Intensive Supervision Practices:
A Preliminary Examination

2003, No. B-31, pdf

————— Research Brief—————

This report is also available in French. Ce rapport est également disponible en français. Veuillez vous adresser à la direction de la recherche, Service Correctionnel du Canada, 340 avenue Laurier ouest, Ottawa (Ontario) K1A 0P9. Should additional copies be required they can be obtained from the Research Branch, Correctional Service of Canada, 340 Laurier Ave., West, Ottawa, Ontario, K1A 0P9.

Intensive Supervision Practices:
A Preliminary Examination

by

Ralph Serin & Ben Vuong
Research Branch
Shandy Briggs
Reintegration Operations
Correctional Service of Canada
July, 2003

TABLE OF CONTENTS

ACKNOWLEDGEMENTS

We would like to thank the District Directors and the Community Council for their comments during the development of this initiative. We would also like to thank the parole officers who completed surveys regarding supervision practices.

EXECUTIVE SUMMARY

Changes in correctional practice over time have influenced community supervision guidelines. In addition, there has been an interest in being able to describe specific strategies that have been developed for distinct samples of offenders to enhance their success on parole. This preliminary examination investigates the application of Intensive Supervision Practices (ISP) to a group of 210 higher risk offenders who were conditionally released between March and June 2002. Their outcome in terms of suspension and revocation rates was compared to a similar group under supervision 12 months previously.

Selection criteria were research-based using reintegration potential ratings and a 3-month follow-up period was used regarding failure rates and time to failure for similar offenders. Some staff training preceded the implementation of the study. A survey instrument was also developed for completion by parole officers to better understand the utility of engagement and supervision strategies in the management of lower reintegration potential offenders.

The results are somewhat encouraging. The ISP group had lower rates of revocation readmission (20% reduction) and longer time until first suspension warrant was issued. For instance, the comparison group had a 24.1% shorter time to revocation or suspension compared to the ISP group. Training and programming issues are briefly discussed regarding possible expansion of ISP.

Introduction

Trends in the field of corrections have varied over time with both rehabilitation and deterrence models having held favour, depending on the jurisdiction and time period. Importantly, both approaches have been proposed as viable strategies to reduce the recidivism of released offenders. Furthermore, proponents have presented empirical and ideological arguments in support of their position for more than four decades. This is particularly striking in the areas of sentencing and community supervision.

The idea for Intensive Supervision Practices (ISPs) began to emerge during the 1950’s and 1960’s with smaller and specialised caseload functions in probation and parole. These specialised caseloads were less concerned with monitoring and enforcement, focusing on treatment of the personal and social problems that were thought to contribute to recidivism. Early examples of ISPs, all of which were grounded in behavioural principles of treatment, took individual differences into account and employed well-trained staff. A Canadian example that has exemplified the merits of ISP was the Canadian Volunteers in Correction Program (Andrews & Keissling, 1980). Programs with demonstrated effectiveness in reducing re-offending developed during this era had several common features: 1) small staff to client caseload rations; 2) well trained and/or committed program staff; 3) a primary focus on providing treatment services to clients; and 4) intensive monitoring and follow-up," reinforcing a balanced approach.

Although predominantly an American trend, with a shift in approach came an expansion of punishment and surveillance-based ISPs in the 1990s. The goals of these newer ISPs have been described to: 1) improve sentencing options; 2) ameliorate prison overcrowding; 3) lower correctional costs; and 4) reduce offender recidivism (Clear and Hardyman, 1990). In terms of prison crowding, research findings indicate that they do not reduce prison overcrowding, but may in fact contribute to the problem (Petersilia and Turner, 1991).

Comparing studies regarding ISPs is problematic for various reasons: there are wide variations in terms of the selection criteria, the training received by staff, the parole officer to offender ratio, the duration of supervision (recidivism rates vary significantly as a function of time at risk), and the actual supervision services provided. As noted above, some newer programs have a greater emphasis on supervision than rehabilitative intervention. Some authors have argued that this difference alone is related to differential effectiveness among ISPs (Gendreau, Goggin, Cullen & Andrews, 2000).

A research review that included 47 different ISP studies involving 19, 403 offenders demonstrated that there was virtually no difference in recidivism rates between offenders who received the ISPs and those who received regular probation or parole (Gendreau et al, 2000). Interestingly, however, ISPs that also included a treatment component generated a 10% reduction in recidivism. It should be noted that these ISP's did not differentiate the application of ISP to high-risk offenders. In fact, the authors expressed concern that intermediate sanctions appear to be widening the net by targeting low-risk offenders who would normally have received periods of regular probation.

A recent review of ISPs in CSC noted that the most successful ISPs appeared to offer more treatment (Paparozzi, 1999). Most program participants attended peer support sessions led by the ISP officer, 60% attended specialised counselling, and over 95% attended a treatment program. Programming, then, appears to be an important adjunct to increased surveillance for an effective ISP. Overall, it was concluded that the apparently successful programs developed during this more recent iteration of ISPs had several common features that contributed to their success: 1) small staff to client caseload ratios; 2) well trained and/or committed program staff; 3) a primary focus on providing treatment services to clients; and 4) intensive monitoring and follow-up.

Moreover, it is important to realise that research has demonstrated that the suspension rate and re-incarceration rate for ISP projects are higher: ISP offenders are "watched" more closely, and as a result, technical violations are more likely to be detected and processed. It would appear that ISP's inevitably result in about the same or slightly higher recidivism rates than comparable regular probation programs, unless ISP incorporates programming elements (Gendreau et al, 2000; Paparozzi, 1999).

For the past two decades, the Service has made a concerted commitment to reintegration programs (Motiuk & Serin, 2001), consistent with its Mission and legislative responsibilities. Most recently, there has interest in addressing responsivity factors in offenders and the development of community-based correctional programs. Strategies to enhance offender response to programs and supervision have been developed, (e.g., motivational interviewing), and the research regarding the identification of discrete groups that vary in terms of reintegration potential (Motiuk & Nafekh, 2001) has informed community supervision approaches, (e.g. frequency of contact). Furthermore, higher risk cases tend to respond better to intensive and extensive service (Kennedy, 2000), indicating that ISPs should be reserved for higher risk offenders.

Background

Against this context, there has been general interest in the evaluation of community supervision of offenders. In April 1999, the Auditor General reported that approaches vary in managing offenders who need a high level of supervision and that it has been a long-standing practice to manage these offenders using different approaches in different locations without any evaluation of their effectiveness. Accordingly, CSC committed to the Auditor General to evaluate existing approaches to managing offenders who require a high level of supervision in the community in order to identify and implement the most cost-effective approaches under different circumstances.

Rationale for Intensive Supervision Practices

Although the Service manages all offenders according to their individual risk and needs level, there is a certain group that it is believed to require more structure and control in the community to succeed on release. As noted above, the application of research on reintegration potential has provided an empirically validated method of identifying those offenders most in need of increased structure and control. The application of a common criterion, (e.g., reintegration potential - RP) is intended to meet the goal that enhanced supervision services are available consistently across the country for a specifically defined target group. Also, the use of RP ensured that these enhanced supervision practices would be allocated for higher risk cases, consistent with the research findings.

Methodology

Following a review of the criteria for inclusion in ISP based on current frequency of contact guidelines, the Research Branch used RP at intake and release to identify ISP candidates. The process for calculating RPs varies for non-Aboriginal and Aboriginal offenders. For non-Aboriginal offenders, the Custody Rating Scale, Statistical Information on Recidivism Scale, and Offender Intake Assessment static and dynamic factor ratings are combined (Motiuk & Nafekh, 2001). For Aboriginal and women offenders, The Statistical Information on Recidivism - Revised is not utilised.

For the purposes of the study period, ISP was applied to Statutory Release (SR) cases meeting the following criteria: 1) low RP rating at Intake, 2) low RP rating at release, and 3) assessed as requiring Level A intervention upon release

An analysis of a snapshot of the community supervision population yielded estimates of the number of ISP candidates for the implementation period. From a community-snapshot sample of 9,899 offenders collected 1 January 2001, there were 2,794 cases on Level A supervision (28.2%). This group comprised 23.5% Day Parole, 19.2% Full Parole, and 57.2% Statutory Release. With this Level A group, 26.1% had low RP, 35.5% had moderate RP, and 38.4% had high RP. Using low RP at intake and release as the criterion for ISP, there were 355 cases or 15.0% of all Level A cases. This yields an estimate of 5.1% of the total community supervision population for whom ISP would be warranted.

The duration of ISP was also empirically derived. Using data from Performance Assurance for time from release to suspension for all revocation admissions between December 1997 and November 1999, rates of failure within time frames were calculated. Of those who failed, 57.7% of all Day parole releases and 59.6% of all Statutory Releases fail within 3 months of release (they constitute the vast majority of Level A cases). The decision was therefore made to evaluate the implementation phase over a 3-month time period. This was a continuous intake model such that offenders were under ISP for 3 months, so data collection occurred for 6 months to ensure offenders released in March 2002 would be followed until June 30, 2002.

Implementation

Following several national consultation meetings, ISP was implemented across all 5 regions on March 26, 2002, such that the study period was from March 26, 2002 to June 30, 2002. Accordingly, the evaluation period continued until September 2002 to ensure that all cases had at least 3 months opportunity under ISP. Guidelines regarding admission, exclusion, and supervision criteria were distributed electronically to all parole offices in a Case Management Bulletin. Also, a survey of ISP practices was developed and parole officers were requested to complete the survey at the conclusion of 3 months or when the offender was revoked or terminated from ISP.

As part of the implementation process, community parole officers (CPO) were required to review each case against the ISP criteria during release preparation and identify target cases in Community Strategy report. Moreover, they were to ensure that the essential components of ISP were provided in community supervision for each case meeting the criteria.

These essential components included the cases being seen as often as necessary to effectively monitor reintegration efforts, but minimally eight face-to-face contacts with the CPO per month. CPOs Conducted increased collateral contacts as well as case conferences with their supervisor. Lastly, it was required that CPO's make every possible effort to implement appropriate treatment interventions in the supervision strategy for ISP cases.

Training

Research from other jurisdictions indicated that for ISP to be maximally effective, ISP should incorporate increased supervisory contact for higher risk cases as well as increased programming (Kennedy, 2000). One aspect of programming considered to be important was the application of specific engagement strategies such as motivational interviewing. Although it is clear that some parole officers are well versed in such approaches, specific training was not provided prior to implementation of ISP as a national practice.

Comparison Groups

Comparison Group 1 (excluded)

There was a group of offenders who met the criteria for ISP, but who were excluded for a variety of reasons by the supervising office. During the ISP pilot, there were 45 offenders who represent this comparison group 1. Their community performance was also compared to the ISP group. It should be noted that the majority of this group had a residency condition imposed upon release by the National Parole Board. Residency is imposed for cases for whom there are specific concerns regarding the likelihood of a future violent offence prior to warrant expiry date. In this respect, this group may be slightly different than the ISP cases per se. Despite meeting the initial entry criteria, given that they had daily contact with staff at the community residential facility, they were excluded from ISP.

Comparison Group 2 (Control)

A comparison group 2 or control group was created using the same selection criteria for ISP and retrospectively applying to a cohort who was under community supervision from March 26 to June 30, 2001. This yielded a group of 244 offenders who met similar criteria to the ISP group, but for whom supervision could be described as status quo. Their community performance and outcome was compared to the ISP group.

Results

One aspect of the results was to determine parole officers' use of different strategies to assist community supervision of these high-risk cases assigned to ISP. There were two related but different sets of strategies that were systematically assessed through use of a survey. Parole officers were asked to report their frequency of use of six different engagement and four different supervision strategies. The former were intended to assist in getting offenders to "buy in" or comply with supervision conditions, while the focus of the latter was monitoring or surveillance.

The survey results are somewhat difficult to interpret, because parole officers indicated their understanding of the utility of a particular strategy independent of whether they used it for the specific offender. Accordingly, cells do not total 100. It is important to note that there is no baseline data with respect to non-ISP cases. Also notwithstanding the provision of definitions with the surveys, it is possible that understanding of the specifics of these strategies could vary among parole officers. Parole officers may rate cases with special conditions differently than other offenders. Lastly, it was possible for a parole officer to determine a particular strategy to be helpful, but not employ it for a variety of reasons not captured by the survey. Nonetheless, the results provide some understanding of parole officers' views about supervision and engagement strategies that form ISP.

Table 1 presents the engagement/compliance strategies in terms of usage and perceived helpfulness by the CPO. Importantly, the full range of strategies were used very often (greater than 80% of the time). In addition, the perceived helpfulness was moderately high (approximately 60-70% rated as very and somewhat helpful). Stated another way, the strategies were seen as not being helpful in fewer than one -third of the cases.

Table 2 presents similar data for supervision strategies. Rates of endorsement by CPOs were not quite as high as engagement strategies. Of note, curfews were used in less than half the ISP cases. Despite this low endorsement rate for curfews, they were seen as helpful in 85% of the cases. That is, in general, curfews are perceived to be helpful for use with ISP cases. Perceived helpfulness for the other strategies was slightly higher than for the engagement strategies (greater than 80%).

Outcomes

Several meetings were convened to arrive at consensus regarding the most appropriate data capture strategy from the Offender management System (OMS) and Criminal Justice Information Library (CJIL). Specifically, supervision status was defined in terms of point in time within a supervision period. Preliminary analyses revealed that failure to attend to common definitions sometimes resulted in capturing from different fields within OMS and, therefore, different rates of suspension.

The three groups - ISP, excluded and control were compared in terms of the proportion for which a suspension warrant was issued, the total number of warrants issued, the average time to first warrant issued, and rates of revocation with violent and nonviolent offence or outstanding charges. These results are presented in Table 3.

There was no statistically significant difference among the three groups in terms of proportion with a suspension warrant issued. For instance, 70% for the ISP group (147 of the 210 ISP cases) had a suspension warrant issued within the first 3 months of release. The rates are 64% for the excluded group and 71% for the control group. Yet, there was a difference among groups in terms of time to first warrant issued. The average time for the ISP group was 29 days, compared to 19 days for the excluded group, and 22 days for the control group. The excluded group was noticeably faster to fail, whereas the ISP group was slower. Relative to the ISP group, the control group had a 24.1% faster average time to fail (29-22/29 x 100).

Results vary when comparing actual offenders versus warrants being issued. For the ISP group, there were 216 warrants issued for the sample of 210 offenders. This represents 147 offenders for whom warrants were issued (70%). Similarly, for the control group, there were 221 warrants issued for the sample of 244 offenders. This represents 173 offenders for whom warrants were issued (71%). There were slight differences between groups in terms of the proportion of warrants withdrawn and cancelled. The ISP group appears to have a slightly greater percentage of warrants withdrawn (13% versus 10% for the control group) and cancelled locally (31% versus 21% for the control group). A speculation is that local cancellations reflect close work with the offender and a wider examination of alternatives to incarceration. The control group has a slightly higher proportion of warrants cancelled by the National Parole Board (6% versus 3% for the ISP group). We have no explanation for these findings.

In terms of revocations, the ISP group had a 16.7% relative reduction compared to the control group (48 - 40/48 x 100). Rate of revocation warrants without an offence was comparable for both groups (57% for the ISP group versus 60% for the control group), but the ISP group had a higher rate of revocation warrants with offence (39% versus 33%). When these results are considered in terms of actual offenders, the ISP group and control group have similar rates of revocations with an offence (16%). However, when considering actual offenders for revocations without an offence, the ISP group had a relative reduction of 25% (32-24/32 x 100).

When inspecting the results in cases with violent and nonviolent revocations, there are some encouraging trends. The ISP group had a 25% relative reduction in the proportion of revocations for a nonviolent offence. In summary, ISP cases appear to remain longer in the community with no increase in revocations for violence offences or outstanding charges.

Discussion

As noted in the introduction, the purpose of ISP is to provide enhanced structured community supervision to offenders with a demonstrated higher risk and need profile. This enhanced intervention includes both supervision strategies and correctional programming. The latter is reflected in the offenders' Correctional Treatment Plan that is developed at the offenders' admission to federal prison and is modified throughout their sentence, culminating at warrant expiry. Given the nature of this sample of offenders, engagement strategies to facilitate programming are seen to be a particularly important aspect of ISP.

Prior research that only increases supervision such as with increased frequency of contact has yielded poorer results than initiatives that have attempted to blend supervision and programming. Research has also identified the importance of staff training and parole officer skills in yielding improved parole outcomes.

As an initial project, some training occurred prior to implementation; however, additional training will incorporated into the parole officer professional development training initiative. Also, although programming was provided consistent with current practices, additional research comparing program attrition among groups would be informative. Notwithstanding these caveats, the current study yielded some promising findings relative to a comparison group. Most importantly, the proportion of offenders who were revoked was lower for the ISP group. Furthermore, time to the first warrant being issued was also longer for the ISP group. This means that during the ISP initiative, the sample of higher risk offenders successfully remained in the community longer without increasing the risk to the public in terms of readmission. Finally, the poorest outcome (time to failure, not failure rates) was for the excluded group, suggesting that a review of the criteria utilised for excluding ISP cases is warranted.

From the survey results, it appears that CPOs frequently utilise both engagement and supervision strategies quite consistently. It was reported that CPOS used engagement strategies slightly more frequently, but they perceived supervision strategies to be slightly more helpful. Further research distinguishing between engagement and supervision and their contribution to successful reintegration would be informative.

This preliminary study has several limitations. Important is the fact that the survey was only applied prospectively for the ISP cases. As a result, we do not know the extent to which these strategies were in place for the control group a year previously. In terms of recidivism data, the actual numbers of cases are also quite small and reflect a short follow-up period. These issues make extrapolation of the data difficult and may have contributed to group differences failing to achieve statistical significance. The findings are encouraging but not striking, raising some concerns regarding the cost effectiveness of ISP. Finally, it is unclear the extent to which ISP practices might be incorporated into existing operations, in terms of entry criteria and engagement and supervision strategies. Extension of ISP should include additional training and more systematic application of programming.

References

Andrews, D. A. & Keissling, J. J. (1980). Program structure and effective correctional practices: A summary of the CAVIC research. In R. R. Ross & P. Gendreau (Eds.), Effective correctional treatment. Toronto: Butterworth.

Clear, T. & Hardyman, P. L. (1990). The new intensive supervision movement. Crime and Delinquency, 36 (1), 42-60.

Gendreau, P., Goggin, C., Cullen, F. T., & Andrews D. A. (2000). The effects of community sanctions and incarceration on recidivism. Forum on Corrections Research, 12 (3), 10-13.

Kennedy, S. M. (2000). Treatment responsivity: Reducing recidivism by enhancing treatment effectiveness. Forum on Corrections Research, 12 (2), 19-23.

Motiuk, L. L. & Nafekh, M. (2001). Using reintegration potential at intake to better identify safe release candidates. Forum on Corrections Research, 13 (1), 11-13.

Motiuk, L. L. & Serin, R. C. (2001). Compendium 2000 on effective corrections. Correctional Service Canada, Ministry of Supply and Services.

Paparozzi, M. (1999). Evaluation of intensive supervision programs (ISP) in Montreal and Toronto. Unpublished manuscript.

Petersilia, J. & Turner, S. (1990). Diverting prisoners to intensive supervision in California. Journal of Law and Criminology, 82 (3), 610-658.