FORUM on Corrections Research

Success rates of female offenders on discretionary versus statutory release: Substance abusers and non-abusers

The dramatic increase in the size of the prison population during the past few decades has heightened interest in appropriate alternatives to incarceration. Moreover, current reintegration efforts stress the value of releasing low-risk offenders at the earliest opportunity while maintaining public safety through assistance and supervision in the community. Recent research supports this approach by demonstrating that day parole, a form of conditional release, is an effective method for releasing lower-risk offenders and can be predictive of post-release outcome². Accordingly, various forms of conditional release can help ease overcrowding in prisons. These release programs have been established to provide optimal circumstances for facilitating reintegration of lower-risk offenders into the community.

A number of studies have focused on factors that predict successful community reintegration accurately. In particular, recent research has demonstrated that an offender's history of substance abuse plays an important role in predicting day parole outcome.³ More specifically, offenders' with substance abuse problems had higher incidence of post-release failure. Research has also suggested that substance abuse is a predominant criminogenic need⁴ among women serving federal terms of incarceration.⁵

As mentioned, current correctional practice mandates the release of offenders at the earliest possible opportunity, while maintaining public safety and minimizing risk to reoffend. To meet this objective, offenders designated lower risk are generally released earlier on day or full parole,⁶ while higher-risk offenders are released later in their sentence and discharged on statutory release. We would expect to find that offenders identified with criminogenic needs such as substance abuse would be less likely than their lower need counterparts to be granted discretionary release.

The present study used a sample of released federal female offenders to compare substance abusers to non-abusers on a variety of criteria, including: demographic profile, offence history and designated risk level, release type, and post-release performance. To examine the association between program participation and success on parole, the substance abuser sample was subdivided according to whether they had completed substance abuse treatment while incarcerated. Subsequent analyses compared treatment completers to non-completers on post-release outcome criteria.

---

The sample for this study was extracted on October 1^st, 1997, from the Correctional Service of Canada Offender Management System, an automated database. All the study participants were federal female offenders who:

• had been admitted to federal custody and received a comprehensive intake assessment to identify risk and needs;
• had an official correctional plan based on the intake assessment; and
• had available Canadian Police Information Centre records (which documents entire official offence history).

The final sample consisted of 251 female offenders. Study participants were then classified as either substance abuser or non-abuser based on whether their correctional plan had recommended substance abuse treatment. This process rendered a group of 143 substance abusers and 108 non-abusers.

Release type

Many female offenders included in this sample had reached their parole eligibility or statutory release dates at the time of data collection. Accordingly, almost half the overall sample had been granted some form of discretionary release (n=105). Of those released, 37.1% were granted day parole, about half (46.7%) were granted full parole, and 16.2% were released on statutory release. As noted, offenders released on either day parole or full parole are generally deemed as lower-risk offenders, as it is believed (and supported by other research) that these individuals have good reintegration potential.

Table 1 displays the various forms of conditional release granted substance abusing and non-abusing federal female offenders. For substance abusers, 41.3% were granted day parole, 39.7% were granted full parole, and 19.0% were granted statutory release. A similar pattern emerged in the non-abusing release sample. These results were not statistically significant -- a somewhat surprising finding, given that previous research produced higher failure rates for substance abusers on day parole. More specifically, we expect to find that substance abusers would not be granted discretionary release as often as non-abusers, but this was clearly not the case here.

Table 1

Release Status: Substance Abusers and Non-Abusers
Release type	Substance abusers (n=63)	Non-abusers (n=42)
Day parole	26 (41.3%)	13 (31.0%)
Full Parole	25 (39.7%)	24 (57.1%)
Statutory release	12 (19.0%)	5 (11.9%)

Demographic information

In all subsequent between-group comparisons, the analyses covered 88 female offenders who had been granted day parole or full parole. The average age was about 35 years, and the parolees identified as substance abusers (mean=34.2) and the parolees who were non-abusers (mean=36.3) did not differ significantly in age.

Analyses comparing Aboriginal to non-Aboriginal female offenders rendered some interesting findings. The vast majority of Aboriginal female offenders (13 of 14) on parole were identified as substance abusers. About half (38 of 74; or 51.4%) of non-Aboriginal female parolees were substance abusers. These results were found to be highly statistically reliable (p<.01). This finding highlights the need for ongoing, culturally specific substance abuse programming for female offenders after release into the community.

Offence history

Female offenders identified as substance abusers were compared with non-abusers on the average number of past and current convictions. Again, comparisons included only female offenders who had been granted either day parole or full parole. Given the criminogenic dimension of substance abuse, it was expected that women with problems in this area would demonstrate more criminal involvement than their non-abuser counterparts.

Results indicated that substance abusers had previously committed a higher number of offences compared with non-abusers. Substance abusers granted discretionary release had committed, on average, 18.6 offences in their criminal history, compared with only 3.7 offences committed by their non-abusing counterparts.

Analyses were also conducted on the number of current offences committed by our sample of female parolees. Once again, substance abusers had committed a higher number of offences at admission (6.1 criminal acts, on average) compared with non-abusers (1.5, on average). These results indicate female offenders identified as substance abusers have had more criminal involvement than non-abusers.

Risk level

An overall risk designation is assigned to each offender at intake, based on a comprehensive analysis of relevant factors such as conviction record, breaches of trust and offence severity. For female offenders granted discretionary release, those identified as substance abusers were compared with non-abusers on overall risk level. Table 2 provides the breakdown of risk level by substance abuse status. Most non-abusers released on parole were categorized as lower risk (91.9%), but only 54.9% of substance abusers were categorized as lower risk.

Table 2

Risk Designation of Parolees: Substance Abusers and Non-Abusers
	Substance abusers (n=51)	Non-abusers (n=37)
Low risk	28(54.9%)	34 (91.8%)
Medium risk	20 (39.2%)	1 (2.7%)
High risk	3 (5.9%)	2 (5.4%)

Post-release outcome

Statistical analyses were also conducted on the recidivism rates of this sample of female offenders. The recidivism rates of those identified as substance abusers were compared with the recidivism rates of non-abusers. For the current study, recidivism was defined as return to federal custody for any reason (including suspensions, revocations and new convictions). Not surprisingly, a higher proportion of female offenders identified as substance abusers failed on discretionary release (25.5%) as compared with non-abusers (8.1%). In general, this result suggests that female offenders identified as substance abusers are a higher-risk group and, therefore, may pose a unique challenge during community reintegration.

Substance abuse programming and recidivism

To examine the relationship between participation in institutional programming and post-release outcome, the sample was reduced to include only those female offenders recommended for treatment (i.e., the substance abusers). Substance abusers who had been released (n=63; which includes those on statutory release) were split into two groups based on whether they had completed substance abuse treatment while incarcerated. This rendered a sample of 29 treatment completers and 34 women who either did not participate in or did not complete substance abuse treatment. These two groups were compared on post-release outcome.

Results indicate that participation in a substance abuse treatment program was associated with a reduced failure rate on discretionary release. Substance abusing women who did not participate in a treatment program were more likely to be returned to custody (44.1%) than those who had participated in some form of substance abuse treatment (10.3%). It should be noted that the failure rate of the treated substance abusers closely approximates the failure rate of the non-abusing offenders. This finding highlights the potential benefits of participating in institutional programming.

Discussion

The findings of the current study should be interpreted with caution due to small sample sizes and other factors such as programming motivation and attrition rates. Results suggest that female parolees with a history of substance abuse are at greater risk for return to custody than non-abusers. Program participation was found to be associated with reduced returns to custody for substance abusers; the rate approximated that of female non-abusers. These results highlight the importance of using appropriate correctional treatment that targets an offender's criminogenic need.

Clearly, substance-abusing offenders possess characteristics that might substantially increase their risk of failure on discretionary release. These needs must be identified properly at admission and treated during the sentence. This will foster community reintegration, thereby mitigating the burden on the offender, the institutional population and society at large.

---

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

2. B. A. Grant, L. Motiuk, L. Brunet, L. Lefebvre and P. Couturier, Day Parole Program Review: Case Management Predictors of Outcome (Ottawa, ON: Correctional Service of Canada, 1996). See also B. A. Grant and C. Gillis, "Gradual release programs: Day parole performance and subsequent release outcome," Forum on Corrections Research, 8, 3 (1996): 19­21.

3. B. A. Grant et al., Day Parole Program Review.

4. Criminogenic needs reflect risk factors of the offender that are changeable and, when modified, reflect changes in the likelihood of recidivism.

5. K. Blanchette, Risk and Need Among Federally Sentenced Female Offenders: A Comparison of Minimum-, Medium-, and Maximum-security Inmates (Ottawa, ON: Correctional Service of Canada, 1997).

6. Full parole eligibility occurs after having served one third of the sentence. Eligibility for day parole is normally six months before full parole eligibility.