Intensive Support Units (ISU) for Federal Offenders with Substance Abuse Problems: An Impact Analysis

2005 N ^o R-151

Brian A. Grant
David D. Varis
&
Derek Lefebvre

Addictions Research Centre
Research Branch, Policy Planning and Coordination
Correctional Service Canada

March 2005

ACKNOWLEDGEMENTS

This research could not have been conducted without the full support of Westmorland, Leclerc, Joyceville, Drumheller and Mission institutions. Wardens and their senior management team, Unit Managers, and front-line staff were instrumental not only in implementing an Intensive Support Unit in their respective facility, but also in the research component of this national pilot initiative. Moreover, many dedicated regional and national headquarters staff provided invaluable project implementation documentation and assistance, in particular, Dr. John Weekes, Reintegration Programs, Dr. Andrea Moser, CSC National Drug Strategy, and Irene Klassen, Performance Assurance, NHQ. As well, the authors would like to thank the many Research Branch staff who played a significant role in this project including Christa Gillis (development of measures), Gilbert Taylor (initial project management), Stafford Murphy (site descriptions and initial co-ordination of research assistants), Sara Johnson (data analysis), Mark Nafekh (statistical analysis advice) and Collette Cousineau (data extraction from OMS). Given the duration of on-site data collection and analysis, there were many research assistants involved in this project and we thank Diane Leblanc, Julie Lambert, Jennifer Ruccolo, Michelle Wylie, Donette Hamula, Darcy Ginn, Joe Yuzik, Richard Kalman, Diana Doherty, Fred Cardwell, Jennifer MacPherson and Mark Edwards. We appreciate greatly the review of the report that Peggy Mullins, Research Analyst, undertook upon its completion. Also, we thank our colleagues at the Addiction Research Centre, who supported and assisted in many ways, while adapting to the new facility in Montague , Prince Edward Island . Finally, we thank the offenders who participated in this study. Their willingness to openly share personal experiences and information about the difficulties and problems associated with substance use was critical to seeing this study to fruition.

  EXECUTIVE SUMMARY

Approximately 80% of offenders admitted to federal penitentiaries are identified as having a substance abuse problem that is associated with their criminal behaviour on admission to prison (Grant, Kunic, MacPherson, McKeown, & Hansen, 2004). To address this challenge, the Correctional Service Canada (CSC) created specialized Intensive Support Units (ISUs). The units were designed to provide offenders: 1) with a supportive environment using specially trained personnel and 2) reduced presence of drugs and alcohol through increased searching and drug testing, beyond what is specified under Canadian law.

Offenders volunteer to live on the ISU to more effectively address their drug and alcohol problems. There are no special programs or privileges and they must sign a contract agreeing to remain drug free, abide by the rules of the units and accept higher rates of searching and drug testing. Offenders living in the ISU continue to participate in regular institutional activities and have been, or are currently involved in substance abuse treatment programs.

The study sample includes 246 male offenders, admitted to an ISU, over a period of 11 months (September 2000 – July 2001) who agreed to participate in the research (of 309 ISU admissions; 80% agreed to participate), The five ISU pilot sites were located in each of CSCs five regions; two in minimum security institutions (Westmorland, Drumheller – Minimum Security Unit) and three in medium security institutions (Leclerc, Joyceville, Mission).

Offenders were followed after the study period for between 8 and 24 months to determine the type of release they received and the amount of time they spent in the community. Offenders completed a test battery at admission to the units and at the time of their discharge from the unit to determine perceptions about the program, their understanding of substance abuse and the issues surrounding their drug and alcohol problems. Changes in behaviour were measured using a monthly behaviour checklist, institutional charges, search and seizure data and results of drug testing. Follow-up outcome was measured using time to release, type of release, and whether they were returned to custody. Appropriate comparison groups were used to evaluate the results.

Most offenders (80%) in the ISUs had a substance abuse problem linked to their criminal behaviour. However, 1 in 5 did not have a substance abuse problem choosing to live in the ISU to reduce their exposure to drugs. ISU offenders were younger, serving shorter sentences (fewer serving life sentences), less likely to be serving a sentence for a violent offence, and were rated as lower risk on static factors (criminal history) than other offenders in minimum and medium security institutions. Aboriginal offenders accounted for 12% of ISU participants, but accounted for 16% of the offender population, indicating they were under-represented in the ISU population.

One of the objectives of the ISUs was to reduce the presence of drugs and alcohol on the units. This was accomplished by a higher level of searching (as compared to other units in the institutions). Even with the higher level of searching less substance-related contraband was found, relative to other units in the institutions. In addition, random drug testing results indicated a decline in drug use detection from 15% before admission to 3% after admission.

Results from questionnaires given to both staff and offenders also indicated that drugs and alcohol were less available in the ISUs than in other parts of the institution. Staff also perceived the increased interdiction activities and offender motivation as contributing to the reduction in drugs and alcohol being available.

Participation in the ISU resulted in a higher rate of discretionary release (day parole, parole) (62%) relative to the matched comparison group (46%); a 34% higher rate of discretionary release for the ISU participants.

ISU participants were 36% less likely to be returned to custody than offenders in the matched comparison group (25% vs. 39%). In addition, ISU participants were 40% less likely to be returned to custody for a new offence than the offenders in the matched comparison group (6.5% vs. 10.9%). Results of questionnaire data indicate that neither staff nor offenders perceived the ISUs as

having a negative impact on the rest of the institution. Other results from questionnaires measuring perceptions of the ISUs indicated that the supportive environment available in the ISUs helped offenders to address their substance abuse problem. The environment was created by a number of factors including the professional orientation of ISU staff (more empathetic, rehabilitation oriented, satisfied with their work) and the use of interdiction activities that reduced the presence of drugs and alcohol.

Potential cost savings of $8,000 per participant were calculated based on decreased incarceration time resulting from earlier release and reduced likelihood of readmission.

Additional work is needed to refine and develop the ISU concept. In particular, work needs to be done to better define how to create and maintain the necessary supportive environment, the level of staff training required, the level of drug interdiction activities, and continued access to programming and work opportunities. The pilot sites were all highly motivated to demonstrate the effectiveness of the concept. There is a need to determine if the benefits identified in this study can be replicated following national implementation.

TABLE OF CONTENTS

ACKNOWLEDGEMENTS

EXECUTIVE SUMMARY

LIST OF TABLES

LIST OF FIGURES

INTRODUCTION

• Overview of Intensive Support Units (ISU)
• Research Questions

METHOD

• Sample
• Study Sites
• Data Sources
• Offender Measures
• Profile Information
  • Static and Dynamic Factor Ratings
  • Severity of Substance Abuse Problem
• Offender Substance Abuse Information
  • Intake Interview
  • Offender Substance Abuse Knowledge
    • Consequences of Alcohol Use
    • Consequences of Drug Use
    • How Much Do They Matter?
    • Relapse Attitude/Knowledge
• Offender Motivation, Expectations, and Response Reliability
  • Readiness for Change
    • University of Rhode Island Change Assessment - URICA
  • Offender Expectation Questionnaire
  • Offender Post Evaluation Questionnaire
  • Balanced Inventory of Desirable Responding
• ISU Outcomes
  • Community Support Visits
  • Program Participation
  • Employment
  • Peer Group Meetings
  • Search and Seizure
  • Offender Behavioural Indicators Assessment Report
• Institutional Outcomes
  • Urinalysis
    • Random Urinalysis
    • Unit Screening Test
  • Institutional Charges
• Release and Readmission Outcomes
  • Type of Release
  • Return to Custody
• ISU Staff Measures
  • Expectation Questionnaire
  • Impact Survey
  • Staff Substance Abuse Knowledge
    • Consequences of Alcohol Use
    • Consequences of Drug Use
    • How Much Do They Matter?
  • Job Orientation
    • Professional Orientation Sub-Scale
    • Total Job Satisfaction Sub-Scale
• Procedure
  • Selection and Participation Process
    • Pre-Admission
    • Post-Admission
    • Discharge
  • Project Management
    • Data Collection
    • Data Coding
    • Data Analyses

RESULTS

• Offender Profile
  • Age, Marital Status, and Race
  • Sentence Length
  • Offence Type
  • Static and Dynamic Factors at Admission
  • Severity of Substance Abuse Problem
• Substance Abuse History
  • Substance Use and Abuse Background
    • Drug and Alcohol Problems
    • Early Alcohol and Drug Use
    • Reasons for Use and Criminal Behaviour
    • Drug and Alcohol Use Prior to Offence
    • Use During Incarceration
    • Treatment
  • Substance Abuse Knowledge
• Offender Motivation, Expectations, and Response Reliability
  • Readiness for Change
  • Reasons for Moving to the ISU
  • Offender Expectations
  • Perceived Effectiveness
  • Expected Benefits
  • Expected ISU Support
  • Impact of the ISU on Institution
  • Response Reliability
• ISU Outcomes
  • Activities (Offender Participation and Association)
  • Behavioural Indicators
  • Search and Seizure Outcomes
• Institutional Outcomes
  • Drug Testing
    • Random Urinalysis
    • Screening Tests
  • Institutional Charges
  • Offender Discharge Data
• Release and Readmission Outcomes
  • Type of Release
  • Return to Custody
  • Survival Analysis
• ISU Staff
  • Staff Expectation and Impact Scales
  • Individual Items
  • Expectation Item Themes
  • Impact Item Themes
  • Staff Knowledge and Job Orientation

DISCUSSION

• Research Questions
• Other findings
• Limitations
• Implications

REFERENCES

APPENDIX A: Informed Consent Form

APPENDIX B: Offender Intake Interview Questionnaire

APPENDIX C: Offender Knowledge Survey

APPENDIX D: University of Rhode Island Change Assessment (URICA)

APPENDIX E: Offender Expectation Questionnaire

APPENDIX F: Offender Post Evaluation Questionnaire

APPENDIX G: Balanced Inventory of Desirable Responding - Version 6

APPENDIX H: Offender Process Measures

APPENDIX I: Offender Behavioural Indicators Assessment

APPENDIX J: Staff Expectation Questionnaire

APPENDIX K: Impact Survey

APPENDIX L: Staff Knowledge Survey

APPENDIX M: Staff Comparison Scales (CSC Staff Survey 1996)

APPENDIX N: Offender Expectation Questionnaire (Admission)

APPENDIX O: Offender Expectation Questionnaire (6 weeks)

APPENDIX P: Offender Post Evaluation Questionnaire

APPENDIX Q: Staff Expectations

APPENDIX R: Staff Impact

APPENDIX S: Process Measures Results

LIST OF TABLES

Table 1: Number of ISU Participants and Staff Who Consented to be Part of the Research

Table 2: Sentence Length for ISU and Comparison Group

Table 3: Static and Dynamic Factor Ratings for ISU and Comparison Group

Table 4: Ratings for Dynamic Factor Domains for ISU and Comparison Group

Table 5: Percentage of Offenders that Scored as Moderate, Substantial or Severe on the ADS, DAST and a Combination of Both Scales

Table 6: Level of Knowledge About Consequences of Substance Use and Coping Challenges for ISU and OSAPP Participants

Table 7: Offender Readiness for Change Scores

Table 8: Reasons for Moving Into an ISU

Table 9: Expectation Scales Comparisons at Admission, Six Weeks, and Upon Discharge

Table 10: Behavioural Ratings Over Five Months

Table 11: Percentage of Positive Urinalysis Tests a Before, During, and After ISU

Table 12: Number of Institutional Charges Received While in the ISU and of Offenders Found Guilty

Table 13: Number of Institutional Charges (Major and Minor) by Type of Charge Received While in the ISU

Table 14: Type of Release for ISU Participants and the Matched Group

Table 15: Returned to Custody With 12-Month Follow-Up

Table 16: Type of Readmission During 12-Month Follow-Up for Voluntarily and Involuntarily Discharged ISU Group and the Matched Comparison Group

Table 17 : Days Out on Release in 12 Month Time Period (Offenders Not Returned Assigned the Full 12 Months)

Table 18: Staff Expectation and Impact Scales

Table 19: Staff Expectations Themes and Items

Table 20: Staff Impact Themes and Items

Table 21: Staff Knowledge of Substance Use Compared to ISU and OSAPP Participants

Table 22: Staff Professional Orientation and Job Satisfaction Scores

LIST OF FIGURES

Figure 1: . Rate of Cell and Individual Searches by Unit Type

Figure 2: . Rate of Substance Related Contraband Seizures from Personal Searches by Unit Type

Figure 3 : Rate of Common Area Searching by Unit Type

Figure 4 : Rate of Common Area Seizure by Unit Type

Figure 5 : Survival Analysis Curve for a 2-Year Follow-Up of Readmission for Voluntary Dischargees, Involuntary Dischargees and Matched Group

INTRODUCTION

The use and abuse of drugs and alcohol has been consistently shown to be associated with criminal behaviour (Andrews & Bonta, 2002; Klingman & Hunt, 1998; Gaines & Kraska, 1997; Walters, 1994). The association between substance abuse and criminal behaviour has been widely examined and, while a causative relationship has not been established, the literature provides substantial information on the link between drug and alcohol abuse and criminal behaviour.

Pernanen and Brochu (1997) present a comprehensive review of studies on the links between alcohol/drugs and crime from a number of European countries, United States and Canada . They report that most of the studies linking substance use to violent crimes show a sizeable presence of alcohol. The percentage of alcohol use in assault and homicide offences are highest in Scandinavian countries with a range of between 70% to 80%, with lesser values in the United States at 50% to 60% and Canada at 40% to 50%. In terms of non-violent crimes, the authors report that, despite the paucity of research in this area, 40% of offenders or more were under the influence of alcohol during the commission of property-related crimes. As well, the authors indicate that in a large number of cases, the use of illicit drugs by offenders committing an offence is often preceded by alcohol consumption and, even when an offender is not under the influence of an illicit drug, property crimes such as robberies are often linked to drug acquisition and use (Pernanen & Brochu, 1997).

The Correctional Service Canada (CSC) has established that 80% of offenders admitted to its penitentiaries have a problem with alcohol and/or other drugs ( Grant, Kunic, MacPherson, McKeown., & Hansen, 2003). Of those, 61% abuse alcohol, 69% abuse drugs, 47% abuse both alcohol and drugs, and 44% are poly-drug users. Additionally, a comprehensive study of 8,598 inmates admitted to Canadian federal institutions between 1993 to 1995 revealed that 54% of offenders reported having been under the influence of a psychoactive substance at the time of their most serious offence (Pernanen, Cousineau, Brochu, & Sun, 2002). An earlier analysis of these data by Brochu, Cousineau, Gillet, Cournoyer, Pernanen, & Motiuk ( 2001) indicated that 44% of the most serious crimes committed by offenders admitted to federal custody during this period were associated with the use of, and addiction to, alcohol, illicit drugs, or both.

Substance abuse places significant demands on criminal justice, social service and health care systems in Canadian society (Zilkowsky, 2001). It was estimated that in 1992, misuse of alcohol and illicit use of drugs cost Canadians $8.9 billion in lost productivity, law enforcement, and direct health care expenditures ( Single, Robson, Xie, & Rehm, 1998). Canadian penal costs associated with alcohol and drug use in 1992 were estimated at over half a billion dollars ( Single, Robson, Xie, & Rehm , 1998).

In an effort to combat both fiscal and social costs associated with substance use, CSC actively encourages and assists offenders to become law-abiding citizens, thereby contributing to the protection of society (Correctional Service Canada, 2004). Moreover, a reduction in recidivism through effective correctional treatment has been shown to be cost effective (Brown, 2001; Grant et al., 2003; T3 Associates, 1999). With a renewed emphasis on combating the substance abuse problem among offenders, many correctional organizations have been examining other strategies beyond traditional treatment and programs.

During the last several years, a number of correctional jurisdictions, including Her Majesty's Prison Service in Great Britain ( Johnson & Farren, 1996) , the South Australian Department of Correctional Services ( Incorvaia & Kirby, 1997), the Dutch Ministry of Justice Services ( Schippers, Van Den Hurk , Breteler, & Meerkerk, 1998; van Doorninck & de Jong, 2001) and several federal and state correctional institutions in the United States ( Peters & Steinburg, 2000), have adopted specialized units that focus on limiting the availability of drugs. These units have been commonly referred to as "drug-free" units. While some focus solely on drug interdiction through increased searching, others provide a multi-faceted approach combining drug interdiction measures with treatment services.

The recent emergence of "drug-free" units within correctional systems has occurred despite limited research. The literature that does exist examines various aspects of these units ranging from degree of support to behavioural outcomes. An inmate survey designed and conducted at a British prison indicated that 61% (453) of the inmate population would prefer to live in a substance free zone ( Johnson & Farren, 1996). A sub-sample (352) of both drug users and non-drug users in this study responded to more detailed questions about substance free zones. Of these respondents, 85% felt that the units were a good idea. When asked to identify the types of supports they felt would be good for the zone, 72% preferred trained staff, 63% favoured one-on-one counselling, 59% indicated the need for a support group, and 57% chose an education/awareness group. Inmates also identified those methods of monitoring that they would support including removal of drug users (77%), urine checks (77%), signed contracts (67%), room searches (47%), body searches (43%) and reporting of rule breakers (25%).

A study in a S outh Australian prison in a Drug Free Unit (DFU) compared drug taking behaviour of DFU inmates with that of the general population ( Incorvaia & Kirby, 1997). DFU inmates reported significantly lower drug use (33%) than other offenders (84%). Even though there was drug taking by DFU inmates, those inmates also reported that the unit was successful in terms of meeting its objective of being an environment absent of drug related and violent incidents ( Incorvaia & Kirby, 1997).

Research on inmate behavioural outcomes relating to psychosocial functioning or criminal recidivism following participation in a "drug-free" environment remains limited. Two studies conducted in the Netherlands (Breteler, Van Den Hurk, Schippers, & Meerkerk, 1996; Schippers et al., 1998) were unable to demonstrate differences in recidivism for offenders who resided in a drug free unit that included additional programming in comparison to addicted offenders who resided in a regular prison unit. In the study by Schippers et al., offenders who had been in the unit were more effective at following a post-release treatment plan, however, this did not impact their criminal behaviour.

Overall, the evidence on drug free units is limited and does not provide strong support for their efficacy. However, other research (Andrews et al., 1990; Gendreau & Ross, 1984; Gendreau et al., 1996) has shown that providing interventions that address criminogenic needs like substance abuse can lead to a reduction in criminal behaviour. The Correctional Service of Canada provides offenders with access to a variety of substance abuse programs; creating living units that focus on substance abuse problems could be one method of increasing the effectiveness of the programs. In February 2000, CSC developed a series of pilot projects to test the concept of Intensive Support Units, units where increased drug interdiction would be carried out while at the same time providing increased support to offenders with substance abuse problems.

Overview of Intensive Support Units (ISU)

The main purpose of the ISU is to provide a safe environment where offenders can live substance-free with enhanced support from staff. The units are available to both offenders with substance abuse problems and to individuals without substance abuse problems who wish to live in an environment that is free of alcohol and drugs and the interpersonal problems associated with their use.

Offenders volunteer to reside in the unit and sign a consent agreement that outlines provisions for additional drug testing and searches in order to minimize the availability of drugs. Offenders who use drugs or alcohol or break the rules of the unit (e.g., positive drug test or possession of drug contraband) are automatically removed from the unit. However, they are eligible to reapply to live in the unit again after a mandatory absence of no less than 30 days.

Staff employ numerous security measures to inhibit drug use and availability, including restricted and monitored access to the unit by other inmates, increased use of drug testing, searching, electronic security devices (ion scanners), and enhanced dynamic security (staff and offender interaction). ISU staff receive an orientation to issues of addiction and recovery, and the legal framework and guiding principles under which the units operate.

A major part of the staff role is to foster a positive environment and to work actively with offenders to assist them in successfully changing their substance use behaviour. Studies on prison environment point to social climate, particularly the dimension of support (defined as reliable and tangible assistance for self-improvement) as a key factor in prison adjustment and healthier behavioural outcomes for inmates ( Wright, 1993).

Offenders living in the unit have access to substance abuse programs including the Offender Substance Abuse Pre-release Program (OSAPP) which is a moderate intensity cognitive behavioural program, and, at one site, Leclerc Institution, the High Intensity Substance Abuse Program (HISAP) that has recently been implemented. While program participation occurs outside of the unit, staff support offenders as they apply learned principles and skills to their daily living. Some offenders participate in Alcoholics Anonymous (AA) and Narcotics Anonymous (NA) and utilize peer support within the unit, which also contributes to the supportive environment.

Research Questions

The study focuses on answering four basic research questions that are central to evaluating the national ISU pilot initiative:

1. What are the characteristics of inmates residing in these units?
   To determine which offenders were accessing these units and whether their profile differed from that of the general offender population, the following variables were examined: age, marital status, race, sentence length, offence type, and risk level based on static and dynamic factors. Since admission to these units was voluntary and did not discriminate between those offenders with substance abuse problems and those without a problem, profile information was useful to determine the percentage of offenders with alcohol and drug problems who were residing in the units and the severity of their problem. The study also examined the drug and alcohol history of ISU offenders, as well as their level of knowledge about the consequences of substance use, readiness for change, reasons for moving to the ISU, and expectations. The characteristic data are particularly important to determine if ISU interventions and strategies are applicable to the general offender population.
   
2. Do these units reduce alcohol and drug availability?
   Since the main purpose of the ISU is to provide a substance free environment for offenders, an examination of data relating to searches conducted in the unit and seizure of drug related contraband was undertaken. Search and seizure data from the ISUs were compared to those from non-ISUs within the same institution.
   
3. Is there an impact on offender behaviour?
   Since providing a supportive and "drug-free" environment where offenders can address their substance abuse problem assumes a potential for positive behavioural change, this study examined specific offender outcomes to determine if there was indeed an impact. These behavioural outcomes included change of behaviour while in the unit, level of misconduct, substance use, and success on release. A comparative analysis of ISU offenders and a matched group of non-ISU offenders was undertaken.
   
4. Is there an impact on the rest of the institution?
   Change within a prison, whether introducing new correctional programs or adopting particular strategies and interventions, can sometimes impact on the environment. Data obtained from pre and post survey measures were examined to determine if offenders and staff perceived an impact of the ISU on the rest of the institution.

  METHOD

Sample

The offender sample includes 246 males admitted to one of five pilot ISUs ¹ between September 1, 2000 and July 31, 2001 who provided consent to participate in the research (See Appendix A for consent form). During the 11 month study period 309 offenders entered the ISU's and 80% agreed to participate in the research. All study participants were in minimum or medium security institutions.

In addition to the ISU sample, three other offender groups were used for purposes of comparison to the ISU participants. The first was a comparison group including all male offenders in custody in minimum and medium security institutions on January 1, 2001 . A second group was derived from the larger comparison group as a matched group to the ISU participants. The variables used to match the groups were: age, security level, sentence length, overall static factor level, overall dynamic factor level, and substance abuse domain level. The third group consisted of offenders who had completed CSC's Offender Substance Abuse Pre-Release Program (OSAPP).

Data were also collected from 99 staff members, who worked in the units or closely with ISU offenders. Staff were primarily correctional officers, parole officers, correctional supervisors and unit managers. The high level of response for Westmorland Institution results from the large number of staff who worked in the unit and not a higher rate of responding. This reflects the staff dynamic of the institution where correctional staff are responsible for the supervision of all offenders, whether they reside in the ISU or in the other living units.

¹ISU`s began operating in February 2000 but full data collection was not available until September 2000. During the February to August period 335 offenders started in the ISU`s

Study Sites

The number of participants from each site are presented in Table 1.

Table 1 : Number of ISU Participants and Staff Who Consented to be Part of the Research

The ISUs operated under similar principles, but each ISU site had slightly different operating characteristics. These are described below.

Westmorland Institution, located in New Brunswick , is a minimum security institution with ten houses (60 beds) assigned ISU status. Being a minimum-security institution, there are no barriers or fences separating the ISU houses from the rest of the institution. There are rules, however, forbidding offenders from entering houses other than their own. These rules apply to the institution as a whole and are not specific to the ISU. All correctional officers have responsibility over the ISU operation given the institutional setting. While not part of the ISU mandate, Westmorland introduced a specialized substance abuse program, called Bridges, that is multi-faceted, incorporating educational, peer support and community readiness components. Other than increased searches and drug testing, the ISU houses are basically managed the same as the rest of the houses in the institution.

Joyceville Institution, located in Ontario , is a medium security institution with one range of a regular unit designated as the ISU (40 beds). This range was formerly a 'drug free' area several years prior to the ISU initiative and was modelled after the " Portage " substance abuse treatment model used in the Quebec region, which relies heavily on peer support. The current ISU still incorporates this strong peer support component with emphasis on group meetings and use of peer facilitators to provide on-going support. There are three Correctional Officers assigned to the range on a full-time basis, working two day shifts and one afternoon/evening shift. The unit is also supervised by regular institutional security staff that rotate through this assigned area. Other than the institutional rule forbidding offenders from going onto ranges other than their own, ISU participants are not isolated; rather, they work and share recreation time with the general population. ISU participants receive some privileges in that they have a bit more freedom on their range compared to other units. On the other hand, they are drug tested more frequently and are expected to participate in unit meetings.

Drumheller Institution, located in Alberta , is a combined medium/minimum security institution. The ISU comprises a major part of the Minimum Security Unit (MSU), which is located outside of the confines of the regular institution. Seven houses with eight offenders each are assigned ISU status (56 beds). Although there was interest from both staff and offenders to declare the entire unit as an ISU, staff need to have at least one of the two remaining houses designated as non-ISU in case an ISU offender is removed for drug involvement. All correctional staff working in the MSU have responsibility of supervising ISU offenders. Being minimum-security, there are no barriers or fences separating the ISU houses from the rest of the houses that make up the MSU. An institutional rule forbids offenders from entering houses other than their own. ISU house phones are activated 24 hours a day instead of being shut down at 10:30 p.m. , as is the case in the other non-ISU houses. Staff work with offender representatives to encourage them to take an active role in managing their own 'community' through special recognition events and awards.

Mission Institution, located in British Columbia , is a medium security facility. The ISU is a separate unit and houses approximately 50 offenders. This unit, commonly referred to as Douglas Manor, was previously designated as 'drug free' and had been in operation as such for several years. There are four other living units at Mission with no barriers or fences separating the ISU from the rest of the institution. The ISU is supervised by a team of correctional staff assigned specifically to that unit. Offenders from the unit work and share recreation time with offenders in general population. There are rules however forbidding offenders from entering units other than their own. Other than increased searches and drug testing, the ISU is basically managed the same as the rest of the units in the institution.

Leclerc Institution, located in the Quebec , officially replaced its ECHO unit (based on a therapeutic community model) with an ISU effective April 1, 2001 . Leclerc is a medium security institution with a dedicated a range for its ISU and has the capacity to accommodate 33 inmates. One cell is reserved for drug testing. The physical layout of the institution is identical to Joyceville Institution. Eight correctional officers were trained and assigned to work in the ISU. A total of three officers provide supervision in the unit during a shift.

Data Sources

The primary sources of data were questionnaires completed by offenders and staff as well as data extracted from the Offender Management System (OMS). OMS is an electronic administrative database maintained by CSC that contains records on all offenders. Other data, such as search and seizure information, were collected from institutional logs for each site. The offender questionnaires were administered by hand or, in some instances, administered face-to-face if the respondent required assistance due to literacy problems. Questionnaires and other specific measures used in this study are presented as appendices (Appendix B through L, inclusive).

Offender Measures

Profile Information

Offender profile information, including age, race, marital status, sentence length, offence type, static and dynamic factors, and severity of substance abuse problems were obtained from the Offender Management System (OMS). All of this information is collected as part of the Offender Intake Assessment (OIA) process completed for offenders admitted to federal prisons in Canada .

Static and Dynamic Factor Ratings

The likelihood of reoffending is measured using both static and dynamic indicators. Static indicators are those that cannot change as a result of treatment programs because they are historical (offence characteristics, offence types, frequency of offending, etc). Dynamic indicators (education, employment, substance abuse problems, etc.) can change through interventions such as treatment programs. The individual static and dynamic indicators are used to determine overall, or global ratings.

The static factor rating is derived from 137 indicator items that are reviewed at the time of admission to a prison. The static indicators provide an assessment of criminal history (including current offence) and the overall rating provides CSC with a measure of risk to reoffend. Offenders are classified as ‘high', ‘medium' or ‘low' on the static factor.

The dynamic factor is assessed using 197 indicator items that have been shown to be linked to criminal behaviour. Review of the dynamic factor indicator items results in a global rating for the dynamic factor (low, medium or high) and ratings in seven domains. The seven domains are defined as follows in CSC documentation, or standard operating procedures (Correctional Service Canada, 2003):

• Employment/Education - the value placed on work and the role of work in one's life.
• Marital/Family - the value placed on being with family and the support one derives from them.
• Associates/Social Interaction - the value placed on non-criminal associates and the opportunity for positive social interaction.
• Substance Abuse - the value placed on living without reliance on alcohol and/or drugs.
• Community Functioning - the value placed on having the knowledge and necessary skills for daily living.
• Personal/Emotional Orientation - the value placed on being in control of one's life.
• Attitude - the value placed on living in law-abiding ways.

Severity of Substance Abuse Problem

A standardized measure of severity of substance abuse problems was obtained using the Alcohol Dependence Scale (ADS) (Skinner & Horn, 1984) and the Drug Abuse Screening Test (DAST) (Skinner, 1982). The ADS measures severity of alcohol problems using 25 items scored on a 2-point, 3-point, or 4-point scale. Results from the scale are ranked on a 5 point scale from 'no problem' to 'severe problem'. Reported reliability for the scale is .92. The DAST measures severity of drug problems using 20 items answered in a yes/no format. Reported reliability for the scale is .92.

These scales are used to measure the overall severity of substance abuse problems, not necessarily associated with the criminal behaviour. This is somewhat different from the assessment of substance abuse problems obtained at the time of offender intake which specifically measures the substance abuse problem in relation to criminal behaviour. The latter results in a higher number of offenders being identified as having a problem.

Offender Substance Abuse Information

Information regarding the offenders' past experiences with substance abuse as well as their knowledge of the effects of the use and abuse of alcohol and drugs was obtained through an intake interview and the administration of questionnaires covering such areas as consequences of alcohol and drug use, the physiological affects of substance abuse and relapse attitudes.

Intake Interview

An Offender Intake Interview was conducted to obtain specific information about an offender's self-reported experience with alcohol and drugs, problems associated with substance use, involvement with substances during incarceration and treatment history. The interview was a 26-item, self-administered questionnaire comprised of questions such as: "Was substance use related to your current offence?" (respondent asked to check either yes or no) and "When in the community what was your reason for using substances?" (respondent asked to check all applicable choices from a list provided). The complete questionnaire can be found in Appendix B.

Offender Substance Abuse Knowledge

Four scales were used to assess offender's attitude toward, and knowledge of, substance use and relapse prevention. Appendix C contains the complete versions of the four questionnaires.

The scale, created by Gunn, Orenstein, Iverson & Mullen (1995), is comprised of 20 statements, each requiring a true or false response. This scale is also used in the Correctional Services Canada's Offender Substance Abuse Pre-Release Program. These explore the respondent's knowledge of the physiological and psychological effects of alcohol use. Examples of the type of statements are: "A person's alcohol tolerance increases with regular drinking" and "The physical withdrawal from heroin is more dangerous than is the withdrawal from alcohol". With respect to this scale's reliability using offender norms, Milson, Weeks, & Lightfoot (1995) reported a Cronbach alpha of .54. For the present study the alpha is .48.

The scale, modified by Millson, Weekes, & Lightfoot (1995) from Gunn, Orenstein, Iverson & Mullen's (1983) 30-item scale, comprises 27 statements requiring a true or false response. The questionnaire investigates the respondent's knowledge of the physiological, psychological, legal and medical consequences of drug use. Examples of the types of statements are: "Moderate marijuana use causes permanent physical damage" and "Physical dependence on barbiturates is as severe as a user's dependence on heroin". The authors reported a Cronbach alpha of .49. For the present study the alpha is .55.

The 14-item scale, modified by Millson, Weekes, & Lightfoot (1995) from Gunn, Orenstein, Iverson & Mullen's (1983) 20-item scale, uses a five point Likert scale that requires respondents to rate their level of agreement for each statement from ‘strongly disagree' to ‘strongly agree'. The scale examines the respondent's perception of how people may be affected physiologically and psychologically by alcohol or drugs not prescribed by a doctor. An example of a statement is: "Using large amounts of drugs or alcohol can damage relationships within a family". The authors reported a Cronbach alpha of .62. For the present study the alpha is .84.

This scale, modified by Millson, Weekes, & Lightfoot (1995) from Lightfoot and Barker's (1989) 47-item Employment Questionnaire, measures an offender's perception of attitudes toward drinking and drug use, coping without alcohol/drugs in various situations, activities promoting healthy lifestyles, and general relapse knowledge. A five point Likert scale, ranging from ‘strongly disagree' to ‘strongly agree', is used to rate respondent's level of agreement for each of the 20 items listed. Examples of items that were rated by the offender are: "When there is a celebration at work, everyone is expected to drink in order to have a good time" and "If someone relapses, other people will think they are a worthless person". The authors reported a Cronbach alpha of .86. The present study found an alpha of .85 for this scale.

Offender Motivation, Expectations, and Response Reliability

Several measures were used to examine an offender's motivation to change behaviour and their level of expectations for the ISU.

Readiness for Change

Prochaska & DiClemente (1992) developed this multi-dimensional assessment questionnaire, used to measure an offender's readiness for change, is comprised of four sub-scales of 8 items each (Precontemplation, Contemplation, Action and Maintenance) for 32 items overall. Using a five point Likert scale (‘strongly agree' to ‘strongly disagree'), this assessment measure also examines an offender's feelings toward therapy and problem resolution. Examples of the items contained in each of the URICA's four stages of change are as follows:

Stage 1 : Precontemplation

"As far as I am concerned, I don't have any problems that need changing";

Stage 2 : Contemplation

"I'm hoping that this place will help me to better understand myself";

Stage 3 : Action

"Anyone can talk about change: I'm actually doing something about it";

Stage 4 : Maintenance

"I'm here to prevent myself from having a relapse of my problem".

The complete scale can be found in Appendix D.

Offender Expectation Questionnaire

Designed specifically for this study, the Offender Expectation Questionnaire consists of 46 items that are rated on a five-point Likert scale ranging from ‘strongly agree' to ‘strongly disagree'. The complete scale is presented in Appendix E. The questionnaire is composed of four scales and 12 items that are analyzed individually. The four scales are:

1. ISU effectiveness measures the offender's perceived effectiveness of the ISU (14 items). Examples of items in this scale include: "I think increased searches in the Intensive Support Unit will help decrease the amount of drugs available" and "I think increased urinalysis in the unit will contribute to a decrease in the number of offenders using drugs". The coefficient alpha was .73 and the scores on this scale can range from 14 to 70.
2. ISU benefits measures the offender's perception of expected benefits of the ISU (12 items). "I think I will have a better chance of getting early release because I live in an Intensive Support Unit" and "I moved to the Intensive Support Unit because I think there will be more freedom on the living unit" are examples of items in this scale. The coefficient alpha was .86 and the scores on this scale can range from 12 to 60.
3. ISU support measures the offender's belief regarding the expected support there will be within the ISU (4 items). "I think staff will be more supportive in the Intensive Support Units" and "I think the Intensive Support Unit will provide a more positive environment for participating in programs" are examples of items in this scale. The coefficient alpha was .71 and the scores on this scale range from 4 to 20.
4. ISU impact measures the offender's perceived impact of the ISU on the rest of the institution (4 items). An example of an item in this scale includes: "I think the Intensive unit will cause problems in the rest of the institution". The coefficient alpha was .54 and the scores on this scale range from 4 to 20.

The individual items measure perceived alcohol and drug problems (4 items) and reasons for residing in the ISU (8 items).

The Offender Expectation Questionnaire was administered to the offender on admission to the unit and six weeks following admission. Negative items were reversed scored. The complete scale can be found in Appendix E.

Offender Post Evaluation Questionnaire

The Offender Post Evaluation Questionnaire is a 48-item questionnaire, which is completed by the offender upon his discharge from the Intensive Support Unit. This measure contains the same four scales found in the Offender Expectation Questionnaire and 14 individual items as opposed to 12. The questionnaire is intended to evaluate the degree to which the ISU meets the offender's expectations, how successful he was at remaining substance free, whether staff were supportive in their interventions and efforts, and whether there were any specific benefits for being in the unit (e.g., an earlier conditional release). The complete scale can be found in Appendix F.

Balanced Inventory of Desirable Responding

This 38-item questionnaire, which explores an offender's honesty in responding, was modified from the original scale, which was developed by Paulhus (1994), of 40 items. Two of the items, “I have sometimes doubted my ability as a lover" and "I never read sexy books or magazines", were dropped due to an expressed discomfort with these items by offenders in previous research (Gillis, 1994). The questionnaire is split into two 19-item sub-scales. The Impression Management (IM) sub-scale measures attempts to impress others in a socially desirable manner. The Self Deceptive Enhancement (SDE) sub-scale measures an offender's tendency to exaggerate or distort unconsciously self-report information in a positive manner. Both utilize a seven-point Likert scale ranging from ‘not true' to ‘very true' for each of the statements (e.g., "I never cover up my mistakes" and "I always obey laws, even if I'm unlikely to get caught."). For this study the Cronbach alphas were .63 for the SDE and .85 for the IM scale. The complete scale can be found in Appendix G.

ISU Outcomes

The Offender Process Measure Survey was designed to track activities for each offender on a weekly basis. Activities refer to pro-social events such as community support visits, program participation, employment, and peer group meetings. The complete survey form is presented in Appendix H. To determine the level of pro-social activities in which offenders were involved, weekly activity data were analyzed and used to create summary measures that reflected the level of activity over a period of time or simply the incidence of activities. The weekly data were summarized to calculate rates and incident levels appropriate for each variable.

Community Support Visits

Data were collected weekly regarding the number of visits offenders received from various groups or individuals. These data were coded as: family/friends, religious, cultural, special interest organizations such as John Howard Society and lifer groups, and other.

Program Participation

Offender participation in programs is directly related to their correctional plan, which is established by the case management team. In order to address identified criminogenic factors, offenders are encouraged to attend programs offered at the institution. Participation in these programs can positively affect their early release from prison. Data were collected to examine the participation level of offenders and the types of programs they attended.

Employment

Data were gathered and analyzed to determine the level of participation in offender employment opportunities within the institution. Data were recorded by employment status including full-time, part-time, unable to work, unemployed, or other.

Peer Group Meetings

Data were collected and analyzed to determine the level and frequency of weekly offender participation in peer group meetings such as Alcoholics Anonymous, Narcotics Anonymous, site specific support programs, and peer group sessions.

Search and Seizure

Search and seizure data from each pilot site were gathered and organized by unit type from September 2000 to July 2001 inclusive. Institutional search plans, logs, and reports provided the necessary data for analysis. Seizure data include only items that were associated with drug and alcohol use (substance use) such as drugs, pipes, syringes, and brew. Searches were grouped into two categories: a) searches of the person and individual cells and b) common area searches for both the ISU and Non-ISU living areas of the institutions. Examples of personal searches include offender 'pat-downs' or 'frisks' and cell searches. Cell searches could be supplemented with drug detecting dogs and ion scanners. Common area searches include those areas within the living units accessible to all offenders.

As a means of comparing search and seizure data between the ISUs and Non-ISUs, monthly rates of cell searching and seizure are used as a measure of comparison. Cell searching/seizure rates are determined by calculating the proportion of monthly searches and seizures per unit (ISU/Non-ISU) to the total unit population for that same month. Due to variation among each institution in the number of offenders per common area in each unit, the percentage of offenders (ISU/Non-ISU) in the institution is used as a proxy measure for calculating the rate instead of the monthly unit population. Accordingly, monthly rates of common area searches and seizures are determined by calculating the proportion of monthly searches and seizures per unit (ISU/Non-ISU) to the percentage of ISU/Non-ISU offenders in the institution.

Offender Behavioural Indicators Assessment Report

Each month, a 24-item behavioural assessment was completed for each offender in the ISU. The 24 behaviourally based items were rated by a correctional officer working in the unit who was assigned to the offender. The items were grouped into seven domains: interpersonal relationships (6 items), attitude (4 items), behaviour (2 items), effort (2 items), motivation (2 items), responsibility (4 items), and communication skills (4 items). All items are presented in Appendix I, but examples for each domain are presented below:

Interpersonal Relationships : interacts well with other residents in the unit and positive towards group participation in group activities in the unit.

Attitude : positive attitude toward living on the units and positive attitude about substance abuse treatment.

Behaviour : influences other offenders positively in the unit and compliant with rules in the unit.

Effort : keeps cell clean and contributes to the cleanliness of the unit/house.

Motivation : interested in learning new information/skills and motivated to change substance abuse problem.

Responsibility : sets and prioritizes goals and accepts responsibility for mistakes.

Communication Skills : listens carefully and conveys information effectively.

Ratings were provided on a five-point Likert scale, ranging from 'hardly ever' to 'most of the time'. The domain score was calculated by summing across the items for that scale and then converting the score to a percentage. The monthly ratings were then compared across the months the offender remained in the ISU. The number of months for which an offender had a monthly rating depended on the number of months he remained in the unit. Results are presented only for the first five months in the unit as the number of offenders who stayed in the units for more than five months was quite small, making the data less reliable beyond this timeframe.

Institutional Outcomes

Urinalysis

Random Urinalysis

Random urinalysis drug testing is conducted on 5% of the inmate population each month. The probability of being tested is not dependent on program participation, institutional activities or offender characteristics. This testing is conducted under legislative authority (Corrections and Conditional Release Act, 1992) through controlled collection procedures and laboratory analysis of urine samples. Refusals to provide a sample or positive tests result in institutional charges and these events are recorded in OMS. The laboratory analysis is for six major classes of intoxicants including amphetamines, benzodiazepines, cocaine, opiates, phencyclidine, and THC (cannabinoid). Urine samples are screened for substitution of urine and adulterated samples. Refusals occur when an offender is unwilling to provide a urine sample. Not all offenders in the ISU were tested in the random testing program.

Unit Screening Test

The process used to collect and record data on urinalysis screening test results was similar to that followed for cell searches. Urinalysis screening data were collected on a weekly basis from unit staff. For positive test results, the drug type was categorized as cannabinoid, benzodiazepine, barbitituate or opiate. As with the cell search process, staff followed a random approach to selecting candidates for testing. Results of this drug testing were not used to discipline offenders as the process did not meet evidentiary standards. However, failure of a test resulted in the offender being involuntarily discharged from the unit.

Institutional Charges

Institutional charges are recorded in the OMS for behaviours that result in formal charges within the institution . A charge may be classified either as minor or major at the discretion of correctional staff. Major charges are adjudicated by an Independent Chairperson and more formal procedures are followed in their processing. Examples of a major charge are fighting or possession of drugs whereas minor charges can be disobeying an order or possession of unauthorized items.

A recent study by French and Gendreau (2003) has linked, through meta-analysis, the impact of effective interventions to reductions in institutional charges. In addition, their analyses indicated that reductions in charges are also linked to reductions in recidivism after release.

Release and Readmission Outcomes

Release outcome was studied in a number of different ways including type of release, length of time on release and readmissions. For these analyses, all offenders who participated in the ISUs were included, not just those who agreed to be part of the research. The data used in these analyses are taken from the administrative database maintained by the Correctional Service Canada and these data may be used for analyses that provide information to the Service about the management of offenders. Only aggregate data are presented. In addition, identifying information and information that was provided only by research participants are not used in these analyses.

Type of Release

Offenders may be granted a discretionary release (day parole or full parole) before they have served two-thirds of their sentence, or a statutory release when they have served two-thirds of their sentence. Statutory release is mandated in law, and only under exceptional circumstances can offenders be kept in custody beyond two-thirds of their sentence. Statutory release is only available for offenders serving determinate sentences, while all releases for those serving indeterminate sentences (mostly offenders serving life sentences) are discretionary.

Offenders who receive a discretionary release are judged by the National Parole Board to be manageable in the community prior to their statutory release date. If living in the ISU had a positive effect on offender's institutional behaviour (such as program completion, positive assessment, and reduction in negative behaviours) they would be more likely to be granted a discretionary release. Discretionary releases granted earlier in the sentence can have a significant impact on the amount of time an offender spends in custody. An effective intervention can substantially reduce the overall cost of managing an offender's sentence by reducing the time they spend in custody.

Return to Custody

Return to custody for offenders may occur as a result of their conditional release being revoked or, after their sentence has been completed, for a new offence. Only the National Parole Board has the authority to revoke a conditional release. The revocation may occur because the offender has failed to meet the conditions of his or her release, or has committed a new offence.

Failure to meet the conditions of release is sometimes referred to as revocation for a technical violation, and usually occurs because the parole officer and the National Parole Board believe the offender's behaviour is deteriorating and they are becoming a risk to themselves or their community. The most common technical violations include: being unlawfully at large (usually the result of failing to meet with the parole officer or having left a prescribed living area), failing to meet an abstinence condition (continuing to drink alcohol or use drugs), and failing to meet other conditions of release.

Revocation with a new offence occurs when the offender is suspected by the parole officer of having committed a new offence, usually resulting from information supplied by the police. In addition, offenders may be readmitted to prison after their sentence is completed. In this case the reason for the readmission is always the result of committing a new offence.

ISU Staff Measures

The contribution of staff to the success of the ISUs was important. Therefore, staff working in the ISU were provided with additional training to sensitize them to the issues around substance abuse and to encourage efforts to detect and deter drug use behaviour as part of creating a safer living environment. To better understand what expectations staff had of the units and the principles under which they operate, data were collected on their expectations for the ISUs and the results they observed (Impacts) after the units had been in place for a year or more. In addition staff were asked about their knowledge of substance abuse, attitudes regarding offenders and job stress.

Expectation Questionnaire

The Staff Expectation Questionnaire was administered during the initial start-up period of the ISU initiative in an effort to measure what staff thought of the units. The questionnaire contains six subscales and three individual items. Each item was scored on a 5-point scale ranging from strongly agree to strongly disagree. The complete questionnaire is presented in Appendix J. The scales are:

1. Effectiveness of the ISU measured staff perception on the efficacy of the ISU in their institution (14 items). Examples of items in this scale included: "I think that the idea of Intensive Support Units is a good one" and "I think the Intensive Support Unit will help offenders because other offenders will be more supportive". The coefficient alpha was .81 and the scores for this scale can range from 14 to 70.
2. Rehabilitation explored the offender related reasons that staff chose to work in an intensive support unit (3 items). Examples of items in this scale included: "I am working in the Intensive Support Unit because I think offenders may be more motivated to change" and "I am working in the Intensive Support Unit because I have more opportunity to work closely with offenders". The coefficient alphas was .77 and the scores for this scale can range from 3 to 15.
3. Positive Environment had items pertaining to the belief that the ISU would be a good place to work (3 items). An example of the items included in this scale is: "I am working in the ISU because I have an interest in helping offenders with substance abuse problems". The Cronbach coefficient alpha for this scale was .79.
4. Increased Interdiction explored Staff opinions about the increased use and effects of using more interdiction methods such as drug testing and searches (4 items). An example of the items included in this scale is: "I am comfortable using increased interdiction methods (e.g., searches, seizures)". The Cronbach coefficient alpha for this scale was .53.
5. Decisions to Work in the ISU explored other reasons why staff might have chosen to work on an ISU (3 items). An example of the items included in this scale is: "I am working in the ISU because I want to stay away from the inmate drug culture". The Cronbach coefficient alpha for this scale was .31.
6. Impact on Institution dealt with potential problems the ISU might have on the rest of the institution (3 items). An example of the items included in this scale is: "I think the ISU's will increase the drug flow in the rest of the institution". The Cronbach coefficient alpha for this scale was .41.
7. Individual items asked staff how they think the ISUs will affect the rest of the institution.

Impact Survey

The Impact Survey was designed specifically for this study, and consists of 30 items that are rated on a five-point Likert scale ranging from ‘strongly agree' to ‘strongly disagree'. The complete scale is presented in Appendix K. The questionnaire has four scales and two items that are analyzed individually, which are:

1. Effectiveness of the ISU measured staff perception on the efficacy of the ISU in their institution (14 items). Examples of items in this scale included: "I think that the idea of Intensive Support Units is a good one" and "I think the Intensive Support Unit will help offenders because other offenders will be more supportive". The Cronbach coefficient alpha was .81 and the scores for this scale can range from 14 to 70.
2. Positive Environment had items pertaining to the belief that the ISU would be a good place to work (3 items). An example of the items included in this scale is: "I am working in the ISU because I have an interest in helping offenders with substance abuse problems". The Cronbach coefficient alpha for this scale was .79 and the scores for this scale can range from 3 to 15.
3. Substance Free (4 items). This scale explored Staff opinions regarding the effectiveness of the ISU's in being 'substance free' and the potential reasons for this success. Items in this scale were: "I do not think the Intensive Support Units were anymore substance free than other units" and "I think the Intensive Support Unit was kept substance free because of peer support". The Cronbach coefficient alpha for this scale was .71.
4. Positive Impact (4 items) explores the present and future benefits the ISU will have on offenders. Items for this scale included: "I think that the experience of the ISU will help offenders after release into the community" and "I think that the ISU helped offenders follow their correctional plan". The Cronbach coefficient alpha was .91.

Staff Substance Abuse Knowledge

The following three substance abuse knowledge questionnaires are the same scales used for the offender knowledge questionnaire. Since these scales have previously been described in detail, only the reliability coefficient alphas for the ISU Staff results for this study are included in this section. The complete scales are presented in Appendix L.

Consequences of Alcohol Use

The Cronbach coefficient alpha reported in this study was 0.42.

Consequences of Drug Use

The Cronbach coefficient alpha reported in this study was 0.86.

How Much Do They Matter?

The Cronbach coefficient alpha reported in this study was 0.85.

Job Orientation

The scales outlined in this section were used to measure various job aspects of the ISU staff. Results from these were compared with data gathered in the Correctional Service of Canada (CSC) All Staff Survey (1996). Appendix M includes the complete scales used in the 1996 survey.

Professional Orientation Sub-Scale

This 17-item scale, created by Klofas and Toch (1982), was modified for this study to create 3 subscales. The scales created were: Empathy towards Offenders (7 items, Cronbach alpha = .66), Punitiveness (4 items, Cronbach alpha = .79), and Rehabilitation (3 items, Cronbach alpha = .74). A four point rating of statements from ‘strongly agree' to ‘strongly disagree', examined a staff member's perception of offenders, rehabilitation and punishment. Examples of these items are: "The way to get respect from offenders is to take an interest in them", "A military regime is the best way of running a prison" and " Rehabilitation programs are a waste of time and money". The complete scale can be found in Appendix L.

Total Job Satisfaction Sub-Scale

This 16-item scale created by Warr, Cook and Wall (1979), is a seven-point rating of statements from ‘extremely dissatisfied' to ‘extremely satisfied' (Likert scale), which examines a staff member's perception of job satisfaction, working conditions, and other features of their present job. Examples of types of items to be rated are: "Your hours of work" and "Now, taking everything into consideration, how do you feel about your job as a whole"? The test-retest reliability coefficient for this sub-scale is .63. The complete scale can be found in Appendix L.

Procedure

Selection and Participation Process

Each offender followed a similar selection and participation process, which are described in terms of pre-admission, post-admission and discharge.

Pre-Admission

Offenders interested in residing in the ISU submitted an application to a unit admission committee, which was established for each pilot site. Selection to the unit was based on three main criteria: 1) no drug or alcohol-related involvement for a minimum of 30 days prior to admission; 2) absence of preventive security information, including illegal or disruptive behaviour, which could result in problems if placed in the unit; and 3) a willingness by the offender to follow his Correctional Plan. If the offender met the admission criteria, he was asked to sign a Unit Consent Agreement which signified his willingness to submit to a more intrusive regime of interaction activities, such as urinalysis screening tests and increased searching, not ordinarily authorized (Corrections and Conditional Release Act, 1992; Corrections and Conditional Release Regulations, 1992 ) . If the offender agreed, he was admitted to the unit and, following a briefing on the research study and where there was interest to participate, offenders signed a research consent form.

Post-Admission

Within one week following admission to the unit, offenders completed the Offender Expectation Questionnaire , the first in a series of measures . During the first three weeks, offenders also completed an intake interview and a battery of seven measures. These measures, described in the method section, were assessment scales examining an offender's knowledge of alcohol and drug use, readiness for change, and ability to affect one's own behaviour (self-efficacy).

Six weeks following admission to the unit, each offender completed the Offender Expectation Questionnaire a second time to determine if their expectations changed or remained the same.

To determine if the ISU had an impact on offender behaviour, an offender's participation in programs, employment, and pro-social activities both within the unit and outside were monitored and recorded on a weekly basis. Each offender was also assigned a staff member, who reviewed the inmate's behaviour on a monthly basis and a rating was recorded across seven behavioural domains. In addition, any involvement that an offender had in relation to misconduct, use of substances, or discharge from the unit, were recorded on a weekly basis.

Discharge

Offenders were discharged from the unit either voluntarily (e.g. granted discretionary release or moved to another unit or prison for personal reasons) or involuntarily (e.g. tested positive for drugs or disruptive behaviour). The specific reason for each discharge was recorded and offenders were asked to complete the Offender Post-Evaluation Questionnaire. Offenders leaving voluntarily were generally more willing to complete the questionnaire than those discharged involuntarily.

Project Management

The research study was project managed and these functions are described in three main areas: data collection, data coding and data analyses.

Data Collection

On-site research assistants were hired for each site. Commencement dates for the research assistants were different, but by September 1, 2000 , a research assistant was assigned to collect data at each pilot site, except Leclerc Institution where a later start-up occurred ( April 1, 2001 ). Each research assistant, working 2 ½ days per week, was given an orientation to the research project, a full explanation of the offender and staff measures, and the research protocol to be followed to collect the data. Given the scope of work and requirement for consistent data recording, a weekly teleconference was instituted in January 2001. This procedure reduced significantly the amount of time spent resolving data quality with each site. In addition, due to a considerable workload, research assistants commenced working full-time (5 days a week) on January 1, 2001 .

Research assistants worked closely with unit staff to obtain offender admission and discharge dates, reason for discharge (where applicable), weekly offender and staff activities, and monthly offender behavioural ratings. In terms of research protocol, each research assistant established a schedule of testing and interviews following the admission of an offender to the ISU. In addition, the research assistants collected weekly process measures data (e.g. drug test results and activities information) and monthly behavioural indicators assessment reports completed by ISU staff.

Research assistants also administered a series of questionnaires to staff (Staff Expectations and Staff Knowledge), who were either permanently assigned to the unit or worked a substantial portion of their shifts in the unit. Staff were primarily correctional officers, parole officers, correctional supervisors, and unit managers.

Data Coding

Each pilot site was provided with an electronic data entry workbook, and coding instructions, designed specifically for this project. Research assistants entered all the collected data into the electronic workbook on a weekly basis. The site data was merged into a master data workbook bi-weekly and was quality controlled by a Project Manager to identify any data quality problems.

Data Analyses

Data from the pilot sites were placed into the Statistical Analysis System (SAS) Version 8.01 (SAS, 1999) database. Analyses were conducted to identify anomalies and inconsistencies with the data. Where data problems, such as duplicate records or missed values, were present, research assistants were instructed to review these data again and make appropriate corrections. Data extracted from the OMS were also placed in the SAS database and analyzed for quality. The principal statistical analyses consisted of frequency distributions and tests of statistical significance using chi-square. F-tests (PROC GLM) were also used when comparing continuous variables (i.e.: age, scale scores). For the survival analysis the SAS procedure Lifetest was used.

RESULTS

The results are presented in eight sections. The first four sections report on the characteristics of offenders in the ISUs, their history and experience with substance abuse, their motivation to change and knowledge of substance abuse, and their expectation for the ISUs. The next three sections present information on the effectiveness of the unit in reducing alcohol and drug availability and whether there was an impact on offender behaviour while in the ISU and on their release and readmission outcomes. The final section presents information on ISU staff expectations, their knowledge of substance abuse, job orientation, and the impact the ISU had on offenders and the rest of the institution.

Offender Profile

To determine if the ISU participants were similar or different from the general offender population, a comparison group was constructed including all male offenders in custody in minimum and medium security institutions on January 1, 2001 .

Age, Marital Status, and Race

The ISU participants were slightly younger than the comparison group at 37 years of age versus 39 years of age ( F (1, 3,736) = 8.74, p <.01). With respect to marital status, almost half of the ISU group, 49%, were single (47% in comparison group), 41%, were either married or common-law (40% in comparison group), and the remaining 10% were previously married (13% in comparison group).

Most of the offenders in the ISU group were Caucasian, 79%, 12% were Aboriginal, and 9% were identified as having other racial origins. These results are similar to what was found for the comparison group (74% Caucasian, 16% Aboriginal, 10% other racial groups). However, there is a slight under-representation of Aboriginal offenders in the ISU that cannot be accounted by regional variations in the sites.

Sentence Length

Offenders in the ISU were more likely to be serving shorter sentences than offenders in the comparison group as shown in Table 2. For example, 43% of offenders in the ISU were serving sentences of less than four years while 32% of offenders in the comparison group were serving sentences of this length. In addition, 17% of ISU offenders compared to 23% of the comparison group were serving life sentences.

Table 2 : Sentence Length for ISU and Comparison Group

Offence Type

Almost half of the ISU offenders (47%) were admitted to prison for violent offences, a slightly lower rate than in the comparison group (51%). Violent offences include homicide, homicide-related offences, robbery, sexual offences, assault, etc. Both groups had an average of five admission offences per offender.

Static and Dynamic Factors at Admission

Overall, the ISU participants score lower on static factors (criminal history) than the comparison group as shown in Table 3. Specifically, the ISU participants were more likely to be rated medium (48%) based on static factors and less likely to be rated high (41%) in relation to the comparison group (medium, 38%; high, 52%). However, there was little difference between the groups for the dynamic factor rating (criminognenic needs), with the majority of ISU participants rated as high on the dynamic factor (55%).

Table 3 : Static and Dynamic Factor Ratings for ISU and Comparison Group

The dynamic factor rating is based on the assessment of seven dynamic domains. The ISU participants and the comparison group were relatively similar in the dynamic need areas identified, except in the substance abuse domain where 79% of the ISU group were identified as having a problem compared to 70% in the comparison group. In addition to substance abuse, Personal and Emotional orientation was identified as a problem area for 90% or more offenders in both groups. This may indicate that the additional support offered by the ISU in this area could be an effective intervention for offenders. Detailed comparisons are presented in Table 4.

Table 4 : Ratings for Dynamic Factor Domains for ISU and Comparison Group

Severity of Substance Abuse Problem

Results from the Alcohol Dependence Scale (ADS) and Drug Abuse Screening Test (DAST) indicate that 54% of offenders in the ISU had a substance problem that was either of moderate severity or greater. This rate is slightly higher than for the comparison group in which 47% of offenders were rated as having a moderate, or greater substance abuse problem. Results also indicated that 45% of the ISU participants had a moderate or higher drug abuse problem, while a much smaller percentage, 19%, had a moderate or higher alcohol abuse problem. Results are presented in Table 5.

Table 5 : Percentage of Offenders that Scored as Moderate, Substantial or Severe on the ADS, DAST and a Combination of Both Scales

Substance Abuse History

This section presents information specific to the history of substance use by ISU offenders as well as problems they encountered, including criminal behaviour, as a result of substance abuse. Also presented is treatment intervention and outcome information.

Substance Use and Abuse Background

Drug and Alcohol Problems

Information on drug and alcohol problems was available from a number of sources including the Offender Intake Assessment that is completed on admission to prison, specific questions in the ISU intake interviews and through responses to questions that were only applicable to those who had an addiction problem. Combining all of these sources, the trend is that about 80% of the ISU offenders had a substance abuse problem, leaving about 20% identified as not having a problem. Details on the different sources of information are presented below.

At the time of admission to prison, offenders are assessed on seven dynamic factors that have been shown to be linked with criminal offending (see the Method section and results presented earlier in this section for details). Substance abuse is assessed as one of the dynamic factors and 79% of ISU offenders were identified as having a substance abuse problem.

On admission to the ISU, offenders were asked if they had an alcohol or drug problem and 55% reported they did. In addition, 58% reported that others thought they had a drug or alcohol problem. This suggests a difference of between 24% and 20% between their assessed level and the self-reported level at admission to the ISU. Interestingly, the difference between level of substance abuse problem self-reported on admission to the ISU and through the dynamic factor assessment was greatest for minimum-security offenders: these offenders have the greatest need to reduce or minimize their substance abuse problem as they are closest to release from prison.

A series of questions asked if moving to the ISU was associated with the offenders' substance abuse problem. Offenders could indicate that they did not have a problem when responding to these questions. The responses provide an estimate of the percentage of offenders who moved to the ISU, but who did not have a drug or alcohol problem. In general, 19% of the offenders indicated they did not have a drug or alcohol problem, leaving 81% identified in this way as having a substance abuse problem. This result is consistent with the percentage of participants (79%) who were identified at admission to federal prison as having a substance abuse. It would appear that when asked directly about their substance abuse problems, offenders tend to minimize the problem, but when asked indirectly, or through the use of multiple sources of information, a more accurate estimate is obtained.

Early Alcohol and Drug Use

Through the Offender Intake Questionnaire, additional information was obtained regarding offender experiences with alcohol and drug use. ISU offenders reported to have tried alcohol at a younger average age than drugs (13 years versus 15 years). Problematic behaviour with alcohol was reported to have emerged at age 20 for alcohol and age 21 for drugs.

Reasons for Use and Criminal Behaviour

The most frequently reported reason for using both drugs and alcohol in the community was to relieve stress and to relax (63% and 67% respectively). Interestingly, addiction and dependence was the second most noted reason for using drugs (65%) but only 41% reported using alcohol due to addiction/dependence. Approximately 67% of offenders indicated an association between substance use and their current offence.

Drug and Alcohol Use Prior to Offence

In the month prior to their most recent offence, 39% of offenders reported abusing drugs more than four days a week and 17% indicated they had abused alcohol for the same number of days.

During the 12- month period prior to their current offence, 68% of offenders reported an addiction to drugs compared to 48% for alcohol. Over one-third of offenders reported that alcohol (37%) and drug (36%) use caused violent and aggressive behaviour. The most frequently endorsed alcohol related problem was conflict with family/friends (45%). For drug related problems, 62% of respondents reported financial problems, 59% experienced conflict with family/friends, and 56% stated drug use interfered with school and work. In addition, offenders consistently endorsed problems such as conflict with strangers, emotional difficulties, and interference with school and work for both alcohol and drugs.

Use During Incarceration

The consumption of drugs or alcohol at least once during any prior incarceration was reported by 47% of offenders and 37% reported consumption at least once during their current incarceration. This is consistent with the 38% of respondents reporting drug use in the 1995 National Inmate Survey (Robinson & Mirabelli, 1996). For those offenders who acknowledged using substances, cannabinoids was the substance most frequently used followed by alcohol or brew as it is commonly referred to in prison.

Of the respondents who indicated using substances while incarcerated, the two main reasons for consuming either alcohol or drugs included peer pressure/ conformity to social norm and managing negative emotions about the environment. Despite these findings, three-quarters of offenders rated their confidence level in dealing with their problem as 'moderate to high'.

Treatment

Three-quarters (77%) of ISU offenders having an identified substance abuse problem (N = 178) were involved in some form of treatment or intervention in the past. The average number of treatment attempts was two. In terms of referral sources, approximately 42% of respondents indicated that they were referred to treatment by an agency of the criminal justice system, followed by self-referred at 32%, and family or friends at 9%. Of those respondents who participated in a substance abuse treatment, 80% reported completing the program. The Offender Substance Abuse Pre-Release Program (OSAPP), CSC's core substance abuse program, was reported as the most recent intervention type for almost one-third of respondents (30%). Over half of respondents (57%) reported that they were addressing either a drug or poly-drug problem and 40% indicated they were addressing an alcohol problem.

With respect to future goals of alcohol and drug use, 66% of offenders reported that they plan to abstain from alcohol after their period of supervision ends, whereas 84% indicated they would like to remain drug free. Twenty percent of offenders reported moderation of alcohol use as a future goal following the expiration of their sentence and 7% indicated drug use moderation. Fourteen percent of offenders indicated that did not envision a change in alcohol use after sentence completion and 9% responded similarly for drug use .

Substance Abuse Knowledge

A set of four scales was used to measure offenders' knowledge of the consequences of substance use and challenges in coping with a substance abuse problem. The scales are defined more completely in the Methods section. Results from the ISU offenders are compared to those obtained from offenders who participated in CSC's Offender Substance Abuse Pre-release Program (OSAPP) between 1992 and 1997 (T3 Associates, 1999). Results for ISU offenders were converted to a percentage for each of the scales to be consistent with the methods used for the OSAPP comparison group.

Overall, the ISU offenders were as knowledgeable as offenders entering the OSAPP program, as shown in Table 6. Comparisons were also made across the minimum and medium security levels and no differences were found.

Table 6 : Level of Knowledge About Consequences of Substance Use and Coping Challenges for ISU and OSAPP Participants

Offender Motivation, Expectations, and Response Reliability

This section presents results on offenders' readiness to change and their expectations of the ISU, including reasons for moving to the ISU, perceived effectiveness, expected benefits, expected support, and perceived impact of the ISU on the rest of the institution. Reliability of the self-reported information is discussed in the last section.

Readiness for Change

As explained earlier, the University of Rhode Island Change Assessment (URICA) was used to estimate which stage of change offenders were at for addressing their substance use and other problems at time of entry to the ISU. There are four stages of change (pre-contemplation, contemplation, action and maintenance) and an individual may be rated, based on the results of the URICA, along this continuum. Offenders in both minimum and medium security prisons were between the contemplation stage (openness to self-discovery) and the action stage (willingness to do something concrete) upon entry into the ISU. There were no differences between offender groups on this measure. The results for the URICA are presented below in Table 7.

Table 7 : Offender Readiness for Change scores

Reasons for Moving to the ISU

Overall, most offenders who had a substance abuse problem, moved to the ISU to address their drug or alcohol problem. Specifically, 56% indicated they moved to the ISU because of their substance abuse problem, 69% indicated they moved to get help with their substance abuse problem, and 63% indicated they moved to "get clean". Interestingly, almost four-fifths, moved to help themselves "stay clean", which is one of the goals of the ISU. The responses to all reasons for moving to the ISU are presented in Table 8.

Table 8 : Reasons for Moving Into an ISU

In addition, 67% of all ISU participants indicated they moved to stay away from the drug culture. Only a very small percentage indicated they moved to get away from muscling and to avoid pressures on their friends and family coming for visits.

While it was anticipated that recommendations from the case management team would be a major influence for ISU participation, only 25% of offenders indicated they moved as a result of a recommendation from the team.

Offender Expectations

Offender expectations about the ISU were measured using four scales; perceived effectiveness, expected benefits, expected ISU support, and impact of the ISU on the institution. Recall that the items are scored on a five-point scale (1 to 5) ranging from ‘strongly agree' to ‘strongly disagree' with the scale midpoint of 3 being neutral. Responses were averaged to produce a score along the five-point scale. The scales were administered at admission to the units, six weeks after admission and again on discharge from the units. Results are presented in Table 9.

The number of cases completing the scales declined across each time period. The loss of cases at six weeks was largely due to administrative problems at two sites and was not the result of discharges from the units. The reduction in the number of offenders completing the Discharge version occurs because many of the study participants were still in the units when the data collection ended and those who left the units involuntarily were reluctant to complete the discharge version of the questionnaire.

Table 9 : Expectation Scales Comparisons at Admission, Six Weeks, and Upon Discharge

Perceived Effectiveness

The mean score for the perceived effectiveness of the ISUs was 3.7, suggesting that the offenders expected the ISU to be effective in achieving its goals through the various controls and procedures put in place.

To determine if those identified as having a substance abuse problem at the time they were admitted responded differently from offenders without a substance abuse problem, a comparison was made between these two groups. There was no statistically reliable difference. In addition, a comparison was made to determine if the institutional security level of the offenders impacted their expectations of effectiveness; again, no differences were found. Examples of specific items producing high levels of agreement at admission include:

• I think Intensive Support Units are a good idea (90% agreed).
• I think the Intensive Support Unit will help other offenders (82% agreed).
• I think increased use of urinalysis in the unit will contribute to a decrease in the number of offenders using drugs (71% agreed).

The perceived effectiveness scale was completed six weeks after admission to the unit and at discharge. The results are presented in Table 9. Results indicate that perceived effectiveness did not change over the three time periods, suggesting that expectations were not inflated when offenders moved to the unit and that what they expected to occur did, in fact, occur.

Expected Benefits

Offenders were asked 12 questions about the benefits they expected to obtain while living in the ISU as well as its impact on their release from prison. The ISUs were not created to provide direct benefits to offenders (no special privileges, improved living conditions, or impact on release decisions).

Overall, the results indicated an average scale score of 3.0 for benefits. That is, they did not expect any direct benefits, but also did not expect any negative impacts of being in the ISU.

Comparisons between offenders with and without a substance abuse problem (identified at admission), and across security levels, indicated no statistically reliable differences. Examples of expected benefit items that showed high levels of endorsement included:

• I think the Intensive Support Units will help prepare me for release (70% agreed).
• I think the living conditions will be better in the Intensive Support Unit (67% agreed).
• I think that living on the Intensive Support Unit will help me do better on release (67% agreed).

Six weeks after admission to the ISU, the average score for expected benefits dropped from 3.0 to 2.7 ( F (1, 453) = 6.67, p <.001), and the average score remained constant until discharge from the units (2.8). This result suggests that initially offenders may have expected more benefits from being in the ISU than they were to receive, but that shortly after joining the unit their expectations became more realistic.

Expected ISU Support

The mean score for expected support from the ISU was 3.6, above the midpoint of 3, suggesting that, on average, offenders felt they would receive support from staff and inmates while in the ISU. Examples of expected ISU support items that showed high levels of endorsement included:

• I think the Intensive Support Unit will provide a more positive environment for participating in programs (71% agreed).
• I think other offenders will be more supportive in the Intensive Support Unit (58% agreed).

The mean scores for expected ISU support did not change at six weeks or at discharge from the ISU. These results suggest that the support expected shortly after joining the units was available, and that it was consistently provided throughout the period of time in the ISU.

Impact of the ISU on Institution

Based on response to the scale measuring Impact of the ISU on the institution, it appears that offender believed the ISU would have little impact on institutional operations. The mean score on the impact scale was 2, on the 5-point Likert scale. Examples of specific items producing low levels of agreement include:

• I think the Intensive Support Unit will cause problems in the rest of the institution (81% disagreed).
• I think other offenders from other units will give me a hard time for moving to the Intensive Support Unit (72% disagreed).

The mean Impact score did not change across time, at six weeks or at discharge.

Response Reliability

Offenders who participated in the study completed the Balanced Inventory of Desirable Responding (BIDR) questionnaire that measures two dimensions of offender self-report information. It was important to know if offenders were responding in a manner that placed them in a more favourable light with others (impression management) or were exaggerating responses in a positive manner (self-deception enhancement). The mean impression management score was 5.5, which is in the average range based on interpretation guidelines (Paulhus, 1998). For self-deception enhancement, the mean score was 6.6, which is in the acceptable range, but on the high side. This suggests that the ISU offenders may believe they better understand their problem, and over-estimate their ability to deal with the challenges they face.

In summary, ISU offenders were at a stage where they were open to self-discovery and willing to address their problems in a concrete way. For those with a substance abuse problem, most offenders moved to the ISU to address their drug or alcohol problem. Offenders also reported that the ISU was effective in meeting its stated objectives, did not impact negatively on the rest of the institution, and did not provide special privileges or benefits to those residing in the unit. Self-reported offender information was determined to be reliable.

ISU Outcomes

Measures of offender activities and behaviour while in the ISU are presented in this section. Included are ISU search and seizure data that are used to measure the effectiveness of the ISUs in achieving one of its principles, a reduction in alcohol and drug availability in the unit.

Activities (Offender Participation and Association)

During each week that offenders were in the ISU, their case was reviewed to determine the institutional activities in which they were involved. Areas looked at included employment, programming, support groups, and visits.

Results indicated that offenders were employed for the majority of time they were in the unit. There were some slight differences between offenders in minimum and medium institutions with those in minimum institutions more likely to be employed full time.

Offenders in the ISUs have access to all programming that is available to other inmates. While in the ISU, 27% of offenders participated in at least one national correctional program. The most frequent program was the Offender Substance Abuse Pre-release Program (OSAPP), which was part of the activities for 13% of ISU offenders (ISU offenders may have taken this program before joining the unit). Minimum-security offenders were more likely to participate in sex offender programs (10% vs. 0%); while offenders in minimum security institutions were less likely to be attending the Cognitive Skills programs (4% vs. 12%). Interestingly, about 3% of ISU offenders were participating in the Methadone Maintenance Program. These percentages of program participation may reflect the population characteristics of the institutions in the sample, and are not representative of the institutional population of the Correctional Service.

Other programming that focuses on peer support such as Alcoholics Anonymous (AA), Narcotics Anonymous (NA), and local substance abuse support groups were available to offenders. Forty-two percent of offenders in the ISU participated in at least one of these groups. One-third of offenders in minimum-security institutions participated in community based support groups such as AA and NA, but no offenders in medium security institutions participated in these community programs. However, 14% of both medium and minimum-security offenders reported attending AA and NA programs that were institutionally based. Access to community programs is more difficult in medium security institutions because fewer temporary absences are granted from medium security institutions.

Access to the community is available through temporary absences and work release. Half of the offenders in minimum-security institutions reported having escorted temporary absences, 22% reported having a work release and 16% reported having at least one unescorted temporary absence. In medium security, these numbers are substantially lower, with 16% of ISU offenders having had an escorted temporary absence, 1% had a work release, and none had an unescorted temporary absence. These differences across security level reflect operational standards, not the offenders in the ISU.

ISU offenders in medium security institutions were more likely to report having visits with family and friends (46%) whereas minimum security offenders (26%) were less likely to have visitors, probably because they had more opportunities for visits in the community through temporary absences. Offenders in both medium and minimum-security institutions also reported access to cultural (8%) and religious (6%) groups. Offenders at both security levels also had private family visits (minimum, 20%; medium, 9%).

These results suggest that ISU offenders remain active in the institution with high levels of employment, participation in core programs, support programs and access to family and friends through visits and temporary absences. Of note is the high level of access to potential support from family and friends. Appendix S has a complete breakdown of the Activities results by institution security level.

Behavioural Indicators

The behavioural ratings for each of the seven domains (interpersonal relations, attitude, behaviour, effort, motivation, responsibility, and communication skills) are presented in Table 10 across five months. The scores in the table represent the sum of the items for each domain converted to a percentage.

Table 10 : Behavioural Ratings Over Five Months

For each of the behavioural domains offenders were rated relatively high in their first month in the ISU, with ratings of 80% or more. In addition, the ratings remained high and stable during the five months. The areas showing the greatest increase were interpersonal relationships and responsibility, with increases of between 5 and 6 percentage points. Unfortunately, the scale was not as sensitive to changes as had been hoped. However, the results indicate that behavioiur remained consistently positive throughout the period in the ISU.

Search and Seizure Outcomes

Offenders in the ISU agreed to increased searches of their person and their living areas as part of the consent agreement. Search and seizure data were gathered from each pilot site over eleven months from September 2000 to July 2001 inclusive. The rates are calculated as the number of actions (searches or seizures) divided by the number of offenders in the unit for that month. Results are presented for personal searches (search of the person or his cell) and for common area searches along with rates of seizure of substance-related contraband. Common area searches take place in any location outside an offender's cell, yet still in the unit, and can include such places as the unit lounge, kitchenette, laundry room, and storage areas.

Results in Figure 1 indicate that, with the exception of month 2, the rate of personal searching in the ISU is higher than for Non-ISU participants over the eleven months of the study. On average, there were 1.6 personal searches per month in the ISU compared to 1.1 in the remainder of the prison.

Figure 1 : Rate of Cell and Individual Searches by Unit Type

Combined with the higher rate of personal searches in the ISUs, the rate of seizure of substance-related contraband was lower for the ISU during the 11 months of the study, as shown in Figure 2. This result suggests that there were fewer drugs and alcohol in the ISUs. In fact, in 7 of the 11 months there was no substance-related contraband found with personal searches in the ISU.

Figure 2 : Rate of Substance Related Contraband Seizures from Personal Searches by Unit Type.

To avoid being caught with illegal substances, offenders may hide contraband away from their cell. Figure 3 shows the results of both ISU and Non-ISU rates of common area searching. On average, the rate of common area searching is higher in the ISU (0.26), than in the Non-ISU (0.21) areas, but the differences are not large. Results in the figure suggest that some months, common area searches were significantly higher in the ISU than in other areas.

Figure 3 : Rate of Common Area Searching by Unit Type

Figure 4 shows the results of both ISU and Non-ISU rates of substance-related contraband seizures in common areas. Overall, the pattern is for higher rates of seizure in the non-ISU areas, with the ISU areas showing no contraband found for 8 of the 11 months.

Figure 4 : Rate of Common Area Seizure by Unit Type

Institutional Outcomes

Presented in this section are institutional outcomes, which comprise primarily drug testing results and frequency of misconduct for ISU offenders and a matched group of non-ISU offenders. In addition, the results of a third group, those involuntarily discharged from the ISU, are presented in relation to drug testing outcomes. These comparative institutional outcomes are important in determining if the ISU had an impact on offender behaviour. A summary of ISU discharge data is also presented in this section.

Drug Testing

There are two types of drug testing data available for ISU participants, random urinalysis and screening tests. Results for both of these are presented below.

Random Urinalysis

Samples for urinalysis to detect drug use are collected randomly on 5% of the inmate population each month. Many of the offenders in the ISU had been tested through the random urinalysis program so it was possible to obtain their results to determine what effect the ISU had on their drug use behaviour. Results for the ISU offenders were compared to a group matched on offender characteristics and risk. Urinalysis results were also compared for those offenders who were voluntarily discharged from the units and those who were involuntarily discharged.

Results presented in Table 11 indicate that, among offenders who were voluntarily discharged from the ISU, drug use detected by random urinalysis was lower, at 3%, than for the two comparison groups. Refusals to provide a urine sample were also lower for the voluntarily discharged group than for the matched comparison group. For offenders voluntarily discharged, drug use detected by urinalysis before and after the ISU experience was shown to be reduced, but the rate of refusals increased after the ISU experience. By comparison, the matched comparison group showed an increase in both the rate of positive tests and the rate of refusals after the ISU period, suggesting that other factors may have been having a negative impact on drug use in the prison.

Screening Tests

Data obtained from unit field tests reveals that of the 1168 tests administered on 221 offenders (no offenders refused to submit a sample, but some may not have remained in the unit long enough to be tested), 18 tests were positive for drugs. This equates to 1.5% of tests being positive. Cannabinoids accounted for 13 of the 18 drugs found within the sample and the remaining two drugs detected were opiates (four cases) and psychedelics (one case).

Table 11 : Percentage of Positive Urinalysis Tests ^a Before, During, and After ISU

Institutional Charges

Overall there were fewer misconducts resulting in minor or major charges in the ISU as compared to the Matched group, as shown in Table 12. There were 282 minor charges for ISU participants and 446 minor charges for the Matched group, representing 164 more charges for the Matched comparison group. In terms of individuals, 125 were charged in the ISU, while 151 were charged in the Matched group. The Matched group was also 50% more likely to receive a major charge than the ISU group (205 vs. 137), but the number of individuals charged was almost the same (79 vs. 80). These results suggest that the ISUs had less disruptive behaviours than the comparison group. However, part of the explanation for this may be that problem offenders were removed from the ISU as part of the contract they signed when becoming ISU participants.

Table 12 : Number of Institutional Charges Received While in the ISU and of Offenders Found Guilty

The number of each type of charge also varied between the ISU and the Comparison groups. In terms of drug and drug related charges, the ISU participants and the Matched group had almost the same number of charges (71 vs. 75). However, the ISU group had 56% fewer charges for violence and aggression, 48% fewer charges for having contraband and unauthorized items and 28% fewer charges for disobeying an order. Overall, the institutional charge data indicate that there was less negative behaviour in the ISU than was observed in the Matched group. An analysis was conducted to ensure that the observed differences did not occur simply because the problem cases were removed from the ISUs . In fact, most offenders receiving charges were able to remain in the ISU for some time after. Results for the charge information are in Table 13.

Table 13 : Number of Institutional Charges (Major and Minor) by Type of Charge Received While in the ISU

Offender Discharge Data

At the time the data collection for the study ended, 53% of the 246 offenders, who had started in the study between September 2000 to July 2001, had left an ISU. Of these, 37% were granted a discretionary release (temporary release, day parole, or full parole), 3% received statutory release, 27% were transferred voluntarily to an institution of similar or lower security, and 33% were transferred to higher security or removed from the ISU involuntarily. Among the 33% of those transferred or removed involuntarily, 25% were removed for substance abuse infractions and the balance (8%) were removed for violence or other serious rule violations. Approximately, one-quarter of those discharged returned to the unit.

Interestingly, those involuntarily discharged from the ISU remained in the ISUs for an average of 7 weeks while those who were released or left voluntarily remained in the ISU for an average of 10 weeks ( F (1,131) = 10.3, p <.001).

Release and Readmission Outcomes

An examination of release and readmission outcomes is presented in this section. To determine if the ISU had an impact on offender behaviour for those released into the community, a comparison of three groups (voluntarily discharged, involuntarily discharged and matched group) was made and outcome data are presented for several types of release.

Type of Release

Almost 2/3 (62%) of the offenders who left the ISUs voluntarily received a discretionary release to the community from the National Parole Board after their time in the ISU (not all releases occurred immediately after their departure from the ISU). By comparison, only one-third (36%) of those who were involuntarily discharged and almost 50% of the matched comparison group received a discretionary release. Offenders who left the ISU voluntarily represented a reduced risk to the National Parole Board, possibly as a result of their behaviour while in the unit, and because it was evident they were working on addressing their substance abuse problem. Results for these data are presented in Table 14.

In addition to looking at type of release, an analysis was conducted to estimate the number of days saved as a result of a discretionary release occurring before a statutory release. Interventions that result in offenders being released, on average, earlier in their sentence, are more effective and provide a cost savings for the correctional system. Offenders who participated in the ISU and who were voluntarily discharged from the ISU were released to the community, on average, 477 days before their statutory release date. By comparison, the Matched group were released 443 days before their statutory release date and the Involuntarily discharged group were released 418 days before their statutory release date. The ISU intervention resulted in a release to the community that occurred between 34 and 59 days earlier than for the comparison groups.

Table 14 : Type of Release for ISU Participants and the Matched Group

Return to Custody

Offenders were followed for the first twelve months they were in the community to determine their rate of readmission as shown in Table 15. Overall, 25% of the ISU participants who left the units voluntarily were readmitted within 12 months of their release from an institution. This compares favourably to the offenders who were discharged involuntarily from the ISU who had 47% return rate within 12 months and the Matched comparison group that had a readmission rate of 39%. The results suggest that the positive impact of ISU was maintained after release.

Table 15 : Returned to Custody with 12-Month Follow-Up

More detail on the reason for being returned to custody is presented in Table 16. The majority of offenders, in all groups, were returned to custody because they violated a condition of their release (revocation without offence) with the ISU participants voluntarily discharged having the lowest revocation rate at 16% compared to the other groups. In terms of a return to custody for committing or being suspected of committing a new offence (revocation with an offence, revocation with outstanding charges and admission with new offence) the differences are greater. While 9% of the ISU participants who left voluntarily were readmitted with a new offence, or suspicion of a new offence, the rate was more than double for the involuntarily discharged offenders (21%) and close to double for the matched comparison group (17%). The ISU participants voluntarily discharged were less likely to be readmitted and were less likely to be readmitted in association with a new offence.

Table 16 : Type of Readmission During 12-Month Follow-Up for Voluntarily and Involuntarily Discharged ISU Group and the Matched Comparison Group

Another measure of program effectiveness is how long offenders are able to remain in the community after their release. Every additional day in the community indicates greater benefit from participation in the program. In addition, each additional day in the community represents a cost saving to the correctional service since community supervision costs less than administering the sentence in the institution. ISU participants who were voluntarily discharged remained in the community for an average of 316 days, whereas those in the matched comparison group remained in the community for only 288 days, a difference of almost one month. The ISU participants who were involuntarily discharged remained in the community for only an average of 265 days, almost two months less than the other groups. Table 17 contains the number of days on release by discharge type and matched group results.

Table 17 : Days Out on Release in 12 Month Time Period (Offenders Not Returned Assigned the Full 12 Months)

Survival Analysis

Survival analysis is a statistical procedure that provides a more graphic view of when offenders return to custody. With survival analysis, the number of cases available at each point in time (monthly in this study) is used to calculate the proportion of people remaining in the community.

Survival analysis has benefits over other types of analyses in that it makes the most effective use of all of the data available. In addition, it is possible to see changes in the rate at which offenders return to custody and compare these changes in rates across the different groups. Results of the survival analysis are presented in Figure 5.

The survival analysis shows clear differences in the rates of return to custody across the three groups, with the ISU participants who were voluntarily discharged remaining in the community the longest and at the highest rate. Statistical analyses indicate that the overall results are statistically reliable (Log-rank test: x 2 (2) = 23.27, p <.0001) and analysis of the differences between pairs of survival curves indicated they are also statistically reliable.

Figure 5: Survival Analysis Curve for a 2-Year Follow-Up of Readmission for Voluntary Dischargees, Involuntary Dischargees and Matched Group

For both the Matched comparison group and the involuntarily discharged group the curve is steeper in the first nine months after release indicating a higher rate of return to custody when compared to the ISU Voluntarily discharged group. In addition, the curves beyond this point, while levelled out, remain different and with increasing loss of cases, that is, increased number of offenders returned to custody.

ISU Staff

Staff perceptions of the ISUs were collected using two questionnaires, the Expectation questionnaire administered shortly after the ISU study was started and the Impact questionnaire which was administered at the end of the study, approximately 11 months later. The Expectation questionnaire provides an indication of what staff thought would happen while the Impact questionnaire provides a measure of what they thought actually happened. The questionnaires were used to construct a series of subscales, some of which could be compared across the questionnaires, and some of which could not be compared. Staff knowledge of substance abuse and job orientation along several dimensions including empathy toward offenders, punitiveness, rehabilitation, and job satisfaction are presented.

Staff Expectation and Impact Scales

Overall, staff assigned to the ISUs indicated that they believed the units would be effective and this perception remained high at the end of the study period (Impact questionnaire). The average Effectiveness of ISU score, on the 5-point rating scales, across 14 items, was 3.7 shortly after the units were started and stayed the same at the end of the study. When the ISUs were started, staff were uncertain about the rehabilitation potential of the units (scoring 3.1, with 3 being the neutral point), but this increased to 3.6 after the units had been operational for a number of months. Results for the two staff scales are summarized in Table 18.

Table 18 : Staff Expectation and Impact scales

Individual Items

Two questions were asked in the Expectation questionnaire and the Impact questionnaire specifically about the impact of the ISU; one asked staff to indicate their level of agreement with the statement, "I think the ISUs cause more disruption in the rest of the institution" and the other asked "I think the ISU's help offenders because it provides a more positive environment for participation in programs". Overall staff believed at the outset and after the program had been in place for a number of months that the ISU would have minimal impact on the remainder of the institution (M = 2, SD = 0.8). In addition, staff believed at the outset that the ISU would have a positive effect on offenders because of the environment they were creating (M = 4.1 SD = .0.8); this view was still supported by staff at the end of the ISU study (M = 3.8, SD = 0.8).

Expectation Item Themes

Overall, only a very small percentage of staff agreed with the statement that it was not possible to establish an effective ISU in a prison setting, as shown for Item 24 in Table 19. In addition to the general question about ISU functioning, 13 other questions about expectations for the ISUs were asked shortly after the ISUs were started. These questions are organized around four themes: a positive work environment, interdiction activities, decision to work in the ISU, and impact of the ISU on the remainder of the Institution. Results are presented separately for minimum and medium security institutions, as they tend to have different characteristics.

Staff in medium security institutions believed that the ISU would provide a more positive and interesting work environment. This result was not found for the minimum-security institutions, possibly because these institutions are already strongly rehabilitation oriented and the addition of the ISU might have little impact on their environment.

Staff were generally comfortable with the need for, and potential effectiveness of, the increased interdiction activities in the ISUs. For example, over 80% of staff were comfortable with the need for more drug testing and searching, and ¾ of staff believed these measures would have a positive impact on the offenders.

In terms of reasons for working in the ISU, staff were not there because they had been coerced and nor were they trying to get away from the general environment of the prison. It would appear that they were choosing to work in the ISU for more positive reasons, because they felt they could have an impact.

Staff also indicated that the ISU would have little impact on the remainder of the institutions, either generally or in terms of other staff or inmates. For example, only 7% thought other staff might react negatively to their work in the ISU, and less than 4% believed there would be a change in drug use in the remainder of the prison.

Table 19 : Staff Expectations Themes and Items

Impact Item Themes

As noted earlier, staff completed the Impact questionnaire at the end of the study. The questionnaire was used to measure their perceptions of what happened in the units. The items from the questionnaire were grouped into two themes: presence of substances in the ISU and positive impact of the ISU on offenders. Results for the individual questions are presented in Table 20.

Overall, 80% of staff believed the ISUs were more substance free than other parts of the prison and they believed that the combination of offender motivation and interdiction activities contributed to this effect. However, staff in minimum security institutions, as compared to staff in medium security institutions, believed more strongly in the positive impact of both interdiction activities and offender motivation in reducing the presence of drugs and alcohol in the units. In general, staff in minimum security institutions were 10 to 30 percentage points higher in their judgements of impacts than those in medium security institutions. Three-quarters of staff believed the interdiction activities were partly responsible for the reduced presence of drugs and alcohol, while at the same time, 67% believed offender motivation was partly responsible for the observed result.

Overall, staff indicated that they believed the ISU experience would have a positive impact on the reintegration of offenders. More specifically, three quarters (77%) believed the ISU helped offenders follow their correctional plan and a similar percentage believed that the ISU experience would help offenders after their release into the community. Again, there were differences between staff in minimum and medium security institutions, with staff at minimum security institutions indicating greater impacts for the ISU.

Table 20 : Staff Impact Themes and Items

Staff Knowledge and Job Orientation

Staff were administered the same questionnaires as offenders for Consequences of Alcohol, Consequences of Drug Use, and the How Much Do They Matter scales. In addition, staff responded to several questionnaires relating to their work, including measures of attitudes toward offenders, job satisfaction, and various dimensions of professional orientation.

Scores for knowledge of the Consequences of Alcohol Use and Consequences of Drug Use, shown in Table 21, indicate that staff are slightly more knowledgeable about the consequences of alcohol use, but similarly informed about drug use, as offenders included in the ISU study and offenders who completed the same scales as part of an evaluation of the Offenders Substance Abuse Pre-Release Program (OSAPP) (T3 Associates, 1999). Overall both offenders and staff are well aware of the consequences of drug and alcohol use as shown by their high scores on these scales. A third scale, entitled how much do they matter, also indicated good awareness of the negative impacts drug and alcohol use can have.

Table 21 : Staff Knowledge of Substance Use Compared to ISU and OSAPP Participants

Staff were also asked a series of questions relating to their professional orientation. Results from these questions were compared to similar scales used in a national survey (see Appendix M for comparison scales) of CSC correctional staff conducted in 1995 (Correctional Service Canada, 1996). Staff working in the ISU were shown to be more empathic towards offenders, had a more rehabilitative orientation and demonstrated greater job satisfaction than correctional officers generally. Ratings on the scale of punitiveness show no differences from other correctional officers. Note that those completing the scales in the ISU were mostly correctional officers but included some supervisory staff. The results from these scales are presented in Table 22.

Table 22 : Staff Professional Orientation and Job Satisfaction Scores

DISCUSSION

Overall, the results of the research indicate that the ISU experience was positive for offenders and the institutions. ISU participants (not involuntarily discharged) were more likely to have received a discretionary release (day parole or full parole), and after release they remained in the community longer than offenders in the comparison groups. In addition, ISU offenders had a lower rate of misconducts and substance use in relation to the comparison group. Surveys conducted at the time of admission to the units indicated that the ISUs were expected to have a positive impact on offenders' behaviour and serve as a source of support. Based on observed results and questionnaire information collected at the time of discharge from the units, these expectations were realized.

Correctional staff also had positive expectations for the units and reported that they perceived the units as having been successful at the end of the evaluation period. Staff working in the units were shown to be more empathic towards offenders, had a greater rehabilitation orientation and were more satisfied with their job in relation to a comparison group.

Overall, the ISUs met the principles under which they were created, attracting mostly offenders with substance abuse problems. However, ISUs were also designed for offenders who did not have a substance abuse problem, but who were seeking an environment that would support them in their drug free lifestyle. Approximately 1 in 5 ISU participants were in this latter group, suggesting that the drug free environment was a positive feature for this group. Offenders in the ISUs continued to participate in institutional work, treatment and social programs. Results from search and drug use data suggest that efforts to reduce the presence of drugs in the units were successful, and this result is supported by questionnaire data from both offenders and staff.

More specific findings of the study are presented in the following sections. First, answers are provided to the four research questions posed in the Introduction. This is followed by highlights of other findings, a discussion of the potential limitations of the study, and finally, the implications of the results for developing the ISUs are discussed.

Research Questions

1. What are the characteristics of inmates residing in these units?

Interventions are generally designed to meet the needs of a specific group of offenders. Determining if, in fact, the intervention attracted, or was used with these offenders is an important part of the evaluation. About 80% of the ISU population were identified at intake as having a substance abuse problem. In addition, offenders in the ISU also self-identified themselves as having alcohol and drug problems, with the majority having drug related problems. However, about 20% of the ISU population did not have identified drug or alcohol problems. At the outset, it was expected that some offenders would choose to live in the ISU even if they did not have a substance abuse problem. For these offenders, the ISU provided an environment where they could more easily stay away from drugs and alcohol.

Offenders in the ISU were similar to the population of offenders in medium and minimum security institutions with a few differences. ISU offenders were younger, serving shorter sentences (fewer serving life sentences), less likely to be serving a sentence for a violent offence, and were rated as lower risk on static factors. None of these differences were large, suggesting that the ISU was successful in attracting a range of offenders. Aboriginal offenders were under-represented in the ISU population, suggesting a need to put greater emphasis on attracting these offenders, or finding similar alternative arrangements.

Almost 80% of offenders moved to the ISU to help themselves “stay clean”, that is they chose the ISU environment to continue to be drug free. This is consistent with the objectives of the ISUs

2. Do these units reduce alcohol and drug availability?

Four sources of information are available to indicate that, in fact, the ISUs were able to reduce the presence of drugs and alcohol. First, in terms of personal and common area searches, the results indicated a higher level of activity consistent with the mandate of the ISUs. However, the higher level of searching produced lower rates of seizures, indicating that there was a reduction in the presence of alcohol, drugs and substance related contraband in the ISUs.

The second source of information on the availability of drugs in the units is from drug testing. Both screening tests and institution based random urinalysis testing indicated that while offenders were in the ISUs the rate of positive tests were significantly lower. While random urinalysis testing in federal correctional institutions generally produces a positive drug test rate of approximately 12% (MacPherson, 2001), this rate was 6% while offenders were in the ISU (only 3% for offenders voluntarily discharged from the ISU), and the more frequently used screening tests, produced a positive rate of 1.5%. Together, these results suggest that drug use was significantly lower in the ISU, and therefore, availability was probably also lower.

The third source of information on drug and alcohol availability comes from the questionnaire data collected from offenders. They indicated, after six weeks in the units and at the time of their discharge, that drugs and alcohol were less available in the ISU than on other units in the prison. Finally, the staff questionnaires, completed at the end of the study, also indicated less availability of drugs and alcohol in the units.

3. Is there an impact on offender behaviour?

An attempt was made to measure changes in offender behaviour across time using a behavioural checklist. These data indicated that there was little change since all behaviours were rated high (80%, or higher) shortly after admission to the units. However, a positive change was observed in the area of interpersonal relationships.

Results of analyses on institutional behaviour indicated a lower rate of misconducts, both minor and major, in the ISUs in relation to a comparison group. Part of this effect may have been due to the stricter rules in the ISU that required discharge from the unit in the event of negative behaviour. However, the effect of this was to create a more stable living environment that would encourage positive behaviour change and create an environment in which offenders are supported in addressing their substance abuse problems.

Drug testing results, presented in the previous section, also support the conclusion that the ISUs resulted in positive changes in behaviour with lower rates of positive tests for drug use.

Data on release decisions indicated that offenders who were discharged voluntarily from the units were more likely, than those in the comparison groups, to have received a discretionary release, the release was received earlier in the sentence and that after release the offenders remained in the community longer.

The majority of offenders, in all groups, were returned to custody because they violated a condition of their release (revocation without offence); with the ISU participants voluntarily discharged having the lowest revocation rate at 16% compared to the other groups. In terms of a return to custody for committing or being suspected of committing a new offence (revocation with an offence, revocation with outstanding charges and admission with new offence), the differences are greater. While 9% of the ISU participants who left voluntarily were readmitted with a new offence, or suspicion of a new offence, the rate was more than double for the involuntarily discharged offenders (21%) and close to double for the matched comparison group (17%). The ISU participants voluntarily discharged were less likely to be readmitted and were less likely to be readmitted in association with a new offence.

Finally, results from the staff questionnaire suggest that the ISU helped offenders address their correctional plan and increased the likelihood of success in the community.

Taken together, these results suggest that the ISUs had a positive impact on the behaviour of offenders, both while they were in the institution and after their release into the community.

4. Is there an impact on the rest of the institution?

Data on the impact of the ISU on other parts of the institution were difficult to obtain. It is therefore necessary to rely on the opinions of staff and inmates to answer this question. Overall, 86% of staff surveyed indicated that they disagreed with the statement, “I think that the ISU caused more disruption in the rest of the institution”. In addition, at the time of discharge from the ISU offenders indicated that they had not perceived any impacts on the rest of the prison. Finally, staff and inmates did not expect, nor perceive, any negative impact on ISU participants from other inmates in the institutions. Overall these results suggest that the ISU likely had minimal impact on correctional operations in the rest of the prison.

Other Findings

Offenders who moved to the ISUs were experienced drug and alcohol abusers. Three-quarters of the participants had participated in treatment programs in the past; generally having been referred by the criminal justice system, but some reported they had self-referred. Participants also reported that they had continued to use drugs and alcohol while incarcerated. However, testing with the URICA, indicated that, on average, they were ready for change in their drug and alcohol use problems, being located between the contemplative and action stages in Prochaska & DiClemente's (1992) model of change.

Participants' motivation to change can be seen in responses to the Expectation questionnaire and the intake interview in which they indicated they were participating in the ISU to get away from drugs and alcohol and that 78% expected the ISU to provide an opportunity to "stay clean".

ISU participants and staff indicated that they expected the units to provide a more supportive environment for change, and after experience with the units this positive attitude was maintained. While in the ISU, 27% of offenders participated in institutional programs, including the Offender Substance Abuse Pre-Release Program (OSAPP). They also participated in self-help programs (42% in minimum security and 33% in medium security) and continued to be employed in institutional jobs. While they recognized that they would receive no special benefits for living in the ISUs, participants who were discharged voluntarily were more likely than other offenders to receive a discretionary release and the release occurred earlier than for other offenders.

It is important to recognize that staff believed the ISU would be effective. They accepted the need for increased interdiction activities, but also recognized that the offenders were motivated to change their behaviour. Staff working in the ISU had more empathy for the offenders and a more rehabilitation orientation than a comparison group and were more satisfied with their work.

The positive environment generated in the ISU likely contributed to the observed reductions in drug use, misconducts and better post-release results. Wright (1993) has argued that a positive and supportive environment can contribute to programming being more successful in correctional settings, which is consistent with the findings in this study.

The benefits of the ISU extend beyond the positive changes in behaviour that were observed. The potential impact of the ISU on correctional operations can be seen in the earlier release of some ISU participants and the longer period of time in the community after release. More specifically, offenders who were voluntarily discharged from the ISUs were released between one and two months earlier than those in the comparison groups and they remained in the community between one and two months longer, on average. This translates into a reduction of time in custody of between two and four months, a difference that would produce a significant cost reduction to the Correctional Service of Canada for these offenders. In addition, safety of the communities was enhanced as ISU offenders were less likely to be readmitted with a new criminal offence than the comparison groups.

The potential cost savings of the ISU can be estimated by comparing the cost of supervising an offender in the community with the cost of managing an offender in a penitentiary. Overall, it costs $46/day to supervise an offender in the community and $182/day for custodial care in a penitentiary (Correctional Service Canada, 2002). The difference between custodial care and community supervision is approximately $136 per day. Results in the study suggest that offenders who are successful in the ISU (voluntarily discharged) are released between 30 and 60 days earlier than those in the comparison groups, and that they remain in the community, on average, between 30 to 60 days longer during a 12 month follow-up. Using the lower number of 30 days saved for early release and 30 days longer in the community the potential is for a saving of 60 days of custody. The potential cost savings is 60 days * 136 = $8,160 per offender.

Operational costs for these units are minimal as the only additional costs are for drug screening tests ($15. per test), staff training and some additional staff costs associated with increased searching. However, these additional staff costs were incorporated into normal operations during the period of the study as searching is part of the prison routine. Given that the average stay in the ISU is 10 weeks and if each offender were tested bi-weekly, the cost for drug screening tests would be less than $100 per offender.

If only 100 offenders participated successfully in the ISU, the savings would exceed $800,000.

Limitations

The study was undertaken on operational units and, while this increases the ecological validity of the findings, it also creates limitations on the generalizability of the observed results. Some of the limitations are discussed briefly below but, as in all applied research, the researchers did not have control over all variables. At times, data collection and intervention integrity were compromised for operational reasons and this may have affected results. However, overall, the results are consistent across different measurement domains, suggesting that results are valid.

The ISUs used in the study were implemented in prison settings that had a strong interest in providing interventions for substance abuse. In a number of instances, interventions that included peer group support were already in existence in these prisons when the ISU model was implemented. Details of these differences are presented in the Method section. The positive expectations and results may have been partly due to specific locations chosen for the ISUs in the study. Additional research may be needed to confirm that the impacts observed in this study are generalizable to other prisons.

Offenders who participated in the ISUs were self-selected, that is, they volunteered to move to the units. Based on the level of motivation observed, these offenders were likely ready for change and some would have had positive results whether or not they participated in the ISU study. However, results were compared to a matched sample of offenders thus reducing the potential bias from this source. In addition, most programming and interventions are provided on a voluntary basis to offenders.

A number of the research instruments used in the study created limitations for the interpretation of results. In general, these were tools specially constructed for the study and there was little opportunity for pre-testing their use. In general, these tools had a ceiling effect in that most offenders scored relatively high at the outset leaving very little room for change to occur over the study period. Examples of these include the Behavioural Rating Scale and the Knowledge of Substance Abuse Scales. While these tools limited the ability of the study to identify findings in these areas, they did not have an impact on the overall study findings.

Implications

Overall the study indicates that the ISUs had positive impacts on the institutional and release behaviour of offenders. Those offenders who were voluntarily discharged from the units showed the greatest positive benefits. The analysis of cost savings also indicates a positive economic effect for the Correctional Service of Canada. The results demonstrate that ISUs should be considered for use on a wider basis.

In fact, early results from the study encouraged the Correctional Service Canada to implement ISUs in all federal penitentiaries. These units are now being studied to determine if the positive impacts observed at the five pilot sites can be maintained with a national implementation.

The research indicates a number of themes that should be critical to the implementation of the ISU concept. While these were believed to be important at the outset of the ISU project, there is now evidence of their importance. Some of these themes include:

1. Creation of a supportive environment that meets the needs of the offenders;
2. Well trained staff who can encourage and support behavioural change;
3. Interdiction activities that ensure a drug and alcohol free environment; and
4. Continued access to programming and work opportunities for offenders in the ISU.

The expected benefits that both staff and offenders anticipated and achieved from the ISUs suggest that there is a great deal of potential in the concept. The implementation of the ISU pilot sites used for the study was not ideal. The concept was new at the time the sites were created and much has been learned through this research. It is time to develop the concept further and to establish more definitive intervention parameters than were used for the study. Work on this is underway, but the development of new model sites could help to further develop the concept. The model sites could serve to guide others in establishing their ISUs and could assist in developing guidelines for searching and drug testing protocols.

REFERENCES

Andrews, D.A. and Bonta J. (2002). The Psychology of Criminal Conduct . Cincinnati, OH, Anderson Publishing Co.

Andrews, D.A., Bonta, J. & Hoge, R.D. (1990). Classification for effective rehabilitation: Rediscovering psychology. Criminal Justice and Behavior , 17, 19-52.

Breteler, M.H., Van Den Hurk, A.A., Schippers, G.M., & Meerkerk, G.J. (1996). Brief Report. Enrolment in a Drug-Free Detention Program: The Prediction of Successful Behavior Change of Drug-Using Inmates . Addictive Behaviors , 21(5), 665-669.

Brochu, S., Cournoyer, L.-G., Pernanen, K. & Motiuk, L. (2001). Drugs, alcohol, and criminal behaviour: A profile of inmates in Canadian federal institutions . FORUM on Corrections Research , 13(3), 20-24.

Brown, S. (2001). Cost-Effective Correctional Treatment . In L. Motiuk & R. Serin (Eds.), Compendium 2000 on Effective Correctional Programming , Vol. 1 . (pp. 221-233). Ottawa , Canada : Ministry of Supply and Services Canada .

Correctional Service of Canada . (1996). Staff Survey Vol 1, May 1996 . Ottawa , Canada : Coopers & Lybrand Consulting.

Correctional Service Canada . (2000a). CSC Intensive Support Units -Backgrounder. Ottawa , Canada : Correctional Service Canada .

Correctional Service Canada . (2000b). News Release - Federal Solicitor General Announces Drug Strategy Initiatives. Ottawa , Canada : Correctional Service Canada .

Correctional Service Canada . (2004). Mission of the Correctional Service of Canada . 7 th Ed. Ottawa , Canada : Minister of Public Works and Government Service Canada .

Correctional Service Canada (2004). Basic Facts About Federal Corrections . Ottawa , Canada : Correctional Service of Canada .

Correctional Service Canada . (2003). Standard Operating Practice 700-04: Of fender Intake Assessment and Correctional Planning . Ottawa , Canada : Correctional Service Canada .

Corrections and Conditional Release Act , RSC, C - 20, (1992).

Corrections and Conditional Release Regulations , SOR/92-620, (1992).

French, S. & Gendreau, P. (2003). Safe and Humane Corrections Through Effective Treatment. Ottawa : Research Branch, Correctional Service Canada

Gaines. L. K. & Kraska, P. (1997). Drugs, Crime and Justice: Contemporary Perspectives . Prospect Heights, IL, Waveland Press Inc

Gendreau, P., Goggin, C. & Little, T. (1996 ). Predicting Adult offender recidivism: What Works! Ottawa , Canada : Ministry of the Solicitor General of Canada .

Gendreau, P. & Ross, R. R. (1984 ). Offender Change: An Effective Alternative to Incarceration . Ottawa , Canada : Ministry of the Solicitor General of Canada .

Gillis , C.A. (1994). The influence of shop supervisor characteristics on employee-reported work attitudes in a prison industry setting . Unpublished Master's thesis, Department of Psychology, Carleton University , Ottawa , Ontario .

Grant, B.A., Kunic, D., MacPherson, P., McKeown, C., & Hansen, E. (2003). The High Intensity Substance Abuse Program (HISAP): Results from the Pilot Program. Research Report R-140 . Ottawa : Correctional Service Canada .

Gunn, Orenstein, Iverson & Mullen (1983). An Evaluation Handbook for Health Education Programs in Alcohol and Substance Abuse . Centre for Disease Control, Atlanta , Georgia .

Incorvaia, D., & Kirby, N. (1997). A Formative Evaluation of a Drug-Free Unit in a Correctional Services Setting. International Journal of Offender Therapy and Comparative Criminology , 41(3), 231-249.

Johnson, G. & Farren, E. (1996). An evaluation of prisoners' views about substance free zones. Issues in Criminological & Legal Psychology , 25, 30-38.

Klingeman, H. & Hunt, G. (1998). Drug Treatment Systems in an International Perspective: Drugs, Demons, and Delinquents . Thousand Oaks , CA , Sage.

Klofas, J. & Toch, H. (1982). The Guard Subculture Myth. Journal of Research in Crime and Delinquency , 19, 238-254.

Lightfoot, L.O. & Barker, J. (1989). A Field Test of the Revised Substance Abuse Pre-Release Program: Joyceville Institution . Correctional Services Canada , Ottawa .

MacPherson, Patricia. (2001). Random Urinalysis Program: Policy, practice, and research results. FORUM on Corrections Research, 13 (3), 54-57.

Melvin, K. B., Gramling, L. K. & Gardiner, W. M. (1985). A Scale to Measure Attitudes Towards Prisoners. Criminal Justice and Behavior , 12 (2), 241-253.

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

Paulhus, D.L. (1994). Balanced Inventory of Desirable Responding: Reference Manual for BIDR Version 6. Toronto , Ontario : Multi-Health Systems, Inc.

Pernanen, K. & Brochu, S. (1997). Attributable Factions for Alcohol and Other Drugs in Relation to Crimes in Canada - Literature Search and Outlines of Data Banks. Ottawa , Ontario : Canadian Centre on Substance Abuse.

Pernanen, K., Cousineau, M.-M., Brochu, S., & Sun, F. (2002). Proportions of Crimes Associated with Alcohol and Other Drugs in Canada . Ottawa , Ontario : Canadian Centre on Substance Abuse.

Peters, R.H. & Steinburg, M.L. (2000). Substance Abuse Treatment in US Prisons. In D. &. D. J. B. Shewan (Ed . ), Drug Use and Prisons - An International Perspective . (pp. 89-116). Amsterdam : Harwood Academic Publishers.

Prochaska, J.O. and DiClemente C.C (1992). Stages of Change in the Modification of Problem Behaviors . Newbury Park , CA , Sage.

Robinson, D. & Mirabelli, L. (1995). Summary of Findings of the 1995 CSC National Inmate Survey. Research Report B-14 . Ottawa : Correctional Service Canada .

SAS Institute Inc. (1999). Statistical Analysis System (SAS), Version 8. Cary , NC .

Schippers, G.M., Van Den Hurk , A.A., Breteler, M.H., & Meerkerk, G.J. (1998). Effectiveness of a Drug-Free Detention Treatment Program in a Dutch Prison. Substance Use & Misuse , 33(4), 1027-1046.

Skinner, H.A. (1982). The Drug Abuse Screening Test. Addictive Behaviors , 7, 363-371.

Skinner, H.A., & Horn, J.L. (1984). Alcohol Dependence Scale ( ADS ): User's Guide. Toronto : Addiction Research Foundation.

Single, E., Robson, L., Xie, X. & Rehm, J. (1998). The Economic Cost of Alcohol, Tobacco and Illicit Drugs in Canada , 1992 . Addiction , 93(7), 991-1007.

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

Van Doorninck, M. &. de Jong, .W. (2001). Development of HIV/AIDS Policy in the Dutch Prison System. In D. &. D. J. B. Shewan (Ed.), Drug Use and Prisons - An International Perspective . (pp. 173-202). Amsterdam : Harwood Academic Publishers.

Walters, G. D. (1994). Drugs and Crime in Lifestyle Perspective . Thousand Oaks , CA , Sage.

Warr, P. Cooke, J. & Wall, T. (1979). Scales for the Measurement of Some Work Attitudes and Aspects of Psychological Well-Being. Journal of Occupational Psychology , 52, 129-148.

Whitehead, J. T., and Lindquist, C. A. (1989). Determinants of correctional officers' professional orientation. Justice Quarterly , 6, 69-87.

Wright K.N. (1993). Prison Environment and Behavioral Outcomes. Journal of Offender Rehabilitation , 20(1/2), 93-113.

Zilkowsky, D. (2001). Canada 's National Drug Strategy . FORUM on Corrections Research , 13(3), 3-4.

APPENDIX A

Informed Consent Form

The purpose of an informed consent is to make sure that you understand the purpose of the study and how you will be involved in the project. The informed consent provides enough information so you have the chance to decide if you want to participate in the research.

Research title:

Evaluation of Intensive Support Units.

Research personnel:

The following people are involved in this study and may be contacted at any time: Stafford Murphy (Research Branch, CSC: 613-536-6538) or Dr. Brian Grant (Research Branch, CSC: 613- 943-8871). If you have any concerns about this study then please contact either of the above.

Purpose and requirements:

The purpose of the study is to examine your ideas and expectations about living in an Intensive Support Unit so we can see if the unit responds to your needs. If you participate, you will be asked to complete some questionnaires during the time you stay in the unit. The first questionnaire explores your first impressions about living in the unit, and will take about 15 minutes to complete. After you have been living on the unit for more time, you will be asked to complete questionnaires that will examine your perceptions of the unit and information on your drug and alcohol use. Also, with your permission, the researchers will follow your performance on the unit and following your release from the institution, to gain a better understanding of how the unit has affected your institutional and community adjustment.

Participation and right to withdraw:

Information gathered in the study will be kept confidential and there is no reward for participating, and no penalty for not participating in the study. If you decide not to participate it will not affect your stay in the unit. You may withdraw at any time from the study. However, your participation will help the researchers to gain a better understanding of how the Intensive Support Units are working and how to improve their operation.

Confidentiality:

Information gathered in the study will be kept confidential and will be made available only to the researchers. Individual information will not be released to unit staff, other institutional staff or anyone not directly involved with the research. All information will be coded so that your name is not associated with the data. Individual offenders will not be identified in any reporting of the results of this research. Only group results will be presented.

While professional confidentiality will be maintained for all participants in this assessment, there are some limits to that confidentiality. There are some situations in which research staff is required by law to break confidentiality. These situations include:

1. if there is concern that a participant might attempt to harm or kill himself or someone else;
2. if there is concern that a participant or other identifiable offenders might attempt to escape;
3. if there is concern that a participant or other known people are involved in on-going criminal activity;
4. if a participant gives specific information about a child who has been abused or who is currently at risk of abuse;
5. if a client gives specific information about prior criminal offences for which he has not been arrested.
6. In some situations, security will be notified and appropriate precautions will be put in place. In others, the Children's Aid Society or police may be notified and an investigation will be initiated. The participant will be advised of the action taken.

Signatures:

I have read the above description of the research evaluating the Intensive Support Units.

I also have read and understand the limits to confidentiality described above.

I give my permission to the researcher to make use of the information gathered in these questionnaires. I agree to allow the researchers to monitor my inmate file for follow-up during my stay on the unit, after I leave the unit, and after my release from the institution.

Full Name (Print): _______________________________________

FPS#: ________________________________________________

Participant Signature: _____________________________________

Witness Name: _________________________________________

Witness Signature: _______________________________________

Date: _________________________________________________

APPENDIX B

Offender Intake Interview Questionnaire

SECTION I : BASIC INFORMATION

For completion within the offender's first 8 weeks in the I.S.U.

OFFENDER NAME: _______________________________

OFFENDER FPS #: ________________________________

INSTITUTION: ___________________________________

DATE OF ARRIVAL
ON THE ISU: ____________________________________

AGE: ___________________________________________

RACE: __________________________________________

MARITAL STATUS: _______________________________

PLANNED DATE OF
RELEASE: ________________________________________

DATE OF INTERVIEW: _____________________________

SECTION II : SURVEY

i) ALCOHOL & DRUG USE HISTORY

How OLD were you when you FIRST TRIED alcohol? ____

How OLD were you when you started to drink REGULARLY? ____

How OLD were you when your drinking became a PROBLEM? ____

How OLD were you when you FIRST TRIED drugs? ____

What type?

___ Benzodiazepines (valium, librium)

___ Barbiturates (secobarbital, reds, red devils, pentobarbital, yellows, phenobarbital, blues)

___ Opiates (opium, codeine, morphine, heroin, percs, Mexican brown, China whites, percodan, demoral )

___ Amphetamines (crank, speed, ice, shabu, meth, black beauties, whites, bennies, beans, diet pills)

___ Cannabinoids (marijuana, thc, grass, pot, hash, hash oil, Acapulco gold, reefer, joint, dubie)

___ Cocaine (coke, blow, crack, etc.)

___ Psychedelics (LSD, mushrooms, acid, mescaline, STP, rave, ecstasy, XTC, PCP, angel dust, krystal)

___ Inhalants (rubber cement, gasoline, laughing gas, nitrous oxide, ether, chloroform, lighter fluid)

___ Other (specify): __________

How OLD were you when you started to use drugs REGULARLY?____

How OLD were you when your drug use became a PROBLEM? ____

What TYPE of drug has caused you the most problems in your life?

___ Benzodiazepines

___ Barbiturates

___ Opiates

___ Amphetamines

___ Cannabinoids

___ Cocaine

___ Psychedelics

___ Inhalants

___ Other (specify): ____________

___ Combinations (specify): __________

Did your substance use cause any problems for you in the 12 MONTHS before you committed your offence?

In the ONE MONTH before you committed your offence, on average, how many days of the week would you have used/abused alcohol and drugs?

ALCOHOL: Used: /7 Abused: /7

DRUGS: Used: /7 Abused: /7

Was substance use related to your current offence?

ALCOHOL: Yes ___ No ___

How related: Crime committed to pay for alcohol ___

Crime committed under the influence ___

Crime committed under the influence and to pay for alcohol ___

DRUGS: Yes ___ No ___

How related: Possession ___

Dealing/Trafficking ___

Crime committed to pay for drugs ___

Crime committed under the influence ___

Crime committed under the influence and to pay for drugs ___

Was substance use related to any prior offences? Yes ___ No ___

Have you used substances during your current incarceration? Yes ___ No ___

What type?

___ Alcohol/Brew

___ Benzodiazepines

___ Barbiturates

___ Opiates

___ Amphetamines

___ Cannabinoids

___ Cocaine

___ Psychedelics

___ Inhalants

___ Other (specify): __________

Have you used substances during previous incarcerations? Yes ___ No ___

Have you been involved in the institutional drug trade at any time during your current incarceration? Yes ___ No ___

Have you been involved in the institutional drug trade during previous incarcerations? Yes ___ No ___

When in the COMMUNITY what was your reason for using substances?

When INCARCERATED what was your reason for using substances?

If you used while incarcerated, do you believe that your substance abuse problem was made worse during your incarceration(s)?

Yes ___ No ___

How? ___________________________________________

Have you experienced withdrawal:

From Alcohol

a. In the community Yes ___ No ___

b. While incarcerated Yes ___ No ___

c. This incarceration Yes ___ No ___

From Drugs

a. In the community Yes ___ No ___

b. While incarcerated Yes ___ No ___

c. This incarceration Yes ___ No ____

ii) SUBSTANCE ABUSE TREATMENT HISTORY

Describe your substance abuse treatment history including the following information:

Age, referral source, type of treatment, approximate length (weeks), site (community vs. institutional), completion (yes/no), motivation for treatment (1-5), substance problem at the time. Include programs you are currently waiting to receive.

Use the following charts to help code the history :

How OLD were you when it was first recommended or ordered that you receive treatment for your drinking/drug use?

AGE (drinking) _____ AGE (drugs) _____

Referral source:

a. Sought it yourself

b. Family/friends

c. Employer

d. Teacher/school

e. Courts

f. Prison/jail

g. Probation/parole

h. Physician/hospital

i. Counsellor/therapist

j. Clergy/elder

k. Other (specify) __________

Type of Treatment :

Residential

Non-residential individual

Non-residential group

A combination of b & c

AA/NA

OSAPP

Substance Abuse Problem (at time of treatment):

Alcohol/Brew

Benzodiazepines

Barbiturates

Opiates

Amphetamines

Cannabinoids

Cocaine

Combinations (specify) ______

Other (specify): __________

Do you believe you have a current problem with alcohol/drugs? Yes ___ No ___

If yes , what type ?

Alcohol/Brew

Benzodiazepines

Barbiturates

Opiates

Amphetamines

Cannabinoids

Cocaine

Combinations (specify) ______

Other (specify): __________

Are you dependent ? Yes ___ No ___

On a scale of 1 – 5, how confident are you that you have your substance abuse problem under control:

1  -  2  -  3  -  4  -  5

Does your substance abuse problem currently interfere with your functioning ?

Have you ever tried to quit using :

ALCOHOL: YES ___ NO ___ How many times? ___

DRUGS: YES ___ NO ___ How many times? ___

How many months has it been since you have used ?

Less than one month

1 - 3 months

3 - 6 months

6 - 12 months

over twelve months

Type?

Alcohol/Brew Benzodiazepines

Barbiturates Opiates

Amphetamines Cannabinoids

Cocaine Psychedelics

Inhalants Combinations (specify) _________

Other (specify): __________

What is your goal with regards to your substance use when you are no longer under supervision ?

ALCOHOL: No Change ___ Abstinence ___ Moderation ___

DRUGS: No Change ___ Abstinence ___ Moderation ___

SECTION III: ADDITIONAL COMMENTS

COMMENTS: ____________________________________________

APPENDIX C

Offender Knowledge Survey

SECTION I : INTRODUCTION

This questionnaire is part of a study being conducted by the Research Branch and Performance Assurance Sector to examine how offenders feel about living in the Intensive Support Units. While the questionnaire is voluntary, your cooperation is important in order for CSC to get a better understanding of how the Intensive Support Units are helping offenders. If you do agree to complete the questionnaire, please answer all of the questions.

This questionnaire is not a personal evaluation.

All of your responses will be treated with confidentiality. The research team will only handle your completed questionnaire. Your answers will only be used within the Research Branch and Performance Assurance Sector and will not be released to other institutional staff, National Headquarters staff or the National Parole Board. Your individual responses will be combined with responses from other inmates at this institution and from other Intensive Support Units participating in the study, so you cannot be identified.

SECTION II : SURVEYS

Consequences of Alcohol

This test consists of 20 statements about the effects of alcohol use. Some of the statements are true while others are false. If you think a statement is true, check the column labelled TRUE . If you think a statement is false put a check in the column labelled FALSE .

Consequences of Drug Use

This test consists of 27 statements about the consequences of drug use. Some of the statements are true and some are false. If you think a statement is true put a check in the column labelled TRUE . If you think a statement is false put a check in the column labelled FALSE .

How Much Do They Matter?

This survey is about how people might be affected by using drugs or alcohol not prescribed by their doctor. Read each statement. Decide the extent to which you agree with it. Circle the appropriate number to the right of the statement. Use the following scale:

Relapse Attitude / Knowledge

On the following pages are some statements that you may agree or disagree. Circle the answer that best represents your feeling about the statement. Pick the answer which best represents your general feeling or the way you usually feel.

Please indicate your feelings about every statement by circling one of the five answers.

APPENDIX D

  University of Rhode Island Change Assessment (URICA)

Instructions: This questionnaire is to help us improve treatment services. Each statement describes how a person might feel when starting therapy or approaching problems in their lives. Please indicate the extent to which you tend to agree or disagree with each statement. In each case, make your choice in terms of how you feel right now, not what you have felt in the past or would like to feel. For all the statements that refer to your "problem", answer in terms of your use of alcohol or drugs. And "here" refers to the Contract Unit.

There are FIVE possible responses to each of the items in the questionnaire: strongly disagree, disagree, unsure, agree and strongly agree. Circle the number to the right of the item that best describes how much you agree or disagree with each statement.

Problem (please check the problem that applies to you):

______ Alcohol use ______ Drug use

______ Alcohol and drug use

APPENDIX E

Offender Expectation Questionnaire

SECTION I : INTRODUCTION

This questionnaire is part of a study being conducted by the Research Branch and Performance Assurance Sector to examine how offenders feel about living in the Intensive Support Units. While the questionnaire is voluntary, your cooperation is important in order for CSC to get a better understanding of how the Intensive Support Units are helping offenders. If you do agree to complete the questionnaire, please answer all of the questions.

This questionnaire is not a personal evaluation.

All of your responses will be treated with confidentiality. The research team will only handle your completed questionnaire. Your answers will only be used within the Research Branch and Performance Assurance Sector and will not be released to other institutional staff, National Headquarters staff or the National Parole Board. Your individual responses will be combined with responses from other inmates at this institution and from other Intensive Support Units participating in the study, so you cannot be identified.

SECTION II : BASIC INFORMATION

Name: _______________________________

FPS: _______________________________

Institution: _______________________________

Date: _______________________________

Date you moved to I. S. Unit: _______________________________

SECTION III : SURVEY

Directions: In the next few pages, you will find a series of statements that you might agree or disagree with. Circle the answer that best reflects how you feel about each statement. Please answer every question, so we can get your opinion on a number of different issues involving the Intensive Support Units.

A = Perceived Effectiveness scale items.

B = Expected Benefits scale items.

C = Expected ISU Support scale items.

D = Impact of ISU on Institution scale items.

SECTION IV: ADDITIONAL COMMENTS

Why do you think the Intensive Support Units will be effective? Please comment.

________________________________________________________________

________________________________________________________________

Do you think there will be any problems with the units? Please comment.

________________________________________________________________

________________________________________________________________

Please add any additional comments in the space below.

________________________________________________________________

________________________________________________________________

Thank you for your cooperation

APPENDIX F

Offender Post Evaluation Questionnaire

SECTION I : INTRODUCTION

This questionnaire is part of a study being conducted by the Research Branch and Performance Assurance Sector to examine how offenders feel about living in the Intensive Support Units. While the questionnaire is voluntary, your cooperation is important in order for CSC to get a better understanding of how the Intensive Support Units are helping offenders. If you do agree to complete the questionnaire, please answer all of the questions.

This questionnaire is not a personal evaluation.

All of your responses will be treated with confidentiality. The research team will only handle your completed questionnaire. Your answers will only be used within the Research Branch and Performance Assurance Sector and will not be released to other institutional staff, National Headquarters staff or the National Parole Board. Your individual responses will be combined with responses from other inmates at this institution and from other Intensive Support Units participating in the study, so you cannot be identified.

SECTION II : BASIC INFORMATION

Name: ______________________________

FPS: ______________________________

Institution: ______________________________

Date: ______________________________

Date you moved to the ISU: ______________________________

Date you left the ISU: ______________________________

SECTION III: SURVEY

Directions: In the next few pages, you will find a series of statements that you might agree or disagree with. Circle the answer that best reflects how you feel about each statement. Please answer every question, so we can get your opinion on a number of different issues involving the Intensive Support Units.

A = Perceived Effectiveness scale items.

B = Expected Benefits scale items.

C = Expected ISU Support scale items.

D = Impact of ISU on Institution scale items.

SECTION IV: ADDITIONAL COMMENTS

What did you like best about the Intensive Support Unit? Please comment.

________________________________________________________________

________________________________________________________________

What did you like the least about the Intensive Support Unit? Please comment.

_________________________________________________________________

_________________________________________________________________

Please add any additional comments in the space below.

_________________________________________________________________

_________________________________________________________________

Thank you for your cooperation.

APPENDIX G

Balanced Inventory of Desirable Responding - Version 6

Instructions: Using the scale below as a guide, circle the number beside each statement to indicate how true it is.

APPENDIX H

Offender Process Measures

UNIT: ______________________

OFFENDER: ______________________

ADMISSION DATE: ______________________

DEPARTURE DATE: ______________________

APPENDIX I

Offender Behavioural Indicators Assessment

Offender: ______________

Date: ______________

The following behaviours may be important in contributing to how well the offender does in the Intensive Support Unit. Please rate the offender's behaviour over the last two weeks for each of the following items:

APPENDIX J

Staff Expectation Questionnaire

SECTION I: INTRODUCTION

The following questionnaire is designed to determine your views regarding your role in the Intensive Support Unit. The Research Branch and Performance Assurance Sector are carrying out this study in order to evaluate how the units are functioning and the impact upon inmates and staff.

Each staff member will be asked to fill out a survey requiring approximately fifteen minutes. Most of the questions deal with your expectations of the impact of the Intensive Support Unit. It is important to note that this is not an evaluation of your performance and will have no impact on your current position. CSC is interested in your views of your role with inmates in the Intensive Support Unit. Later, surveys will also include some standard questions that have been used in other studies of the workplace.

You are under no obligation to participate in this survey. However, if you do complete the questionnaire, we ask that you answer all of the questions. After you have completed it, please place the completed survey in the attached pre-addressed envelope and seal. All of your responses will be treated with confidentiality. The research team will only handle your completed questionnaire. Under no conditions will information about individual supervisors be released to any other CSC employees.

If you have any questions or concerns about the study, please contact Stafford Murphy at (613) 536-6538.

SECTION II: SURVEY

Institution: _____________________________________

Date: _____________________________________

Directions: In the next few pages, you will find a series of statements that you might agree or disagree with. Circle the answer that best reflects how you feel about each statement. Please answer every question, as it is important to get staff input on a number of different issues involving the Intensive Support Units. Seal your completed questionnaire in the envelope provided.