroin injection in New South Wales prisons. Drug and Alcohol Review, 17, 153–158.

15.

Dole

V. P.

Robinson

J. W.

Orraca

Towns

Searcy

Caine

(1969). Methadone treatment of randomly selected criminal addicts. New England Journal of Medicine, 280, 1372–1375.

16.

Fazel

Bains

Doll

(2006). Substance abuse and dependence in prisoners: A systematic review. Addiction, 101, 181–191.

17.

Fischer

(2000). Prescriptions, power and politics: The turbulent history of methadone maintenance in Canada. Journal of Public Health Policy, 21, 187–210.

18.

Fischer

Argento

(2012). Prescription opioid related misuse, harms, diversion and interventions in Canada: A review. Pain Physician, 15, ES191-ES203.

19.

Fischer

Cruz

M. F.

Rehm

(2006). Illicit opioid use and its key characteristics: A select overview and evidence from a Canadian multisite cohort of illicit opioid users (OPICAN). Canadian Journal of Psychiatry, 51, 624–634.

20.

Fischer

Rehm

Brissette

Brochu

Bruneau

El-Guebaly

Baliunas

(2005). Illicit opioid use in Canada: Comparing social, health and drug use characteristics of untreated users in five cities (OPICAN study). Journal of Urban Health, 82, 250–266.

21.

Fischer

Rehm

Patra

Firestone Cruz

(2006). Changes in illicit opioid use profiles across Canada. Canadian Medical Association Journal, 175, 1385–1387.

22.

Gordon

M. S.

Kinlock

T. W.

Schwartz

R. P.

O’Grady

K. E.

(2008). A randomized clinical trial of methadone maintenance for prisoners: Findings at 6 months post-release. Addiction, 103, 1333–1342.

23.

Hammett

T. M.

Gaiter

J. L.

Crawford

(1998). Reaching seriously at-risk populations: Health interventions in criminal justice settings. Health Education & Behavior, 25, 99–120.

24.

Hammett

T. M.

Roberts

Kennedy

(2001). Health-related issues in prisoner reentry. Crime & Delinquency, 47, 390–409.

25.

Hedrich

Alves

Farrell

Stover

Moller

Mayet

(2012). The effectiveness of opioid maintenance treatment in prison settings: A systematic review. Addiction, 107, 501–517.

26.

Hyman

S. E.

Malenka

R. C.

(2001). Addiction and the brain: The neurobiology of compulsion and its persistence. Nature Reviews Neuroscience, 2, 695–703.

27.

Johnson

MacDonald

S. F.

Cheverie

Myrick

Fischer

(2012). Prevalence and trends of non-medical opioid and other drug use histories among federal correctional inmates in methadone maintenance treatment in Canada. Drug and Alcohol Dependence, 124, 172–176.

28.

Johnson

van de Ven

J. T. C.

Grant

B. A.

(2001). Institutional methadone maintenance treatment: Impact on release outcome and institutional behaviour (Report No. 119). Ottawa, Ontario: Correctional Service Canada.

29.

Kinlock

T. W.

Gordon

M. S.

Schwartz

R. P.

Fitzgerald

T. T.

O’Grady

K. E.

(2009). Randomized clinical trial of methadone maintenance for prisoners: Results at twelve months post-release. Journal of Substance Abuse Treatment, 37, 277–285.

30.

Kinlock

T. W.

Gordon

M. S.

Schwartz

R. P.

O’Grady

(2008). A study of methadone maintenance for male prisoners: 3 month postrelease outcomes. Criminal Justice and Behavior, 35, 34–47.

31.

Kinlock

T. W.

Gordon

M. S.

Schwartz

R. P.

O’Grady

Fitzgerald

T. T.

Wilson

(2007). A randomized control trial of methadone maintenance for prisoners: Results at one-month post-release. Drug and Alcohol Dependence, 91, 220–227.

32.

Kritz

Chu

John-Hull

Madray

Louie

Brown

L. S.

(2009). Opioid dependence as a chronic disease: The interrelationships between length of stay, methadone dose, and age on treatment outcome at an urban opioid-treatment program. Journal of Addictive Diseases, 28, 53–56.

33.

Kunic

Grant

B. A.

(2006). The computerized assessment of substance abuse (CASA): Results from the demonstration project (Report No. 173). Ottawa, Ontario: Correctional Service Canada.

34.

Larney

Toson

Burns

Dolan

(2012). Effect of prison-based opioid-substitution treatment and post-release retention in treatment on risk of re-incarceration. Addiction, 107, 372–380.

35.

Leri

Bruneau

Stewart

(2003). Understanding polydrug use: Review of heroin and cocaine use. Addiction, 98, 7–22.

36.

Lind

Chen

Weatherburn

Mattick

(2005). The effectiveness of methadone maintenance treatment in controlling crime: An Australian aggregate-level analysis. British Journal of Criminology, 45, 201–211.

37.

Luce

Strike

(2011). A cross-Canada scan of methadone maintenance treatment policy developments. Toronto, Ontario, Canada: Canadian Executive Council on Addictions.

38.

Luyt

W. F. M.

(2007). Harm reduction, substance abuse and methadone maintenance in Scottish prisons. European Journal of Crime, Criminal Law, and Criminal Justice, 15, 205–226.

39.

Magura

Rosenblum

Lewis

Joseph

(1993). The effectiveness of in-jail methadone maintenance. Journal of Drug Issues, 23, 75–99.

40.

Marsch

L. A.

(1998). The efficacy of methadone maintenance interventions in reducing illicit opiate use, HIV risk behavior and criminality: A meta-analysis. Addiction, 93, 515–532.

41.

Marzo

J. N.

Rotily

Meroueh

Varastet

Hunault

Obradovic

Zin

(2009). Maintenance therapy and 3-year outcome of opioid-dependent prisoners: A prospective study in France (2003-06). Addiction, 104, 1233–1240.

42.

Massing

(1998). The fix. Berkeley: University of California Press.

43.

Mattick

R. P.

Breen

Kimber

Davoli

(2009). Methadone maintenance therapy versus no opioid replacement therapy for opioid dependence. Cochrane Database of Systematic Reviews, 3, 1–31.

44.

McLellan

A. T.

Lewis

D. C.

O’Brien

C. P.

Kleber

H. D.

(2000). Drug dependence, a chronic medical illness: Implications for treatment, insurance, and outcomes evaluation. Journal of the American Medical Association, 284, 1689–1695.

45.

McMillan

G. P.

Lapham

Lackey

(2008). The effect of a jail methadone maintenance therapy (MMT) program on inmate recidivism. Addiction, 103, 2017–2023.

46.

Merrill

Alterman

Cacciola

Rutherford

(1999). Prior treatment history and its impact on criminal recidivism. Journal of Substance Abuse Treatment, 17, 313–319.

47.

Motiuk

Dowden

Nafekh

(1999). Methadone maintenance treatment (MMT) programming for federal prisoners: A preliminary investigation. Ottawa, Ontario: Correctional Service of Canada.

48.

Nicholls

Bragaw

Ruetsch

(2010). Opioid-dependence treatment and guidelines. Journal of Managed Care Pharmacy, 16(1 Suppl B), S14–S21.

49.

Oliver

Keen

Rowse

Ewins

Griffiths

Mathers

(2010). The effect of time spent in treatment and dropout status on rates of convictions, cautions, and imprisonment over five years in a primary care-led methadone maintenance service. Addiction, 105, 732–739.

50.

Osher

Steadman

H. J.

Barr

(2003). A best practice approach to community reentry from jails for inmates with co-occurring disorders: The apic model. Crime & Delinquency, 49, 79–96.

51.

Pearson

F. S.

Lipton

D. S.

(1999). A meta-analytic review of the effectiveness of correction-based treatments for drug abuse. The Prison Journal, 79, 384–410.

52.

Plourde

Brochu

(2002). Drugs in prison: A break in the pathway. Substance Use & Misuse, 37, 47–63.

53.

Prendergast

M. L.

(2009). Interventions to promote successful reentry among drug-abusing parolees. Addiction Science & Clinical Practice, 5, 4–13.

54.

Public Health Agency of Canada. (2006). I-Track: Enhanced surveillance of risk behaviours among people who inject drugs - phase I report (Publication No. HP40–4/1-2006E-PDF). Ottawa, Ontario, Canada: Surveillance and Risk Assessment Division, Centre for Infectious Disease Prevention and Control, Public Health Agency of Canada.

55.

Public Safety Canada. (2011). Corrections and conditional release statistical overview-Annual report 2011 (Publication No. PS1-3/2011E-PDF). Ottawa, Ontario, Canada: Public Works and Government Services Canada.

56.

Reist

(2010). Methadone maintenance treatment in British Columbia, 1996–2008. Victoria, BC, Canada: Centre for Addictions Research of BC.

57.

Rich

J. D.

Holmes

Salas

Macalino

Davis

Ryczek

Flanigan

(2001). Successful linkage of medical care and community services for HIV-positive offenders being released from prison. Journal of Urban Health, 78, 279–289.

58.

Sibbald

(2002). Methadone maintenance expands inside federal prisons. Canadian Medical Association Journal, 167, 1154.

59.

Stallwitz

Stover

(2007). The impact of substitution treatment in prisons: A literature review. International Journal of Drug Policy, 18, 464–474.

60.

Stones

(1999). Dynamic and static predictors of methadone maintenance treatment compliance among forensic clients (Unpublished Master’s research paper). Department of Applied Psychology, University of Toronto, Ontario, Canada.

61.

Stover

Casselman

Hennebel

(2006). Substitution treatment in European prisons: A study of policies and practices in 18 European countries. International Journal of Prisoner Health, 2, 3–12.

62.

Strang

Gossop

Heuston

Green

Whiteley

Maden

(2006). Persistence of drug use during imprisonment: Relationship of drug type, recency of use, and severity of dependence to use of heroin, cocaine, and amphetamine in prison. Addiction, 101, 1125–1132.

63.

Strike

Urbanoski

Fischer

Marsh

Millson

(2005). Policy changes and the methadone maintenance treatment system for opioid dependence in Ontario, 1996 to 2001. Journal of Addictive Diseases, 24, 39–51.

64.

Ward

Hall

Mattick

(1999). Role of maintenance treatment in opioid dependence. Lancet, 353, 221–226.

65.

World Health Organization. (2009). Guidelines for the psychosocially assisted pharmacological treatment of opioid dependence. Geneva, Switzerland: Author.

66.

Zakaria

Thompson

Jarvis

Borgatta

(2010). Summary of emerging findings from the 2007 national inmate infectious disease and risk-behaviours survey (Report No. 211). Ottawa, Ontario, Canada: Correctional Service of Canada.

67.

Zhang

Friedmann

P. D.

Gerstein

D. R.

(2003). Does retention matter? Treatment duration and improvement in drug use. Addiction, 98, 673–684.