Correctional Service Canada
Symbol of the Government of Canada
Skip to content | Skip to institutional links

Common menu bar links

Home > Research > Research Reports

Institutional links

Research Reports

Warning This Web page has been archived on the Web.

Accelerated Parole Review:

Were the Objectives Met?

By:
Brian A. Grant
Research Branch
Correctional Service of Canada
in co-operation with
the National Parole Board

February 1998

This report is part of a series of 24 research/evaluation reports (listed below) that were prepared as background to the Consolidated Report of the Working Group studying the provisions and operations of the Corrections and Conditional Release Act and related Consultation Paper.

The Working Group is composed for representatives from the following agencies:

Correctional Service of Canada

National Parole Board

Correctional Investigator

Justice

Department of the Solicitor General


Research/Evaluation Reports:

Information about Offenders

Security Classification of Inmates

Judicial Determination

The Temporary Absense Program: A Descriptive Analysis

Personal Development Temporary Absences

Work Release Program: How it is used and for what purposes

Day Parole: effects of the CCRA (1992)

Case Management: Preparation for Relase and Day Parole Outcome

Accelerated Parole Review

Statutory Release and Detention Provisions

Community Supervision Provisions

Provisions Relating to Victims

Observers at National Parole Board Hearings

The National Parole Board Registry of Decisions

CSC Human Resources

Administrative Segregation

Search, Seizure and Inmate Discipline

Offender Grievance System

Urinanalysis Testing Program

Inmate's Input in Decision-making

Information to Offenders

Aboriginal Offenders

Health Services

Women Offenders

Executive Summary

As part of the five year review of the Corrections and Conditional Release Act (CCRA), a study was conducted to determine if the APR provisions of the Act were effective in selecting low risk non-violent offenders for release on parole and to determine the impact of the change on the National Parole Board (NPB) and the Correctional Service of Canada (CSC).

The APR provisions of the CCRA provide for the selection of first time federal offenders who have not committed an offence listed in Schedule I (violent offences) of the Act and who have not received judicial determination of their parole eligibility with an offence on Schedule II (drug offence) of the Act. If the offender meets the APR criteria they are reviewed by the NPB for parole without the need for a hearing. Where CSC and the Board can find no evidence in the criminal and social history to indicate that the offender is likely to commit a violent offence before the completion of the sentence, then the Board must direct the release on parole. The parole criteria are different in that the Board need only satisfy the condition that the offender is unlikely to commit a violent offence, rather than the normal parole criteria which requires that the offender be unlikely to commit a new offence. In addition, no parole hearing is required if the offender is directed for release.

The study used information for all offenders released from January 1, 1990 to March 31, 1996. Offenders were divided first into pre- and post-CCRA groups based on the date of their parole hearing, then they were divided into four groups based on their eligibility and ineligibility for APR.

  • APR eligible
  • APR ineligible-offence; first time federal admission but non-eligible offence
  • APR ineligible-admission; multiple federal admissions but APR eligible offence.
  • APR ineligible-both; multiple federal admissions and non-eligible offence

The study was designed to provide answers to seven questions related to APR:

  1. Are APR cases properly identified and referred as per the CCRA?

    While problems in identifying offenders were noted following the implementation of APR, these problems were eliminated early and APR cases are being identified correctly.

  2. Are those who are eligible for APR being released at their parole eligibility date?

    The results indicated that 82% of the offenders eligible for APR were released on their parole eligibility date. The NPB did not direct the release of 18% of the APR-eligible offenders.

  3. Has APR reduced the length of incarceration for non-violent first time federal offenders?

    While the results indicate that first time, non-violent offenders are released earlier than before the introduction of APR in the CCRA, there is also evidence that these offenders are actually spending more time incarcerated than before APR was introduced. This is occurring because APR eligible-offenders are less likely to apply for day parole which could result in an earlier release, are more likely to be returned to custody and to remain in custody longer than before the CCRA.

  4. At what rate do those released following accelerated parole review recidivate as compared with other offenders?

    Compared with similar offenders prior to the CCRA, APR eligible-offenders had a higher readmission rate. However, readmission rates also increased for the comparison groups. The rate of violent reoffending for APR offenders, both pre- and post-CCRA was the lowest of the four groups in the study, and although it did increase, the increase was significantly lower than for the comparison groups.

  5. Are the more focused criteria for APR (parole directed if there is no evidence for violent offending) resulting in the release of offenders who may have a high risk of reoffending, but a low risk of reoffending violently?

    The more focused criteria for APR eligible-offenders did not result in a higher rate of reoffending, and the rate of violent reoffending remained lower for the APR-eligible offenders than for the comparison groups.

  6. Has APR resulted in a decrease in the use of institutional program resources by low risk non-violent offenders?

    This question was not addressed directly in the study, but it was assumed that if APR offenders spent less time in custody and more time in the community, then there should be a reduction in the use of resources. However, the results indicate that APR offenders spent more time in custody than similar offenders prior to the CCRA.

  7. Has APR changed the work load of CSC and the Board in reviewing low risk cases?

    It is unlikely that APR has resulted in a reduction of the work required to review cases given that day parole hearings were needed for almost half of APR eligible cases and 20% required a parole hearing because they were not directed for release at the initial in-office review. In addition, a high percentage of the offenders released would require a revocation hearing and then possibly another hearing if they again applied for parole.

The study also investigated the criteria used to select offenders for APR. Selecting offenders on the basis of first federal sentence appears to result in the selection of offenders less likely to have committed previous violent offences and less likely to commit new offences after release. In addition, the criteria of a non-violent admitting offence also resulted in the selection of offenders less likely to be readmitted, and to commit a new violent offence.

Comparisons between offenders directed for parole release and not directed for parole release indicate offenders not directed for parole release were more likely to be readmitted, to commit a new offence and to commit a new violent offence after their release than offenders directed for parole release. These results indicate that the NPB is making appropriate decisions in not directing parole release.

Acknowledgements

A study of this type requires assistance from many people whose contribution to the final product is seldom recognized by readers. R.L. Belcourt, Director, Research Information Centre, developed and provided the various data bases used for the study. Without these data bases the study would have taken considerably longer and it would not have been possible to include all cases in the analyses. Bart Millson merged, manipulated, calculated and cleaned the raw data bases in the process of developing the data we analayzed for the report. He also was responsible for doing the analyses, providing input into the selection and development of new variables and he was the sounding board for many ideas. Sara Johnson and Chris Beal coded over 2,000 criminal histories so we would have background data on provincial sentences served by offenders and follow-up results for offenders after release. Chris Beal also conducted the analyses on these data. Marlo Gal and Chris Beal reviewed a sample of cases to determine if the selection algorithm was working to confirm problems with the selection of APR cases. Finally, Laura Vandette formatted early drafts of the report and created all the tables. Without the help of each one of these people this study would not have been completed.

Table of Contents

List of Tables

List of Figures

Chapter 1 : Introduction

The Corrections and Conditional Release Act became law in November of 1992. The Act combined the former Penitentiaries Penitentiary Act (1985) and the Parole Act (1985). One of the goals of the new Act was to improve the efficiency of the federal correctional system consisting of the Correctional Service of Canada and the National Parole Board. An innovation introduced in the Act was Accelerated Parole Review, which provides for a more efficient method of reviewing low risk offenders for parole. However, it does not provide for earlier release on parole, only a change in the criteria and method used for the parole review for certain offenders.

Accelerated Parole Review was designed to ensure that low risk, non-violent, offenders would be released from prison at the earliest possible date to serve the remainder of their sentence in the community. Federal offenders must serve a minimum of one-third of their sentence in custody before being allowed to serve the remainder in the community under appropriate parole supervision. Six months prior to parole, offenders may be granted day parole that normally requires them to reside at a halfway house.

Accelerated Parole Review

All offenders serving their first federal sentence who have not been convicted of a violent or serious drug offence must have their parole eligibility reviewed by the National Parole Board using the Accelerated Parole Review (APR) process and criteria. More specifically, offenders are excluded from APR if they have been convicted of murder, any other violent offence (listed in Schedule I of the Act), or a drug offence (listed in Schedule II) with the sentencing judge setting parole eligibility at one-half of the sentence rather than at one-third of the sentence (section 741.2 of the Criminal Code, judicial determination).

Process and Criteria

The Correctional Service of Canada (CSC) must review and refer the case of any offender eligible for APR to the National Parole Board (NPB) at least three months prior to the offender's parole eligibility date (PED). The review must be based on all available information including social and criminal history, and performance and behaviour while in custody. Of particular importance is any information that suggests a potential for violent behaviour (offences listed in Schedule I) prior to the end of the sentence.

The National Parole Board conducts a file review of these cases without a hearing at least seven weeks before the eligibility date for full parole. If there is no information to indicate that the offender will commit a violent offence before the end of the sentence, the National Parole Board must direct the Correctional Service to release the offender at the full parole eligibility date. If following the file review, the Board does not direct the release of the offender, a hearing must be held for the offender before the parole eligibility date with the APR criteria used to judge eligibility for parole.

The process and the criteria are different in two ways from the normal parole review. First, there is no need for a parole hearing if the offender is to be released. The normal process requires a hearing with the National Parole Board before granting parole. The second difference is in the criteria used to determine if the offender should be released. The normal criteria for parole requires that the offender will not, by reoffending, present an undue risk to society and that release will facilitate reintegration into the community (Sec. 102 of the CCRA). The APR criteria state that the offender is unlikely to commit a violent offence prior to the end of the sentence. If there is no evidence to indicate that violence is likely, then the Board has no choice, but to direct release, even if there is an expectation that other non-violent crimes may occur.

In early 1997, changes were made to the CCRA through Bill C55. One of the amendments was to introduce Accelerated Day Parole Review which came into effect on July 3, 1997. No data are available to evaluate the effects on this provision. The changes require that APR eligible offenders be reviewed for day parole using the APR criteria. In addition, day parole eligibility is set at one-sixth of the sentence rather than six months prior to the parole eligibility date.

Review of the Corrections and Conditional Release Act (1992)

The Corrections and Conditional Release Act (1992), Section 233 (1) requires that Parliament review the provisions and operation of the Act five years after it comes into effect. As part of this review process, the Correctional Service and the National Parole Board have established a number of review projects, including the present study. An analysis of Parliamentary Committee proceedings and briefings indicated that the review should address four general themes11:

  • Public Safety and Reintegration;
  • Openness and Accountability;
  • Fair processes, Equitable decisions; and,
  • Special Groups, Special Needs.

The main focus of APR was to address public safety and reintegration. It was designed to ensure that lower risk offenders were released at the earliest possible date in their sentence to allow the Correctional Service and the National Parole Board additional time for dealing with more serious offenders.

The procedures established in the APR sections of the legislation clearly defined the roles and responsibilities for the APR process and the criteria that should be used to judge eligibility for parole release.

By providing a means for the efficient release of low risk offenders, the legislation attempts to meet the needs of offenders for whom serving time in custody may be of limited value, but for whom faster re-integration into society may serve both their needs and those of the community.

The intent of Accelerated parole review is to provide for formal recognition in law that non-violent and violent offenders should not be subject to the same conditional release process. All relevant information in the offender's criminal and social history that could bear on the offender's risk of committing a violent offence upon release is assessed. Offenders released through the APR process are supervised in the community in accordance with the risk and needs noted in their correctional plan. Any new criminal offence or breach of the conditions of release could potentially lead to a revocation of parole.

1 National Parole Board, Context for the CCRA review, 1996, unpublished report

Issues

Seven issues related to the application of the APR provisions of the CCRA were identified for review in this study.

  1. Are APR cases properly identified and referred as per the CCRA?
  2. Are those who are eligible for APR being released at their parole eligibility date?
  3. Has APR reduced the length of incarceration for non-violent first time federal offenders?
  4. At what rate do those released following accelerated parole review recidivate as compared with other offenders?
  5. Are the more focused criteria for APR (parole directed if there is no evidence for violent offending) resulting in the release of offenders who may have a high risk of re-offending, but a low risk of re-offending violently?
  6. Has APR resulted in a decrease in the use of institutional program resources by low-risk, non-violent, offenders?
  7. Has APR changed the workload of CSC and the Board in reviewing low risk cases?

Issues 1 through 5 can be addressed directly with the results that follow. Issues 6 and 7 are not addressed directly in the study. Rather, the results of the study are used to infer changes in the use of resources and workload. If offenders are released at their parole eligibility date as a result of APR, then it may be assumed that they are using fewer institutional resources prior to release. However, this analysis does not address the issue of changes in resource usage in the community. Similarly, if hearings are not needed for most of the APR cases, then it may be assumed that the workload of the Board has been reduced. However, if a large percentage of the APR cases require hearings for day parole or because they were not directed for parole, then the workload of the Board will not have been reduced.

In addition to the above issues related to APR, two broader questions about the selection of offenders for special review are addressed in the report. First is the impact of selecting only first time federal offenders for special review . Most offenders whose crimes result in a sentence requiring custody in a federal institution already have extensive criminal record. In fact, an analysis conducted while preparing for this study indicated that 54% of offenders had more than five previous criminal offences on admission to federal custody. Therefore, the question was raised as to the effectiveness of using `first federal admission' as a criteria for selecting offenders who might receive special review.

The second broader issue addressed is whether selecting offenders for special review on the basis of the type of offence committed is an effective way of identifying offenders who are unlikely to be returned to custody or to commit a new offence. Research (Andrews & Bonta, 1994) has shown that offence type, by itself, is not a very good indicator of future criminality, but the CCRA uses such a criteria for both detention and APR. The Act also does not make a distinction across levels of violence such as between a minor assault and a major assault where serious injury has occurred. If offenders are convicted of assault and receive a sentence of two years or more, they are not eligible for parole review using the APR criteria even if it is their first federal sentence.

To address these issues, it is necessary to compare the release and recidivism history of different groups of offenders. More specifically, the study provides for comparisons across time, number of federal admissions and type of offence. For analyses relating to time, comparisons are made between the pre-CCRA period and the post-CCRA period. For number of admissions, comparisons are made between first federal admission and multiple federal admissions. Finally, for type of offence, comparisons are made between offenders convicted of offences that make them eligible for APR and offenders convicted of violent or serious drug offences, which make them ineligible for APR.

Criminal history prior to a federal admission is an important factor in determining the effectiveness of APR since the selection process excludes offenders with previous federal admissions, but includes offenders without regard to the types of offences that resulted in provincial sentences. Therefore, efforts were made to determine all previous offences for a sample of offenders. In this way, it will be possible to determine if the APR selection criteria result in offenders with violent offence histories being released, or if the National Parole Board decisions result in these offenders not being directed for parole release.

Report Outline

The report is divided into four additional chapters. The next chapter presents the methodology used in developing the data for the study. In particular, it explains how cases were selected and how the comparison groups were constructed. Information is also presented on how the criminal history data were coded. The next chapter presents the descriptive data for APR cases showing the types of previous offences, previous admissions and other related information. This chapter also presents data on the time of release on full parole. The following chapter presents information on the types of release and readmissions of APR eligible-offenders and other offenders. Chapter 5 presents responses to questions raised earlier in this chapter and discusses issues more generally associated with APR.

Chapter 2: Methodology

Data Sources

The data used in the study were derived from two sources: the Offender Management System and criminal records maintained by Canadian Police Information Centre (CPIC). The Offender Management System (OMS) is an administrative data system containing all offender records. Its main purpose is to provide information to operational personnel to permit them to carry out their duties efficiently. Data from the system can be extracted for use in research analyses and this has been done for the current study. Factors extracted from OMS include admission and release dates and related information, warrant expiry dates, offence history and demographic data.

A limitation of the OMS is that it only contains information on offences for which offenders received a term in a federal prison, normally a sentence of two years of more. Alternatively, a number of offences with a consecutive sentence of more than two years, or an offence committed while under the supervision of the Correctional Service of Canada can result in the sentence being served in a federal penitentiary and inclusion in OMS. This system excludes offences for which offenders served time in a provincial jail, as normally they have sentences of less than two years.

The second source of data is the criminal history data from the Canadian Police Information Centre. This system lists all adult offences for which an offender has been convicted. Offences for which the sentence was served in a provincial institution are included in the file along with those for which only fines, probation or other sanctions were imposed. These data must be coded and summarised in a format that can be analyzed. Two university students completed the coding of CPIC information using the coding manual presented in Appendix 1. The coding provides counts of all previous offences, previous violent offences, previous sexual offences, age at first adult offence, and follow-up data for all offences.

Analysis Groups and Samples

There are 23,968 offenders included in the study. They account for all warrant of committal admissions from April 1, 1990 to March 31, 1996 and are divided into two groups: pre-CCRA and post-CCRA. Offenders who had a parole eligibility date or who had a parole decision prior to the CCRA coming into effect (November, 1992) were not eligible for consideration under the criteria for APR. Therefore, any offender who had a parole eligibility date prior to November 2, 1992 was placed in the pre-CCRA group. All other offenders were placed in the post-CCRA group since they would, in theory, be eligible for APR providing they met the APR criteria. This classification of offenders makes it possible to determine if there are differences in characteristics and outcome for offenders between the period before and after the CCRA.

An algorithm to identify APR cases was developed and tested. The algorithm provided the opportunity to identify cases in the pre-CCRA period that would have been eligible for APR in the post-CCRA period. The algorithm was tested by comparing the identified cases with decisions made by the National Parole Board in the post-CCRA period. The testing indicated that the algorithm could effectively identify APR eligible cases pre- and post-CCRA. These pre-CCRA, APR eligible-offenders provided a comparison group for determining if APR resulted in earlier release of offenders, and if the APR release process affected the recidivism rate of offenders.

To be eligible for APR, offenders must be serving their first federal sentence and must be incarcerated for a non-violent offence (not included on Schedule I of the Act). While these relatively straightforward criteria are useful for inclusion in legislation they may not be the best predictors of release outcome. More inclusive measures of recidivism risk have been developed and are in use in the Correctional Service. These measures rely on more detailed analysis of factors shown through research to be associated with recidivism. They include factors such as number of previous convictions, age at on-set of criminal activity, degree of violence used, variety of criminal offences, and type of offence based on all previous offences, not just the current offence.

To determine if the criteria used in the CCRA are appropriate, three comparison groups were developed based on type of offence and number of federal sentences served. A comparison group was created that included offenders who were serving their first federal offence, but who had been convicted of a violent offence as specified on Schedule I (Ineligible-offence).

Offenders serving their second or subsequent sentence in a federal prison are also ineligible for APR. However, many of these offenders are serving a sentence for a non-violent offence. They are ineligible for APR because of a previous federal admission and are therefore referred to as Ineligible-admissions.

A fourth comparison group includes offenders who have multiple federal sentences and who have been convicted of a violent or very serious drug offence. These offenders are referred to as Ineligible-both because they do not meet either of the APR criteria.

The classification of offenders produced eight analysis groups: four pre-CCRA and four post-CCRA groups. The four groups within the pre- and post-CCRA periods and the labels used to describe them are summarized below:

  1. APR-eligible: meets APR eligibility criteria; although only post-CCRA offenders are truly eligible for APR, pre-CCRA offenders are included for comparison
  2. Ineligible-offence: not eligible for APR because of the type of offence committed, but serving a first federal sentence
  3. Ineligible-admission: convicted of an APR eligible offence (not a violent or very serious drug offence) but this is not the first federal sentence
  4. Ineligible-both: not a first federal sentence and convicted of a violent offence or very serious drug offence

The number of cases in each of the analysis groups is presented in Table 2-1. The table also presents the basic layout of all the tables used in the study. One feature of the results in Table 2-1 is the high percentage (56%) of offenders who are first time federal admissions, but are ineligible for APR because the offence they committed is listed in Schedule 1.

To ensure that offenders were followed properly, a fixed two year follow-up period was used. However, this limited the number of offenders who could be studied because offenders with sentences of less than three years would be unlikely to have a full two year follow-up. To compensate for this, analyses were conducted with offenders who reached the end of their sentence in less than two years and the results for the two analyses are presented separately.

Results in Table 2-1 indicate that between the introduction of the CCRA and March 31, 1996 there were 3,888 offenders eligible for APR review. Of those who were eligible for APR, 82% were directed for parole release, while the release of the remaining 18% (692) was not directed, even after a hearing.

Table 2-1: Number of cases in each analysis group
APR Eligible1 Ineligible
Offence2 Admission3 Both4
Pre CCRA 1,924 2,466 602 679
Post CCRA 3,888 10,210 1,578 2,621

1Offenders released prior to the CCRA could not receive an APR, but they met the criteria as specified in the CCRA.

2 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

3 Ineligible admissions are not serving their first federal sentence.

4 Ineligible because of their offence and their admission.


In addition to analysis conducted on all admissions, a sample of cases from the four post-CCRA groups was selected for more detailed analyses of criminal history. This group of offenders is referred to as the Criminal History (CH) sample. As noted previously, complete criminal history data was obtained from the Canadian Police Information Centre (CPIC) and coded for analysis. For each of the four groups, a random sample of approximately 600 cases was selected. APR cases were over-sampled to include a larger sample of APR cases not directed for release.

This produced a sample of approximately 2500 cases including 211, or 26% of the APR group, whose release had not been directed following their APR review. The distribution of the groups is presented in Table 2-2.

Table 2-2: Distribution of cases in criminal history sample by analysis
Analysis Group Number of Cases % of Cases
APR eligible: directed 585 23.7
APR eligible: not directed 211 8.5
Ineligible - offence 575 23.3
Ineligible – admission 550 22.3
Ineligible - both 549 22.2

Analyses

Results obtained from the two data sets require slightly different interpretation. Data from the full sample include all admissions for the period under study and therefore these data represent a population, rather than a sample. Therefore, the usual statistical tests do not provide additional information for the purpose of interpretation since observed differences are real, and not estimates of population differences. In deciding whether differences between groups are important, one must rely on professional judgement as to whether the observed result is meaningful in the context. Some statistical results are presented with these data because in certain cases, a statistic like the correlation coefficient provides a useful summary of data. Results from the CH sample require statistical tests since they are derived from samples. The statistical tests do not provide an estimate of the meaningful of the observed results, but help to determine if an observed difference is likely to have occurred by chance or is a true difference in the population.

The statistics used most commonly in the report are the Chi-square statistic, which tests whether observed distributions are likely to have occurred by chance; the correlation coefficient, which provides a measure of the degree of association between variables; and the F test, which provides a test of differences between averages. All analyses were conducted using the Statistical Analysis System (SAS) version 6.11 (SAS, 1996).

Identification of APR Cases

One of the issues raised in the introduction was whether the Service properly identified APR cases. The following analyses address the identification issues to ensure that all APR cases are included in the sample.

A review of the OMS information related to APR indicated that the APR flag was unreliable. Therefore, APR cases were identified using a custom computer program that selected APR cases from all admissions. The selection program was written in SAS and used data from three data sets (admissions, offences and releases) created from OMS data by the Research Branch. These data sets contain a record of all releases and all admissions to or from CSC prisons from January 1, 1990 to March 31, 1996. In addition, all offences committed by offenders before March 31, 1996, and recorded in OMS were available in the offence data base.

The selection program used the legislated APR criteria as follows:

  • offenders who were serving their first federal sentence;
  • offenders who had not committed a Schedule I offence;
  • offenders who, having committed a Schedule II offence, had not received judicial determination (the requirement that an offender serve 50% of his or her sentence prior to being eligible for parole);
  • offenders who were not convicted of murder or serving an indeterminate sentence.

In addition, the following conditions were also used in the selection of cases and may make the number of cases differ slightly from those identified using other methods:

  • only warrant of committal admissions were used (provincial and foreign transfers excluded);
  • only admission on or after January 1, 1990;
  • the parole eligibility date was before March 31, 1996;
  • schedule I and II offences were those defined in the CCRA (1992), not those added by Bill C45, which amended the offence schedules included in the CCRA. Almost all cases with a parole eligibility between the implementation of Bill C45 (January 1996) and March 31, 1996 would have had a parole hearing prior to the implementation of Bill C45.

Based on these criteria 5,017 cases were identified as eligible for APR from a total of 26,932 admissions in the period from April 1, 1990 to March 31, 1996. Note that only offenders who had their parole hearing after the implementation of the CCRA on November 1, 1992 could be considered for APR. These 5,017 cases were then examined to determine if they had an APR decision, parole directed or a referral for a hearing. The results of this process showed a discrepancy between the number of cases identified as APR-eligible and the number of cases with an APR decision.

Overall, 22% of the cases identified as APR-eligible did not have an APR decision and 7% of the cases with a decision for APR were not identified as APR-eligible. Further analyses, including file reviews of a sample of cases, were conducted to identify the reasons for the discrepancies.

Cases with APR Decisions, but not Identified as APR-Eligible

Of the 7% (270) of cases with an APR decision but not identified as APR-eligible by the selection criteria, just over half (55%) were excluded due to data base selection criteria and other limitations (non-warrant of committal cases, offenders admitted prior to January 1, 1990, and cases where there was an error in the OMS data). The remaining cases had OMS information that would exclude them from APR eligibility (an offence on the CCRA schedules, previous admission or parole eligibility date prior to CCRA).

These errors could reflect problems in OMS data, rather than errors in decision-making. A small percentage of cases (4%) were in fact mis-identified as APR by the selection program.

Cases Identified as APR-Eligible, but Without Decision

The most frequent problem was with cases identified as APR-eligible for whom there was no APR decision on record (1,106 of 5,017, or 22%). To study the problems with these cases, a sample of 144, or 13% of the mismatches, was selected.

For each case, two research assistants reviewed the correctional plans, the sentence administration information (types and dates of offences), APR flags, and other information in OMS to determine if the cases were eligible for APR.

Of the 144 cases reviewed, it was concluded that 111, or 77%, were APR-eligible. An additional 5% had offences that were listed on Schedule I, but did not have the schedule flag set in OMS; these may have been offences prosecuted as summary offences rather than by way of indictment. For 16% of the cases it was not possible to determine if the offender was APR eligible because there was insufficient information in OMS. Most of these cases were admitted prior to January 1, 1990 or were female offenders. For cases prior to January 1, 1990 the information probably exists in paper files, but has not been entered into OMS. Data on many female offenders are missing in OMS because they frequently serve time in provincial institutions. In 3% of the cases, there was a previous admission, prior to the period covered by this study; these cases represent true errors from the APR identification program.

Analysis of the year the offender was released show that most of the errors in identifying APR cases occurred in the first year or two following the introduction of APR and these results are presented in Table 2-3. By fiscal year 1994-95, the inconsistencies between the identified APR cases and those with decisions had dropped to 7% and there was a further decline the following year to 5%.

A rate this low can be accounted for simply as mis-identification errors within the program, and are not necessarily offenders who did not receive an APR review. Remedial action has been taken to identify cases that did not receive an APR review.

Table 2-3: Distribution of cases by year of release
Year of Release
1992-93 1993-94 1994-95 1995-96
APR cases identified by research 416 1,146 1,179 975
APR cases without APR decision 236 218 184 45
Percent of cases without APR decision 57% 19% 7% 5%

These analyses suggest that there was some difficulty in identifying APR cases during the implementation of the CCRA. The file review results confirm that the program used to identify APR cases was reliable and accurate in its selection of cases. For the purposes of the study, only those cases identified as APR and who had a decision by the National Parole Board were included in further analyses.

Chapter 3: Description of APR Offenders

A description of the offenders eligible for APR is provided in this chapter along with a description of the offenders in the comparison groups. Factors that are presented include age, gender, race (Aboriginal/non-Aboriginal), sentence length, type of offences, and regional distribution. Comparisons are also presented for the APR offenders who were directed for parole release and those who were not directed for release.

Age at Admission

The average age at admission for offenders eligible for APR remained constant before and after the CCRA, at about 30 to 31 years (see Table 3-1). The offenders ineligible for APR because of their offence were very similar in age to the APR-eligible group. Offenders in the other two groups were about three years older with ineligible-admissions group being approximately 35 years at admission while the ineligible-both group was about 34 years. All of the groups were older in the post-CCRA period.

Table 3-1: Average age at admission by analysis group
APR Eligible1 Ineligible
Offence2 Admission3 Both4
Pre CCRA Average 30.4 30.3 34.5 33.1
Num. of cases 1,924 2,466 602 679
Post CCRA
Average 30.9 31.7 35.9 35.1
Num. of cases 3,888 10,210 1,578 2,621

1 Offenders released prior to the CCRA could not receive APR, but they met the criteria as specified in the CCRA.

2 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

3 Ineligible admissions are not serving their first federal sentence.

4 Ineligible because of their offence and their admission.


Gender

Males account for 96% of the APR eligible-offenders in the post-CCRA period and about 95% in the pre-CCRA period as shown in Table 3-2. Female offenders account for about 4% of the APR cases, a percentage slightly higher than their representation in the inmate population (2% to 3%). Females were most likely to be ineligible for APR because they had committed a violent or serious drug offence rather than because of multiple federal admissions.

Table 3-2: Gender distribution of offenders in the study groups
APR Eligible1 Ineligible
Offence2 Admission3 Both4
Pre CCRA Male 95.4 97.0 99.0 99.0
Female 4.6 3.0 1.0 1.0
Num. of cases 1,924 2,466 602 679
Post CCRA Male 96.5 97.4 99.3 99.0
Female 3.5 2.6 0.7 1.0
Num. of cases 3,888 10,210 1,578 2,621

1 Offenders released prior to the CCRA could not receive an APR, but they met the criteria as specified in the CCRA.

2 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

3 Ineligible admissions are not serving their first federal sentence.

4 Ineligible because of their offence and their admission.


Aboriginal/non-Aboriginal Status

Aboriginal offenders account for between 11% and 12% of the inmate population, but only 7% of the offenders eligible for APR as shown in Table 3-3. Aboriginal offenders account for 15% of the first time federal admissions who are excluded from APR because the type of offence they committed was on the CCRA schedule. These exclusions account for 67% of the Aboriginal offenders. There are no differences between the pre- and post-CCRA groups.

Table 3-3: Percentage of Aboriginal and Non-aboriginal offenders in each group
APR Eligible1 Ineligible
Offence2 Admission3 Both4
Pre CCRA Non-Aboriginal 93.0 84.1 92.2 83.5
Aboriginal 7.0 15.9 7.8 16.5
Num. of cases 1,924 2,466 602 679
Post CCRA Non-Aboriginal 93.3 84.5 92.0 84.6
Aboriginal 6.7 15.4 8.0 15.3
Num. of cases 3,888 10,210 1,578 2,621

1Offenders released prior to the CCRA could not receive an APR, but they met the criteria as specified in the CCRA.

2 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

3 Ineligible admissions are not serving their first federal sentence.

4 Ineligible because of their offence and their admission.


Sentence Length

The average sentence length for offenders in each group was about three years, with the exception of post-CCRA offenders ineligible because of their offence (4.3 years) and post-CCRA offenders ineligible because of both their offence and previous admissions (5.0 years). Detailed results are presented in Table 3-4. Post-CCRA APR eligible-offenders had an average sentence length of 3.1 years and offenders who were APR ineligible because of previous admissions had an average sentence length of 3.2 years indicating that these two groups do not differ in terms of sentence length. Offenders who were APR ineligible because of their offence had a average sentence length of 4.3 years, a full year longer than the APR eligible cases, but this large difference in sentence length did not exist prior to the CCRA. Overall there is a trend towards longer average sentences after CCRA. However, CCRA had no impact on the judicial sentencing.

Table 3-4: Average sentence length (number of years ) by group
APR Eligible1 Ineligible
Offence2 Admission3 Both4
(n) (n) (n) (n)
Pre CCRA 2.8 3.1 2.8 3.4
(1,024) (1,129) (1,033) (1,227)
Post CCRA 3.1 4.3 3.2 5.0
(1,140) (1,563) (1,166) (1,818)

1Offenders released prior to the CCRA could not receive an APR, but they met the criteria as specified in the CCRA.

2 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

3 Ineligible admissions are not serving their first federal sentence.

4 Ineligible because of their offence and their admission.


Number of Previous Convictions

The number of all offences for which offenders received a federal sentence was calculated and the results are presented in Table 3-5. Prior to the CCRA, the average number of offences did not differ across the four groups but there was an increase in the number of offences from pre-CCRA to post-CCRA for all groups. There was also a differentiation between the APR-eligible and APR-ineligible-admissions groups and the two APR-ineligible-offence groups (offence and both) after the CCRA. That is, prior to the CCRA, the APR eligible and APR-ineligible-admission groups had an average of five previous convictions with federal sentences, but after the CCRA this increased to an average of seven previous convictions. The change for the other groups was from an average of five convictions to an average of six convictions. These differences could have an impact on the length of sentence served pre- and post-CCRA. To control for the impact, the percentage of sentence served is used for comparisons. Additional information on previous convictions, including convictions with non-federal sentences is presented later in the report.

Table 3-5: Average number of convictions and scheduled offences
APR Eligible1 Ineligible
Offence2 Admission3 Both4
Pre CCRA Convictions 4.9 4.7 4.9 4.8
Scheduled offences 0.7 2.0 0.3 1.8
Post CCRA Convictions 6.9 5.8 7.3 5.7
Scheduled offences 0.9 2.8 0.5 2.5

1Offenders released prior to the CCRA could not receive an APR, but they met the criteria as specified in the CCRA.

2 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

3 Ineligible admissions are not serving their first federal sentence.

4 Ineligible because of their offence and their admission.


Table 3-5 presents the average number of Scheduled offences, that is, violent and drug offences. Offenders in the APR-eligible and APR-ineligible-admissions groups have no violent scheduled offences and a small number of scheduled drug offences, which would not exclude them from APR eligibility. The other two groups, consisting of ineligible-offence and ineligible-both, have an average of about three scheduled offences each. There is also evidence of an increase in the number of scheduled offences from the pre- to post-CCRA periods.

Type of Offence

The type of offence(s) offenders in the post-CCRA period were convicted of are presented in Table 3-6. Offenders in the APR-eligible group have offence profiles similar to offenders in the APR-ineligible-admission group. This indicates that even though the ineligible-admissions group had multiple federal admissions, they did not have a criminal record with more violent offences than the APR eligible-offenders. The main differences between these two groups was that the ineligible-admission group was more likely to have been convicted of a break and enter or theft offence. As shown in the table, APR eligible-offenders may be serving a sentence for a violent offence, but these offences would have been prosecuted by way of summary conviction rather than indictment.2 Only offences processed by the courts by way of indictment restrict offenders from APR eligibility.

Results in also indicate that offenders serving their first federal sentence and convicted of ineligible offences (mostly violent offences) are more likely to have been convicted of sexual offences (18% vs. 12%) than offenders with multiple federal sentences. Offenders who were ineligible because of the offences they committed and have multiple convictions (ineligible both) are more likely to have been convicted of robbery (51% vs. 35%) than offenders who are serving their first federal sentence.

2 Less serious violations of an offence are prosecuted as summary convictions, while indictments are used for more serious violations.
Table 3-6:Percentages of cases with a conviction for each type of offence (for post- CCRA cases only)
APR Eligible1 Ineligible
Offence2 Admission3 Both4
Type of offence
Murder 0.0 7.2 0.0 5.7
Manslaughter 0.0 6.7 0.0 4.7
Sexual offences 0.5 17.5 0.4 11.6
Robbery 0.0 35.8 0.0 51.0
Other violent offences 4.9 32.8 6.8 27.9
Drug offences 10.2 4.3 9.9 6.0
Fraud offences 8.8 2.6 10.7 3.4
Break & enter, & theft offences 40.1 20.9 62.9 27.6
Non-violent offences 48.7 30.2 27.8 31.8
Non-violent offences 48.7 30.2 27.8 31.8
Num. of cases 3,888 10,210 1,578 2,621

Note. Columns do not add to 100% because the types of offences are not mutually exclusive. That is, offenders may have more than one type of offence.

1 Offenders released prior to the CCRA could not receive an APR, but they met the criteria as specified in the CCRA.

2 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

3 Ineligible admissions are not serving their first federal sentence.

4 Ineligible because of their offence and their admission.


Regional Distribution

The Atlantic region has the highest percentage of offenders (25%) who are eligible for APR on admission, while the other regions (except Pacific) have only slightly less at 20% to 24%. The Pacific region has only 12% of admissions eligible for APR as shown in . The ineligible-offence group accounts for 60% of admissions in Ontario and the western regions while this group accounts for only 50% of new offenders in the Quebec and the Atlantic regions. The Quebec and Atlantic regions' admissions are more likely to be in the Ineligible-admission group (11% and 14%, respectively) whereas this group accounts for only about 7% of admissions in the other regions.

Table 3-7: Regional distribution of analysis groups for post-CCRA period only
APR Eligible1 Post-CCRA
Ineligible-offence2 Ineligible-Admission3 Ineligible-Both4
Atlantic % 25.0 50.6 13.5 10.9
No of cases 517 1,048 280 226
Quebec % 23.9 47.3 10.7 18.1
No of cases 1,168 2,311 524 882
Ontario % 21.6 59.8 6.7 11.9
No of cases 1,098 3,033 342 602
Prairies % 19.5 60.7 6.6 13.2
No of cases 894 2,781 303 605
Pacific % 12.5 61.6 7.7 18.2
No of cases 211 1,037 129 306
Total 3,888 10,210 1,578 2,621

1 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

2 Ineligible admissions are not serving their first federal sentence.

3 Ineligible because of their offence and their admission.


Selection for parole release

Characteristics of offenders directed for parole release following parole review using the APR criteria are presented below. Overall, 82% of eligible offenders were directed for release on parole while the remaining 18%, or 683 offenders, were not directed for release. Offenders not directed for release following APR could apply at a later date for full parole and be reviewed using standard parole criteria or they could remain in custody until their statutory release date.

Comparisons of the percentage of offenders directed and not directed for parole are presented in Table 3-8. While 82% of male offenders are directed for parole release, almost all (94%) of the female offenders are directed for release following review of their files by the NPB.

Table 3-8: Characteristics of offenders directed for parole release
Parole directed Parole not directed Total
Male 81.8 18.2 3,751
Female 94.2 5.8 137
Aboriginal 58.0 42.0 262
Non Aboriginal 84.0 16.1 3,626
Average age 31.4 28.6
Average sentence length (days) 1,163 1,032
Number of offenders 3,196 692

Aboriginal offenders are least likely to have parole directed when they are reviewed by the APR criteria. Only 58% of the Aboriginal offenders are directed for release, while 84% of all other offenders are directed.

Offenders who are directed for release are almost three years older than offenders who are not directed for release (31.4 years vs. 28.6 years). Oddly, offenders who are directed for parole release have longer sentences (1,163 days) than offenders who are not directed for release (1,032 days). The difference in sentence length is approximately four months.

The percentage of cases directed for parole varies by 20 percentage points across the regions as shown in Table 3-9. The Quebec region has the highest rate of parole directed at just 90%, while the Atlantic region has the lowest rate at about 70%. In both the Prairie and Pacific regions, approximately three-quarters of offenders eligible for release are directed, while in Ontario approximately 83% of eligible offenders are directed for release. The differences suggest that there are variations in the interpretation of the decision criteria being used to direct release on parole for these offenders. It is unlikely that the differences of the magnitude observed would occur because of different risk profiles for offenders in the different regions. In 1996/1997 the variations in the parole directed rate were reduced significantly to the point where regional rates varied by 7 percentage points or less.

Table 3-9: Regional distribution of parole not directed and parole directed cases
Parole directed Parole not directed Total
Atlantic 70.8 29.2 517
Quebec 91.4 8.6 1,168
Ontario 83.4 16.6 1,098
Prairies 77.2 22.8 894
Pacific 73.9 26.1 211
Total 3,196 692 3,888

Criminal History (CH) Sample

The CCRA states that decisions relating to APR should be based on the entire criminal and social history of the offender, not just the admission offence. Therefore, samples of the APR cases and the comparison groups were selected for analysis of their complete criminal record. Results of these analyses are presented in the following tables.

Table 3-10: Number and percentage of previous offences for which provincial sentences were served
Number of offences with provincial sentences APR Eligible1 Ineligible
Offence2 Admission3 Both4
None 19.8 24.9 4.0 4.0
1 - 5 26.3 30.6 18.9 20.0
6 - 10 13.6 15.7 16.4 19.8
11 - 20 20.4 15.8 31.3 30.7
20+ 20.0 13.0 29.5 25.5
No of cases 796 575 550 550

1 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

2 Ineligible admissions are not serving their first federal sentence.

3 Ineligible because of their offence and their admission.


Only 20% of the APR eligible-offenders had no previous convictions with provincial sentences, whereas 25% of offenders who were ineligible for APR because of the type of admitting offence had no previous provincial sentences as shown in Table 3-10. Only 4% of the other two comparison groups had no previous provincial sentences. In addition, while 54% of the APR eligible-offenders had six or more convictions with provincial sentences, only 44% of the ineligible-offence group had six or more previous convictions with provincial sentences. Three-quarters of the other two comparison groups had six or more previous convictions with provincial sentences. Clearly, being a first time federal offender, as in the case of APR eligible-offenders, is not an indicator of a lack of criminal history. However, offenders with multiple previous federal admissions have the highest number of previous offences with provincial sentences.

The APR eligible-offenders had the lowest rate of convictions with provincial sentences for violent offences at 31% while approximately 39% of ineligible-offence and ineligible-admission groups had previous violent offences with provincial sentences (see Table 3-11). Of offenders with multiple federal admissions and an admission for a violent offence (ineligible-both), 56% had previous violent offences. Of the APR eligible-offenders, 9% had been convicted of three or more violent offences for which they served provincial sentences. Of the ineligible-admissions and ineligible-offence groups, 12% had been convicted of three or more violent offences for which they served a provincial sentences.

Table 3-11:Number and percentage of violent offences (Schedule I) with provincial sentence
Number of offences with provincial sentences APR Eligible1 Ineligible
Offence2 Admission3 Both4
None 68.5 60.5 61.5 43.6
1 - 2 22.5 27.1 27.1 37.8
3 or more 9.0 12.4 11.5 18.5
Number of cases 796 575 550 550

1 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II offences (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

2 Ineligible admissions are not serving their first federal sentence.

3 Ineligible because of their offence and their admission.


Comparisons of the number of provincial offences between the parole directed and parole not-directed groups are presented in . These results indicate clear differences between the groups, with the parole directed group having fewer offences overall and substantially fewer violent offences. For example, while almost one-quarter of offenders directed for release had no previous offences only one-eighth of the not-directed offenders had no previous offences. Similarly, while 80% of the directed offenders had no previous violent offences, only 40% of the not-directed group had no previous violent offences. A more outstanding difference between directed and not directed offenders is the percentage of offenders who had three or more violent offences. Only 4% of the directed group had three or more previous violent offences, while six times that many (24%) of the not-directed group had three or more previous violent offences.

Table 3-12: Number of offences with provincial sentences and number of violent offences (Schedule I) for APR eligible-offenders directed and not directed for release
Number of offences with provincial sentences Parole directed Parole not directed
None 22.4 12.8
1 - 5 29.2 18.0
6 - 10 13.3 14.2
11 - 20 19.8 21.8
20+ 15.2 33.2
Number of violent offences(Schedule I) with provincial sentences
None 79.3 38.4
1 - 2 16.9 37.9
3 - 10+ 3.8 23.7
Total number of cases 585 211

Comparisons were also made across the four study groups for drug offences with provincial sentences. Results of these comparisons indicated there were minimal differences in the number of previous drug offences. In addition, the parole directed and not-directed groups did not differ on the number of previous drug offences.

Criminal History Risk

Two measures of offender criminal history risk were used in the study, the SIR (Statistical Information on Recidivism) scale (Nuffield, 1982) and the CHR (Criminal History Risk). In Table 3-13 the full distribution of SIR scores in the post-CCRA period is presented.

However, the results were analysed using only two levels of risk (high and low) since having many levels make the number of offenders in each category too small to allow for meaningful analyses.

Table 3-13:SIR score level for eligible and non-eligible offenders post-CCRA only
Risk level APR Eligible1 Ineligible
Offence2 Admission3 Both4
Post-CCRA
Low (4/5 will not reoffend) 42.3 39.4 10.1 9.3
Low-medium (2/3 will not reoffend) 11.0 15.5 7.9 6.8
Medium (1/2 will not reoffend) 11.4 16.7 11.2 19.8
High medium (2/5 will not reoffend) 11.7 12.3 15.0 19.8
High (1/3 will not reoffend) 23.4 16.2 55.8 44.3
Number of cases 1,571 1,960 446 474

Note: The SIR scale was available for less than one-third of offenders.

1 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II offences (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

2 Ineligible admissions are not serving their first federal sentence.

3 Ineligible because of their offence and their admission.


As can be seen in Table 3-13, approximately 42% of the APR eligible-offenders were rated as low risk to reoffend. The ineligible-offence group had a similar percentage of low risk offenders (39%) while only 10% of offenders in the other groups were in the low risk category. Based on the SIR scale, 23% of the APR eligible-offenders were at higher risk to reoffend generally while only 16% of the ineligible-offence group were in the high risk category. Approximately 50% of the offenders in the other groups, ineligible-admissions and ineligible-both, were rated as high risk to reoffend, and over 80% of offenders in these groups were classified as medium to high risk by the SIR scale.

Given that the SIR scale was not available for a large percentage of offenders in the post-CCRA period, and was unavailable for most offenders in the pre-CCRA, a comparison of the risk ratings provided by the two risk scales is provided in Table 3-14. Analyses in the remainder of the report use the Criminal History Risk (CHR) scale developed by Robinson (1995).

The percentage of offenders who were low risk on the basis of the SIR and CHR scales are presented in Table 3-14. This table allows for comparisons of the SIR and CHR scales in the post-CCRA period, and the CHR scores in the pre- and post-CCRA periods. Both scales indicate that slightly more than half (54% and 56%) of the APR eligible-offenders are low risk for reoffending in the post-CCRA period. Both scales also indicate that the ineligible-offence group is of no higher risk to reoffend than are the APR eligible-offenders. However, the two scales differ on their classification of ineligible-offence offenders in the post-CCRA period with the SIR indicating that 55% are low risk and the CHR indicating 73% are low risk.

Table 3-14: Criminal history risk (SIR scale and CHR scale) for offenders released pre and post-CCRA
Risk level APR Eligible1 Ineligible
Offence2 Admission3 Both4
Pre-CCRA
SIR-Low n/a n/a n/a n/a
CHR-Low 54.1 71.4 30.6 45.8
Number of cases 1,918 2,443 601 655
Post-CCRA
SIR-Low 53.5 54.9 17.9 16.0
Number of cases 1,517 1,960 446 474
CHR-Low 55.7 72.9 28.6 40.4
Number of cases 3,701 5,308 1,153 1,203

1 Offenders released prior to the CCRA could not receive an APR, but they met the criteria as specified in the CCRA.

2 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

3 Ineligible admissions are not serving their first federal sentence.

4 Ineligible because of their offence and their admission.


The CHR tends to rate more offenders who had multiple federal admissions as low risk than the SIR scale does, probably because the SIR scale uses more information on the offenders' criminal history. The CHR may not be as accurate a measure of criminal history risk as the SIR for groups with higher proportions of high risk offenders.

Comparisons between pre- and post-CCRA indicate that approximately the same percentage of offenders were rated as low risk at both times by the CHR scale. This would imply that release patterns should have been similar for the groups pre- and post-CCRA and that differences were likely the result of changes in legislation or other factors, not differences in the characteristics of the offenders.

Of the APR eligible-offenders directed for release, 57% had a SIR score in the low range while only 36% of the offenders not directed for release had a low SIR score. Looked at slightly differently, 90% of the offenders rated as low risk to reoffend were directed for release while in comparison 79% of the offenders at a high risk to reoffend were not directed. The APR criteria set out in the legislation, and used by the National Parole Board, takes into account risk to reoffend violently, not simply risk to reoffend. Therefore, the high percentage of cases with a high risk to reoffend that were directed for release is not surprising.

Summary and Discussion

A number of factors were used to compare the APR-eligible and ineligible groups. These factors included age, race, sentence length, previous offences with federal sentences, previous offences with provincial sentences, previous violent offences and offender risk level as determined by the SIR scale and the CHR. The comparisons indicated differences in offenders from the pre- to post-CCRA periods, with increased criminality (more offences and more serious offences) in the post-CCRA period. The results also indicated, that on many of the variables, the APR group did not differ significantly from either the ineligible-offence or ineligible-admissions groups, but generally these three groups did differ from the ineligible-both group. However, the APR group showed similarities to the ineligible-offence and ineligible-admissions groups on different factors, thus reducing the clarity of the findings.

The APR eligible-offenders were similar in age at admission to the ineligible-offence group (approximately 30 years of age), but these two groups of offenders were younger than the ineligible-admissions and ineligible-both groups, who were three to four years older. The results also indicated that there was a slight increase in age from pre- to post-CCRA.

Analyses of gender differences indicated that similar percentages of male and female offenders are eligible for APR. However, racial comparisons indicated some differences between Aboriginal and non-Aboriginal offenders, with only 7% of APR eligible-offenders being Aboriginal. Analyses for the Aboriginal offenders indicated that 67% of these offenders were ineligible because they had committed an offence from Schedule I of the CCRA (a violent offence) at the time of their first federal admission.

Significant differences were observed in the proportion of offenders in each region who were APR eligible. In the Atlantic region, one quarter of the admissions were eligible for APR whereas in the Pacific region only 12% of new offenders were eligible for APR. Differences in the rate at which parole was directed for eligible offenders were also noted. While in the Atlantic, Pacific and Prairie regions approximately one-quarter of the APR eligible cases were not directed for parole, in Quebec region only 9% were not directed for parole release. The rate of not directing parole in the Ontario region was 17%. These results suggest differences in the admissions to each region which need to be taken into account when studying APR, but they also suggest that the decision process for APR may vary across the regions. Given an APR-eligible offender, it seems problematic that one region would not direct release at a rate three times higher than in another region (9% in Quebec and 29% in Atlantic). Interestingly, the Atlantic region also had the largest proportion of offenders eligible for APR.

Prior to the CCRA, the average sentence length for offenders was approximately three years, but this increased after the CCRA. The largest increases were for the ineligible-offence and ineligible-both offender categories, with increases in sentence length to four and five years, respectively. In the post-CCRA period, APR eligible-offenders had an average sentence length similar to offenders in the ineligible-admissions group.

Analyses of the average number of offences with federal sentences indicated that there was an increase in the average from the pre-CCRA period to the post-CCRA period, rising from about five to between six and seven offences. The largest increase in the average number of offences was for the APR-eligible and ineligible-admissions groups, and these two groups also had the highest average number of offences at about seven. There was also an increase in the number of violent and drug offences (Schedule I and Schedule II) within the four comparison groups.

In addition to looking at the number of offences for which federal sentences were served, analyses were conducted using the number of offences with provincial sentences. APR offenders were more likely to have served sentences in provincial institutions and have a higher number of convictions. While 32% of the APR eligible-offenders had committed violent offences for which they served provincial sentences, approximately 40% of the ineligible-offence and ineligible-admissions groups had served provincial sentences for violent offences, which does not constitute a large difference. Overall, these results lead to the conclusion that the APR eligible-offenders have more previous offences than other offenders, but they have a slightly lower number of violent offences. However, they are not violence free and the rate of violent offences is not dramatically lower than the rate for offenders in the comparison groups.

Risk of reoffending was measured using the standardized SIR scale and a custom developed scale, the CHR. Both of these scales measure risk on the basis of criminal history variables. To accurately predict risk, it is optimal to have both criminal history risk measures and measures of criminogenic needs, however, the latter information was not available for the study groups. The risk analyses indicated that over half of the offenders in the APR eligible group were low risk, but over half of the ineligible-offence group were also rated as low risk. The risk level of the other two comparison groups was significantly higher. Interestingly, unlike the observed increase in the number of offences and the increase in the number of violent offences, the risk measures did not indicate an increase in level of risk for the offenders between the pre- and post-CCRA periods.

Analyses were conducted to determine if the parole directed offenders were different from the not directed offenders. These analyses indicated that female offenders were more likely than males to have parole directed and Aboriginal offenders were less likely than non-Aboriginal offenders to have parole directed. Offenders with parole directed were younger (on average) and oddly, the average sentence length of offenders with parole directed was longer than for offenders with parole not directed. Parole directed offenders had fewer previous offences and were much less likely to have had a previous violent offence. Specifically, 79% of parole directed offenders had no previous violent offences while only 38% of the not directed offenders had no previous violent convictions. Finally, as expected, the parole directed offenders were rated as lower risk to reoffend than offenders who were not directed for release. However, even among the directed group, 43% were rated as high risk by the criminal history measures.

The results suggest that selecting offenders for special treatment because they are first time federal offenders may be a reasonable approach. In a number of comparisons the first time federal offenders (APR eligible and ineligible-offence groups) were quite similar, whereas they differed from the offenders with multiple admissions. However, there were some comparisons in which offenders who had multiple admissions were also similar to the APR eligible-offenders.

Chapter 4: Release and Readmission

The purpose of this chapter is to determine how well offenders released on APR did after their release in comparison to offenders who were not eligible for APR. More specifically, the results in this chapter look at the type of release offenders received, the time of release within the sentence, the percentage of the sentence completed at the time of the release, recidivism rates (readmission, technical violations, new offences and new violent offences) and the relationship between offenders' risk level and recidivism. Comparisons are also made between the APR eligible-offenders directed for release and offenders not directed for release.

Type of Release

Prior to the CCRA, 68% of APR eligible-offenders were granted a day parole as their first release and 15% received full parole as a first release (Table 4-1). After the CCRA, only 44% of APR eligible-offenders were granted day parole as their first release, while 48% were granted full parole as their first release. That is, after the CCRA offenders were more likely to wait until their parole eligibility date for release, rather than taking an earlier release on day parole. Another difference between pre- and post-CCRA periods is the decline in the percentage of APR eligible-offenders who remained in custody until their statutory release date. Prior to the CCRA, 16% of APR eligible-offenders remained in custody until their statutory release date, but after the CCRA only 7% remained in custody until their statutory release date.

Table 4-1: Type of first release pre and post CCRA
APR Eligible1 Ineligible
Offence2 Admission3 Both4
Pre CCRA Day parole 68.1 55.5 56.4 45.0
Full parole 14.6 9.0 7.0 4.7
Statutory release 15.8 28.7 33.8 41.2
Other 1.5 6.8 2.8 9.0
Number of cases 1,918 2,443 601 655
Post CCRA Day parole 44.1 56.0 57.7 46.7
Full parole 48.3 7.1 4.8 4.0
Statutory release 7.3 31.0 35.9 41.6
Other 0.2 5.9 1.7 7.7
Number of cases 3,701 5,308 1,153 1,203

1 Offenders released prior to the CCRA could not receive an APR, but they met the criteria as specified in the CCRA.

2 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

3 Ineligible admissions are not serving their first federal sentence.

4 Ineligible because of their offence and their admission.


The comparison groups did not show the same changes in type of release from pre- to post-CCRA, and instead were very consistent across the time periods. Approximately 56% of offenders in the ineligible-offence and ineligible-admission groups received day parole as their first release, while another 30% of these offenders were released at their statutory release date. The percentages were slightly different for offenders in the ineligible-both group (multiple federal admissions and non-eligible offence) with 46% released on day parole and 41% released at their statutory release date.

While only 7% of the APR eligible-offenders had statutory release as their first release, 25% had a statutory release during their sentence. This would imply they were returned to custody from a conditional release and then were released at their statutory release date. Prior to CCRA, 48% of APR eligible-offenders were eventually released on SR, although only 16% had statutory release as their first release.

APR eligible-offenders who were not directed for release were about equally likely to be released first on day parole (45%) or at their statutory release date (50%). However, 81% of these offenders were eventually released at their statutory release date indicating that most of those granted day parole were either readmitted during day parole, or the day parole period immediately preceded their statutory release date.

Time to release

Prior to the CCRA, offenders who would have been eligible for APR were released, on average, 114 days after their parole eligibility date, but after the implementation of the APR selection criteria, the same type of offenders were released, on average, within 15 days of their parole eligibility date, a decrease of 86%. The time from parole eligibility to parole release declined for the comparison groups, the decline ranged from a 3% to 15% decrease and that is relatively minor compared to that for the APR eligible-offenders. Note that these times are based only on cases that were released on parole. Many offenders in the comparison groups may not have been released on parole, as were some of the APR eligible-offenders who were held until their statutory release date.

Table 4-2: Average time (in days) from parole eligibility date to release of full parole
APR Eligible1 Ineligible
Offence2 Admission3 Both4
Pre-CCRA
Average time (in days) 115 173 149 241
Number of cases 178 962 823 185
Post-CCRA
Average time (in days) 15 158 144 205
Number of cases 2,723 1,485 242 178

Note: most cases are released at parole eligibility date

1 Offenders released prior to the CCRA could not receive an APR, but they met the criteria as specified in the CCRA.

2 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

3 Ineligible admissions are not serving their first federal sentence.

4 Ineligible because of their offence and their admission.


Another measure of how APR affected the release of offenders is the proportion of their sentence served until their first release. Prior to CCRA, APR eligible-offenders served approximately 39% of their sentence before being released. After the CCRA this decreased to 34%. The comparison groups did not show a change from pre- to post-CCRA in terms of the percentage of time served with the ineligible-offence and ineligible-admissions groups serving less than 50% of their custody until their first release and the ineligible-both group serving more than 50% of their sentence in custody. In terms of percentage of time served to first release, the APR-eligible group pre- and post-CCRA served less of their sentence prior to release than the comparison groups. These results are presented in Table 4-3.

Table 4-3: Percentage of sentence served at time of first release
APR Eligible1 Ineligible
Offence2 Admission3 Both4
Pre-CCRA
Average time (in days) 115 173 149 241
Number of cases 178 962 823 185
Post-CCRA
Average time (in days) 15 158 144 205
Number of cases 2,723 1,485 242 178

1 Offenders released prior to the CCRA could not receive an APR, but they met the criteria as specified in the CCRA.

2 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

3 Ineligible admissions are not serving their first federal sentence.

4 Ineligible because of their offence and their admission.

Post-release Outcome

Four measures of post-release outcome were used: readmissions, violations of release conditions, new offences and new violent offences. Counts and percentages for the measures are not mutually exclusive so an offender who committed a violent offence is included in the readmission, new offence and new violent offence measures. Two different follow-up periods were explored:

  1. Variable follow-up within two years of release (includes APR eligible-offenders whose sentence expired within two years of their release);
  2. Fixed follow-u of two years (excludes APR eligible-offenders with shorter sentences, less than three years).

If offenders committed a new offence after the two year period it was not included in the results.

Variable follow-up Within two Years of Release

As can be seen in Table 4-4, APR eligible-offenders (with a variable follow-up) released after the CCRA were more likely to be readmitted, have their conditional release revoked for technical violations, and commit new offences than similar offenders released prior to the CCRA. While 23% of pre-CCRA APR eligible-offenders were readmitted to prison within two years of their release, 39% were readmitted after the CCRA was implemented. The percentage of offenders with technical violations almost doubled from 15% to 28% while the percentage with a new offence increased from 11% to 14%. Increases were also noted for the comparison groups, although the size of the increase was larger for the APR group. Also noteworthy among APR eligible cases was the decline in new violent offence from 2.4% pre-CCRA to 2.0% post-CCRA.

Table 4-4: Type of readmission within two years following release
APR Eligible1 Ineligible
Offence2 Admission3 Both4
Pre-CCRA
Readmission 23.0 29.2 42.2 43.1
Technical violation 14.6 20.8 29.7 30.1
New offence 10.7 11.2 17.3 18.2
New violent offence 2.4 4.3 2.0 4.4
Number of cases 1,181 1,663 398 455
Post-CCRA
Readmission 38.7 32.6 49.6 44.8
Technical violation 28.4 25.4 35.7 33.3
New offence 14.1 10.4 19.8 13.9
New violent offence 2.0 2.8 4.6 4.0
Number of cases 1,392 2,316 565 547

1 Offenders released prior to the CCRA could not receive an APR, but they met the criteria as specified in the CCRA.

2 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

3 Ineligible admissions are not serving their first federal sentence.

4 Ineligible because of their offence and their admission.


Most of the increase in readmissions for APR offenders can be attributed to the increase in technical violations. While the other groups showed increases of around five percentage points from pre- to post-CCRA, these increases were small compared to that observed for the APR group.

Fixed follow-up Two years

Using a follow-up period of two years, APR eligible-offenders released after the CCRA were more likely to be readmitted, have their conditional release revoked for technical violations, and commit new offences than similar offenders released prior to the CCRA. While 32% of pre-CCRA APR eligible-offenders were readmitted to prison within two years of their release, 43% were readmitted after the CCRA was implemented. The percentage of offenders with technical violations more than doubled from 7% to 15% while the percentage with a new offence increased from 25% to 32%. There was a small increase in the percentage of offenders who committed new violent offences from 4.1% to 4.7% (a 15% increase), but this increase was much lower than the comparison groups.

Using this fixed-time period methodology, rates of readmission, technical violations, new offences and new violent offences increased for the comparison groups from the pre- to post-CCRA period. Again, relative to the comparison groups, the APR group had the largest increase in technical violations of conditional release.

Table 4-5: Type of readmission (within two years following release)
APR Eligible1 Ineligible
Offence2 Admission3 Both4
Pre-CCRA
Readmission 31.5 35.7 52.7 52.7
Technical violation 7.2 13.7 16.8 22.6
New offence 25.4 24.9 40.9 35.2
New violent offence 4.1 10.3 6.4 15.1
Number of cases 737 775 203 199
Post-CCRA
Readmission 42.6 49.2 66.0 66.6
Technical violation 14.8 23.0 20.9 25.2
New offence 31.8 30.0 52.0 46.4
New violent offence 4.7 17.0 11.1 25.2
Number of cases 1,096 1,495 344 388

1 Offenders released prior to the CCRA could not receive an APR, but they met the criteria as specified in the CCRA.

2 Ineligible offences are those listed on Schedule I (violent offences) and Schedule II (drug offences) for which the judge orders eligibility for parole at 50% of the sentence rather than at 33%.

3 Ineligible admissions are not serving their first federal sentence.

4 Ineligible because of their offence and their admission.


Time in custody

Two measures of the effect of APR on correctional operations are the percentage of time in custody from release into the community and the percentage of time in custody for the entire sentence. These measures are calculated using percentage of time, rather than actual time, in order to control for the increased length of sentences in the post-CCRA period. The percentage of time in custody from release indicates the amount of time offenders spent in custody after failing on their conditional release and is a measure of how effective their reintegration was. The need to spend longer time in custody after the conditional release may indicate that offenders were not ready for release. The percentage of the entire sentence spent in custody takes into account time at release on day parole and time spent after the offender was returned to custody.

Prior to the CCRA, APR eligible-offenders spent, on average, 11% of their available community time in custody after their full release. After the CCRA, they spent about 18% of their available community time in custody. However, as can be seen in Figure 4-1, all of the comparison groups showed an increase in post-release time in custody with the percentage change being approximately the same for all groups except the ineligible-both group, which had only a 17% increase. All of the ineligible groups spent more time in custody after their first full release, although the difference between the APR eligible and ineligible-offence group was minimal (11% vs. 14% pre-CCRA and 18% vs. 22% post CCRA). In more concrete terms, APR eligible-offenders, prior to the CCRA, spent an average of 82 days in custody after their release while they spent 115 days in custody after the CCRA, a difference of 33 days or one month.

Figure 4-1:Percentage of time in custody (days) from the time of full

While the percentage of time in custody after release reflects the offender's response to reintegration into the community, an important measure for the Service is the percentage of time in custody for the entire sentence. The results indicate that prior to the CCRA, APR eligible-offenders spent, on average 43% of their sentence in custody, while after the CCRA they spent, on average, 46% of their sentence in custody, an increase of 7%. Increases in the average amount of time in custody were evident for all groups after the CCRA, ranging from 8% to 12% for the comparison groups. The increased time in custody for the APR eligible-offenders results from an increase in the average time in custody after release (returns to custody) and a decrease in the use of day parole by APR eligible-offenders.

Figure 4-2: Percentage of sentence served in custody

National Parole Board selection

The preceding sections present the return rates for all offenders eligible for APR. However, not all APR eligible-offenders are released. The National Parole Board must make a decision whether to direct parole or not direct parole, in the latter cases the Parole Board must have evidence to suggest that the offender is likely to commit a violent offence upon release. In addition to looking at the overall return rates for APR eligible-offenders, an analysis was conducted to determine how effective the National Parole Board was in selecting the most serious offenders.

Almost all (98%) of the offenders directed for release on parole were released on full parole. The small percentage that was not released committed an offence while on day parole or while in the institution awaiting parole release. However, only 19% of offenders not directed for release were ever released on full parole, with most (81%) released at their statutory release date, and some with a preceding day parole. In addition, offenders directed for release served 30% of their sentence prior to their release, while those not directed served an average of 53% of their sentence prior to their release.

While 41% of the offenders directed for release by the National Parole Board were readmitted within two years of their release, 66% of offenders not directed for release were readmitted. Similarly, 31% of directed offenders committed a new offence within two years of their release, and 44% of the not directed offenders committed a new offence. In terms of violent offending, the rate of violent offending for the not directed group was almost triple that of the directed group (11% vs. 4%). Readmission and other recidivism measures are presented in Table 4-6.

Table 4-6: Percentage of outcomes by APR decision for two year follow up
APR Eligible
Parole directed Parole not directed
Readmitted 40.7 65.9
Technical violation 13.9 25.9
New offence 30.9 43.9
New violent offence 4.1 11.0
Days to new violent offence 190 140
Number of cases 1,014 82

Additional results on new offences were obtained by looking at all offences, including both provincial and federal sentences. Of the parole directed offenders, 38% committed a new offence within one year of their release, while 69% of the offenders not directed for release committed another offence. Likewise, for violent offences, 7% of directed offenders committed a violent offence after release while three times as many non-directed offenders (21%) committed a new violent offence. Clearly, the offenders who were not directed for release were more likely to commit new offences and were more likely to commit violent offences.

Analyses of time to failure for parole directed and non-directed offenders indicates that not only were the not directed offenders more likely to be readmitted and to be convicted of new offences, they also returned more quickly. The full results are presented in Table 4-7. While the average time to recidivate for directed offenders was around 200 days (7 months) it was less than 160 days (5 months) for offenders not directed for release.

Table 4-7: Time to failure for APR eligible-offenders within two years of release
APR Eligible
Parole directed
(n1)		 Parole not directed
(n1)
Days to readmission 234 151
(413) (54)
Days to new offence 219 159
(313) (36)
Days to new violent offence 190 140
(42) (9)

1 (n) are the number of cases with a readmission, new offence or new violent offence


Summary and Discussion

Compared to a similar group of offenders prior to the CCRA, the group of offenders who were eligible for APR showed a decrease in the use of day parole, a decrease in the number of days to first release, and a decrease in the percentage of the sentence served at the time of their release. They also demonstrated, on average, an increase in the percentage of time spent in custody throughout their sentence and this increase was evident for both low and high risk offenders. In addition, APR eligible-offenders (post-CCRA) were more likely to have their conditional release revoked for failing to meet the conditions of the release, for committing a new offence, and committing a new violent offence. However, increases in negative outcomes were also observed for the comparison groups from pre- to post-CCRA. The results also indicated that the offenders that the National Parole Board did not direct for parole release were the most likely to recidivate when they were eventually released, suggesting effective selection of problem cases by the Board.

The analyses revealed that there was a significant decrease in the number of APR eligible-offenders who had day parole as their first release and a consequent increase in the number of offenders who had full parole as their first release. These results suggest APR eligible-offenders chose not to apply for day parole (National Parole Board decisions for day parole also decreased after the CCRA), accepting instead full parole release at one-third of their sentence. This is unfortunate because day parole is a very effective form of early release which provides a structured environment in a halfway house for offenders on their re-entry into the community. Another effect of the decline in day parole use for APR eligible-offenders is that before the CCRA, these offenders had the highest percentage of day parole use, but after the CCRA, offenders in the ineligible-offence and ineligible-admissions groups were more likely to have been granted day parole as their first release. However, modifications to the CCRA in Bill C-55 (1997) ensures that APR eligible-offenders who are directed for release are first released on day parole at one-sixth of their sentence, rather than on parole at one-third of their sentence.

The results indicate that APR had a positive effect on the number of offenders released at their statutory release date. That is, the percentage of APR eligible-offenders whose first release occurred at their statutory release date decreased from 15% to 7% and for any release, from 50% to 25%. These comparisons are between similar offenders in the pre- and post-CCRA periods and indicate that fewer APR eligible-offenders were released at the statutory release date after the CCRA.

The results clearly indicate that the APR legislation resulted in earlier release for APR eligible-offenders. On average, APR eligible-offenders were released almost 100 days earlier post-CCRA, and they were released after serving an average of 13% less of their sentence.

Overall, there was an increase in the return to custody rates (readmission, violation of conditions, new offence and new violent offence) for all offender groups in the study from the pre- to post-CCRA period for both the less than two year and two year follow-up groups. However, the results for the two year follow-up show the increase in return to custody was greatest for the APR eligible-offenders. From pre- to post-CCRA, the readmission rate and the rate of violation of release conditions almost doubled for the APR eligible-offenders. The new offence rate for APR eligible-offenders also increased and was higher, in the post-CCRA period, than that of offenders in the ineligible-offence group. Violent offending for APR eligible-offenders also increased by 15%, although in absolute terms the increase was only from 4% to 5% and was low compared to the other groups. The increase in violent re-offending was higher for the comparison groups. Results for offenders without a full two year follow-up period indicate that technical violations were the major cause of the higher post-CCRA admission rate. While the new offence rate increased slightly for these offenders, the new violent offence rate remained virtually unchanged for all groups.

Results for APR eligible-offenders indicated that after the CCRA there was an increase in the time in custody after their initial release. That is, there was an increase in the time offenders spent incarcerated after failing while on conditional release, an increase of an average of 32 days per offender. In addition, there was also an increase in the percentage of their sentence served in custody for APR offenders, from 43% to 46%. This result needs to be viewed in light of the earlier releases these offenders had (at about one-third of their sentence) compared to similar offenders in the pre-CCRA period. The increased days in custody occurs because of fewer day parole releases that allow the offender to be released earlier in the sentence, increased returns to custody, and longer periods in custody for some offenders following a return. The results also indicate increased time in custody for offenders in the comparison groups from pre- to post-CCRA.

Results indicate that the decisions by the National Parole Board not to direct parole release for some offenders were supported by the outcomes for these offenders. Offenders who were not directed for parole release were more likely to remain in custody until their statutory release date, on average they served over half of their sentence before release, and two-thirds were readmitted within two years of release. Offenders not directed for release were also more likely to commit a new offence and a new violent offence while released.

Chapter 5: Discussion

The goal of Accelerated Parole Review, to ensure low risk offenders are released from custody early, appears to have been achieved with APR eligible-offenders being released earlier than similar offenders prior to the implementation of the CCRA. However, the evidence also suggests that many of these early releases result in readmission while in the community and often a return to custody. In fact, among the group of APR eligible-offenders in this study, there was an overall increase in the average amount of time spent in custody after the CCRA rather than a decrease. In addition, the readmission, violation of conditional release and new offence rates increased for the APR eligible-offenders in the post-CCRA period.

Answers to questions

The introduction included a set of seven questions to be answered from this research. These questions are presented below and answers are provided. The results of the study are discussed further after the responses to the questions.

  1. Are APR cases properly identified and referred as per the CCRA?

    Analyses indicated that when the CCRA was first implemented in 1992-93 there were some problems in identifying offenders who were eligible for APR, but these problems were largely eliminated within a year. The results indicate that APR cases are now being identified correctly and the National Parole Board is making appropriate decisions applying the legislative criteria.

  2. Are those who are eligible for APR being released at their parole eligibility date?

    The results indicated that 82% of the offenders eligible for APR were released on their parole eligibility date. In addition, 44% were released before their parole eligibility date on day parole. While the remaining 18% were not directed for release . The average percentage of sentence served prior to release for offenders directed for parole release was approximately 30%, less than one-third of their sentence because of day parole releases.

  3. Has APR reduced the length of incarceration for non-violent first time federal offenders?

    While the results indicate that first time, non-violent offenders are released earlier than before the introduction of APR in the CCRA, there is also evidence that on average this total group are actually spending more time incarcerated than before APR was introduced. This appears to be occurring for two reasons. First, APR eligible-offenders are less likely to apply for a day parole, which can permit their release six months before their parole eligibility date. Prior to the CCRA over two-thirds of offenders who met the APR eligibility criteria received a day parole release, but after the CCRA this declined to less than half. The second reason for increased incarceration is the increased rate of readmission following release among APR eligible-offenders. These readmissions resulted in an increase in the number of offenders returned to custody and therefore, an increase in the average amount of time incarcerated.

  4. At what rate do those released following accelerated parole review recidivate as compared with other offenders?

    Compared with similar offenders prior to the CCRA, the readmission rate of APR eligible-offenders increased. However, readmission rates for the comparison groups also increased, indicating that there was a general increase in readmissions for all offenders. Using a variable length follow-up period, the largest increase for APR offenders was in the area of violations of parole conditions, increasing from 14.6% to 28.4%. While there was a lower increase for new offences (10.7% to 14.1%), the rate of violent reoffending for APR eligible-offenders, both pre- and post-CCRA was the lowest of the four groups in the study.

  5. Are the more focused criteria for APR (parole directed if there is no evidence for violent offending) resulting in the release of offenders who may have a high risk of reoffending, but a low risk of reoffending violently?

    The increase in the rate of readmission with a new offence from pre- to post- CCRA was similar for the APR eligible and ineligible offenders. Therefore, the more focused APR criteria did not result in a proportionately higher increase in new offences for these offenders. In addition, the rate of new violent offending was lowest for the APR eligible-offenders, and remained considerably lower than for the comparison groups (5% vs. between 11% and 25%).

  6. Has APR resulted in a decrease in the use of institutional program resources by low risk non-violent offenders?

    This question was not addressed directly in the study, but it was assumed that if APR offenders spent less time in custody and more time in the community then there should be a reduction in the use of resources. However, the results indicated that APR offenders, on average, spent more time in custody than similar offenders prior to the CCRA did. Therefore, it would appear that the resources they used would be the same or greater than the resources utilized by similar offenders prior to the CCRA.

  7. Has APR changed the work load of CSC and the Board in reviewing low risk cases?

    It is unlikely that APR has resulted in a reduction of the work required to review APR cases. Normally, all offenders who apply for parole have a hearing with members of the NPB. The APR legislation allows the Board to review the offenders documentation without the need for a hearing. If there is no evidence that the offender is likely to commit a new violent offence, then the Board directs the offenders to be released. If the Board decides the offender is likely to commit a violent offence then a hearing must be held. The reduction in workload was supposed to result from the reduced number of hearings needed for these potentially low risk offenders.

Given that almost half of APR offenders applied for day parole, and an additional 20% were not directed for release and would require a parole review with a hearing, there would not have been a dramatic reduction in the number of hearings for release. In addition, those offenders readmitted following release would require a revocation hearing and then possibly an additional hearing if they again applied for parole.

Prior to the CCRA, case management prepared documentation for both day parole and parole review at the same time since day parole review was automatic and a parole review would be done at that same hearing. After the introduction of APR, case managers in the institutions were required to prepare the same reports for the APR process, and then to update the reports for a later parole hearing if the offender was not directed for release. While there may not have been an increase in the work load for case management, it is unlikely that there was a significant decrease.

References

Andrews, D.A., and Bonta, J. (1994). The Psychology of Criminal Conduct.
Cincinnati: Anderson

Nuffield, J. (1982). Parole Decision Making in Canada. Ottawa: Communication
Division. Solicitor General Canada.

Parole Act, RSC, C-2 (1985).

Penitentiary Act, RSC, C-5 (1985).

Robinson, D. (1995). The impact of Cognitive Skills Training on Post-Release Recidivism
among Canadian Federal Offenders, R-41. Ottawa: Correctional Service of Canada

Chapter 6: Appendix 1

Coding Manual

  1. FPS Number
    _______________


  2. AGE AT FIRST OFFENCE
    Date at first Criminal Code conviction3
    ___________________


    Date of first institutional sentence (30 days or more)
    ___________________


  3. CRIMINAL CODE CONVICTIONS UP TO INDEX ADMISSION DATE
    (see CRD classification of offences)
     
    PROVINCIAL
    FEDERAL


    Number of Criminal Code convictions
    __________
    __________


    Number of violent Criminal Code convictions
    __________
    __________


    Number of sexual4 Criminal Code convictions
    __________
    __________


    Number of UAL or escape convictions
    __________
    __________


  4. CRIMINAL CODE CHARGES DROPPED UP TO INDEX ADMISSION DATE

    Number of Criminal Code charges dropped
    ___________________


    Number of violent Criminal Code charges dropped
    ___________________


    Number of drug offence charges dropped
    ___________________


    Number of violent Criminal Code charges dropped
    ___________________


  5. SCHEDULED OFFENCES UP TO INDEX ADMISSION DATE
    (SEE SCHEDULED OFFENCES LIST)
     
    PROVINCIAL
    FEDERAL


    NUMBER OF PREVIOUS SCHEDULE 1 CONVICTIONS
    __________
    __________


    NUMBER OF PREVIOUS SCHEDULE 2 CONVICTIONS
    __________
    __________


  6. ESTIMATED INCARCERATION DAYS UP TO INDEX ADMISSION DATE
    (CHECK ONE)
     
    PROVINCIAL
    FEDERAL


    0 DAYS
    __________
    PREVIOUS FEDERAL TIME


    1 TO 365
    __________
    NO (DON'T ADD TOTAL DAYS)
    __________


    366 TO 730
    __________
    YES (ADD TOTAL DAYS)
    __________


    731 OR MORE
    __________


    TOTAL DAYS SENTENCED
    __________
    TOTAL DAYS SENTENCED
    __________


    TIME INCARCERATED REFERS TO TIME SENTENCED TO JAIL. IF AN OFFENDER HAS RECEIVED MULTIPLE OFFENCES AND SENTENCES ON THE SAME DATE ASSUME THAT THE LONGEST SENTENCE RECEIVED HAS PRECEDENCE AND THAT THE OTHER SENTENCES ARE BEING SERVED CONCURRENTLY, UNLESS SPECIFIED THAT THE SENTENCES ARE BEING SERVED CONSECUTIVELY.

    IF AN OFFENDER RECEIVES ANOTHER SENTENCE AFTER HIS INITIAL SENTENCE AND BEFORE THAT SENTENCE EXPIRES ASSUME THAT SENTENCE IS BEING SERVED CONCURRENTLY, UNLESS SPECIFIED THAT THE SENTENCE IS TO BE SERVED CONSECUTIVE TO THEIR CURRENT SENTENCE. TO GET AN ESTIMATE OF TOTAL INCARCERATION TIME ADD UP THE AMOUNT OF TIME THAT THE OFFENDER HAS SPENT INCARCERATED KEEPING IN MIND THAT MULTIPLE SENTENCES MAY BE SERVED CONCURRENTLY.

    "CONC" REFERS TO SENTENCES BEING SERVED CONCURRENTLY

    "CONSEC" REFERS TO SENTENCES BEING SERVED CONSECUTIVELY

    NOTE THAT "CONC" MAY NOT ALWAYS APPEAR. WHEN IN DOUBT ASSUME THAT THE SENTENCE IS BEING SERVED CONCURRENTLY.


  7. ANY CONVICTIONS AFTER INDEX RELEASE
    OFFENDER'S INDEX RELEASE DATE WAS BEFORE JUNE 30, 1995 (CODE ANY OFFENCE AFTER INDEX RELEASE)

    No (finished)
    _______________


    Yes (code)
    _______________


    Number of Criminal Code convictions
    __________


    Number of violent Criminal Code convictions
    __________


    Number of sexual Criminal Code convictions
    __________


    Number of drug offence convictions
    __________


    First sentence date
    __________


    First sentence date (30 days or more)
    __________
    sentence length (days)
    __________


    First sentence date for violent offence
    __________
    sentence length (days)
    __________


    First sentence date for sexual offence
    __________
    sentence length (days)
    __________


    First sentence date for drug offence
    __________
    sentence length (days)
    __________


  8. CONVICTIONS UP TO 1 YEAR AFTER INDEX RELEASE
    FOLLUP1 IS BEFORE JUNE 30, 1995 (CODE FROM INDEX RELEASE TO FOLLUP1)

    No (finished)
    _______________


    Yes (code)
    _______________


    Number of Criminal Code convictions
    __________


    Number of violent Criminal Code convictions
    __________


    Number of sexual Criminal Code convictions
    __________


    Number of drug offence convictions
    __________


    First sentence date
    __________


    First sentence date (30 days or more)
    __________
    sentence length (days)
    __________


    First sentence date for violent offence
    __________
    sentence length (days)
    __________


    First sentence date for sexual offence
    __________
    sentence length (days)
    __________


    First sentence date for drug offence
    __________
    sentence length (days)
    __________


  9. CONVICTIONS UP TO 2 YEARS AFTER RELEASE
    FOLLUP2 IS BEFORE JUNE 30, 1995 (CODE FROM INDEX RELEASE TO FOLLUP2)

    No (finished)
    _______________


    Yes (code)
    _______________


    Number of Criminal Code convictions
    __________


    Number of violent Criminal Code convictions
    __________


    Number of sexual Criminal Code convictions
    __________


    Number of drug offence convictions
    __________


    First sentence date
    __________


    First sentence date (30 days or more)
    __________
    sentence length (days)
    __________


    First sentence date for violent offence
    __________
    sentence length (days)
    __________


    First sentence date for sexual offence
    __________
    sentence length (days)
    __________


    First sentence date for drug offence
    __________
    sentence length (days)
    __________


3 Criminal Code offence includes Narcotic Control Act offences.
4Prostitution is not a sexual offence.
Date Modified: 2015-03-05

Top of Page
Important Notices