Performance Assurance

Evaluation Report: The Section 81 Agreement between the Native Counselling Services of Alberta and the Correctional Services of Canada; The Stan Daniels Healing Centre

File #394-2-30

Evaluation Branch
Performance Assurance Sector
October 7, 2005

KEY FINDINGS

The following results are presented under their respective Evaluation Objectives.

Objective 1: Success:

FINDING 1: The Stan Daniels Healing Centre generally operates below optimum capacity despite a good pool of potential candidates.

The current recruitment strategy employed to identify and recruit appropriate candidates for residency at the Centre involves staff members regularly conducting information sessions at minimum security institutions that target institutional parole officers as well as inmates. The goal of the information sessions is to encourage referrals of appropriate candidates.

Analyses of automated data revealed that, despite a pool of potential candidates available for transfer to the Centre, an average of 9 beds remained vacant at the Centre per month between May 1^st, 1999 and June 1^st, 2005. The Stan Daniels Healing Centre has a bed capacity of 73, of which an average of 64 beds was occupied per month. Examination of a series of monthly offender profiles revealed there was an average of 26 potential Aboriginal candidates to consider for residency at the Centre. Potential candidates were defined using the selection criteria agreed upon by CSC and the NCSA^[41]. Specifically, Aboriginal inmates were selected if they were within 6 months of a discretionary release, rated as having high motivation, high reintegration potential levels, low institutional adjustment risk and low security risk^[42], and were residing in a minimum security institution (n=11). As well, Aboriginal offenders were selected if they were on discretionary release (day or full parole), residing in a Community Correctional Centre or a Community Residential Facility in the Prairie region, and rated as being highly motivated and having high reintegration potential (n=15). It is important to note that, in comparison to actual protocols, this definition is conservative with respect to offender ethnicity^[43] and release conditions^[44]. If the offender ethnicity criterion were expanded to include non-Aboriginal offenders, the candidate pool would increase substantially (to an estimated 107 offenders, 57 inmates and 52 offenders). Given that transfers to the Centre are voluntary, it should be noted that the absence of information confirming the willingness of theses potential candidates to transfer to the Centre limits the analyses to the “identification” of possible transfers.

Themes generated from interviews with key sources regarding the level of efficiency of the Centre support the finding that the Centre is under-utilized. When sources were asked whether they thought the Centre was being used to maximum capacity, the majority of Stan Daniels staff did not consider the Center to be utilized to its maximum capacity (11/17), and 5 of 8 CSC staff members responded similarly.

RECOMMENDATION 1: To ensure the Agreement operates more efficiently, CSC and the NCSA should enhance the current identification and recruitment processes through a formal review mechanism of all cases and a formal referral protocol.

FINDING 2: The funding formula for the Centre was agreed upon at a point in time where the resident profile was comprised of approximately 50% discretionary release residents, 20% statutory release residents and 30% inmates. However, there has been an increasing proportion of residents with higher supervision needs (inmate and statutory release residents). This trend is not factored in to the funding formula. This situation may impact the Centre's capacity to meet the needs of this changed profile of residents.

As described in the Program Profile section of this report, the funding formula for the Centre seems to have been based on a per diem rate for the supervision of conditionally released offenders at Community Residential Facilities (CRFs) and Community Correctional Centres (CCCs) (see Table 5). However, the Centre has also been accepting clients with more intensive supervision needs. More specifically, there has been an influx of minimum and even medium security inmates, as well as statutory release cases with residency conditions (see Figure 6). Typically, more staff members are assigned to supervise incarcerated inmates in CSC's facilities relative to those on discretionary release (see Table 6).

Resource issues were raised in a document provided to the evaluation team by the Native Counselling Service of Alberta (NCSA) Director of Operations. In general, the document indicates that the nature of the current funding structure (salaries are determined largely by bed utilization and per diem rates), is not conducive to recruiting and retaining staff members required for the resident groups (inmates and parolees). The document suggests that, although the current staffing pattern meets the dynamic security requirements, there are significant challenges in meeting the Centre's reintegration strategy. Specifically, the document recommends salaries for an additional two full-time-equivalent (FTE) employees in order to meet the demands for escorted temporary absences and to assist with pre-release planning, as well as an additional Parole Officer FTE employee to meet community and institutional parole supervision duties. These concerns were also expressed by the Chief Executive Officer (CEO) of the NCSA. In a letter to the Commissioner, CSC (2001) the CEO expressed that a funding formula based on the supervision of conditionally released offenders does not include costs associated to the escorted temporary absences (ETAs) required for inmates.

Although key sources did not explicitly discern between the needs of parolees versus inmates, themes generated from key-source interviews suggest a gap in human resources associated with the needs of the resident inmates. Interviews with Stan Daniels staff members revealed that a majority, 76% (n =17), felt that the resources were inadequate to facilitate the timely release of inmates and offenders into the community. The vast majority (13 of 17) indicated a need for more resources and/or funding in order to recruit and retain staff, such as psychologists, councillors and Elders. One staff member stated that “people come here specifically to prepare for a release, and could be waiting up to six months for a psychological risk assessment”. Two of the Centre's staff members also indicated more human resources were required for escorted temporary absence (ETAs), a type of release granted to inmates. Although half of CSC staff members indicated they thought that resources were adequate, they still expressed a need for additional funding and the presence of more Elders. Lastly, while three of six resident inmates interviewed identified ETAs as release strategies they have been successfully granted, 2 residents expressed a concern over ETAs, either not getting them or having them cancelled.

Table 5: Expenditure per diems

	Stan Daniels Daniels Centre	Community Residential Facility (CRF)	Community Correctional Centre (CCC)		Minimum Security Institution					Medium Security Institution
			Osborne	Oskana	Willow Cree	Riverbend	Rockwood	Grand Cache	PeSakestew	Drumheller	Bowden
Salary expense	916,885	808,582	450,347	359,549	2,286,975	3,932,714	4,030,384	8,632,458	2,547,474	19,633,424	19,778,925
Salary per diem	xx1	xx1	xx1	xx1	xx1	xx1	xx1	xx1	xx1	xx1	xx1
Operating expense	920,634	280,140	355,615	317,780	653,658	1,298,299	1,662,647	4,355,263	815,018	6,853,374	7,033,196
Operations per diem	xx1	xx1	xx1	xx1	xx1	xx1	xx1	xx1	xx1	xx1	xx1
Total expense	1,837,519	1,088,722	805,963	677,329	2,940,634	5,231,013	5,693,031	12,987,721	3,362,492	26,486,797	26,812,121
Expense per diem	xx1	xx1	xx1	xx1	xx1	xx1	xx1	xx1	xx1	xx1	xx1

1: Access to Information Act, Sect. 20(1)(b) and 20(1)(c)

Expenses per diem for the Centre, calculated as the actual expense divided by the actual number of bed days used, fall within the range of those for Community Residential Facilities and Community Correctional Centres in the Prairie region. Notably, the per diem expenditures for the Centre are less than those of minimum security institutions in the Prairie region, as well as the medium security institutions from which the Centre has taken residents in the past.

Figure 6: Average Number of Residents

The proportion of inmate and statutory release residents was just under 50% in 1999/2000, peaked to 70% in 2002/2003 and has remained over 60% for the last two years of the analyses. The number of staff members at the Centre over that time period fluctuated between 25 and 27.

Table 6: Staff to Inmate/Offender Ratios

	Stan Daniels Healing Centre	Community Residential Facilities (CRF)	CCC		Minimum					Medium
			Osborne	Oskana	Willow Cree	River-bend	Rockwood	Grand Cache	Pe Sâkâstêw	Drum-heller	Bowden
# Staff	21	15.5	7	8	20.8	78	82.2	131.5	44.6	322.5	354.5
# Residents	59	40	30	17	39	88	90	235	39	507	533
Staff/Offender Ratio	1:2.8	1:2.6	1:4.3	1:2.1	1:1.9	1:1.1	1:1.1	1:1.8	1:0.9	1:1.6	1:1.5

During the fiscal year 2004/2005, the Stan Daniels Healing Centre had 21 staff who supervised an average of 59 offenders (21 inmates, 34 day parolees, and 4 full parolees). Thus, the staff/offender ratio was 1 staff member to 2.8 offenders. This ratio is just below that of the standard requirement for a Community Residential Facility (CRF). Specifically, according to CSC National Guidelines, a CRF with 36 to 40 beds capacity requires 15.5 staff, a ratio of 1 staff member to 2.6 offenders^[45]. Notably, the ratio is far below that of minimum security institutions in the Prairie region. In 2004/2005, minimum institutions in the Prairie Region had a staff/inmate ratio of 1:1.9 or higher^[46].

RECOMMENDATION 2: CSC and the NCSA should review the current per diem to ascertain its precision and explore the option of a specific per diem for each status type (i.e. inmate and offender) for candidates housed at the Centre.

FINDING 3: The Centre successfully provides alternatives to incarceration, particularly for the Aboriginal demographic of the resident population, while maintaining levels of reintegration commensurate with other similar correctional facilities.

Discretionary Release Grants:

Between May 1999 and June 2005, there were 372 discretionary releases granted to residents at the Centre. Approximately 61% (n=226) were granted to Aboriginal residents. When compared to the matched^[47] group, Chi-square analyses revealed that inmate and day parole residents (both Aboriginal and non-Aboriginal) at the Centre were more likely to be granted a discretionary release^[48] than their counterparts (45% vs. 13% respectively, ²(1, n = 794) = 103.44, p<0.001). Further, within the Aboriginal demographic of the sample, resident inmates and day parolees were more likely to be granted a discretionary release^[49] than their matched Aboriginal counterparts (50% vs. 9% respectively, ²(1, n = 370) = 73.157, p<0.001).

Technical Revocation

The median time to technical revocation^[50] was used as a measure of central tendency for the survival analyses. An examination of the matched sample revealed that residents at the Centre (Aboriginal and non-Aboriginal) were more likely to return with a technical revocation than their matched counterparts over time [Χ² (1, n=983)=17.693, p<0.001)]. This finding was replicated for the Aboriginal demographic of the matched sample [Χ²(1, n=557)=5.7308, p<.05)]. For both analyses, low base rates for technical revocations precluded any estimation of the median length of time in the community^[51]. These findings should be interpreted with caution, as residents at the Centre were more likely to be released to urban locations when compared to their matched counterparts (88.2% vs. 76.5%, Χ²(1, n=940)=22.39, p<.001). It is possible that, due to geographic disparity and population density, residents released from the Centre may have been more highly scrutinized and/or have had more opportunity to breach conditions of their parole release.

Reconviction:

An examination of the matched sample revealed there were no significant between-group differences in survival curves for reconviction^[52] [Χ²(1, n=983)=2.139, p=0.14]. Thus, there were no between group differences in the probability of returning with a new offence over length of time. Low base rates for returns with a new offence precluded any estimation of the median time in the community. Finally, this finding was replicated for within-between group analyses. Particularly, an examination of the Aboriginal demographic of the matched sample did not reveal any significant between group differences in survival curves for reconviction [Χ²(1, n=557)=2.403, p=0.12]^[53].

Figure 7: Survival Time to Reconviction: Residents versus Non-residents

Program Participation

Although correctional program assignment and completion information is incorporated into CSC's Offender Management System (OMS), the evaluation report did not conduct comparisons with the matched group sample on correctional program outcome due to between group disparities. Considering that additional constraints controlling for the opportunity to take and complete specific correctional programs would have been required to facilitate between group comparisons, the additional matching criteria^[54] would have resulted in attrition in the sample to a degree that would render comparisons insubstantial.

Results from interviews with key sources indicated that they believe the enhanced activities and programs offered at the Center provides a secure and supportive environment for its residents, and effectively prepared residents for release into the community. The majority of Stan Daniels staff members (82%; n= 14), CSC staff members (59%; n= 10), and residents (97%; n = 28) agreed that activities provided for residents of the Centre created a safer and more encouraging environment. The vast majority of offenders at the Centre reported that they feel “safe, secure, and supported” by the Centre (97%; n= 28). When residents were asked how the Centre compared to other facilities, the majority (79%; n= 23) responded favourably towards Stan Daniels. Resident sources indicated that they felt that staff members were approachable and accessible (61%, n=28).

The Centre is performing many activities and programs to respond to the individual needs of offenders. Under the Agreement, the Centre is expected to provide programs designed to address the mental, emotional, physical and spiritual needs of Aboriginal offenders. These programs should also allow offenders to connect to the history, their culture and provide the opportunity to participate in a personal healing process. The majority of respondents indicated that they participate in Aboriginal-specific programming (76%; n= 22) and consider this activity important (83%; n=24). As one resident stated, “interactions with Elders have helped me make changes in life. [These] changes would never have taken place without spiritual/cultural guidance.”

Programs were perceived by all key source groups interviewed to meet the needs of inmates and offenders. The majority of CSC staff members (9 of 12) reported that programs helped meet the overall needs of participants. Resident respondents (72%; n= 21) stated that the activities and programs had assisted them with community reintegration. The overwhelming majority of respondents from Stan Daniels (15 of 16) and CSC (6 of 7) thought the Stan Daniels philosophy towards offender reintegration is continuously meeting the reintegration needs of the Aboriginal offenders.

The overall theme throughout much of the qualitative data collection was that the array of programs and activities offered through the Centre is what sets it apart from other facilities and it is what helps offenders most significantly in their efforts toward successful reintegration. Over 90% of staff members at the Center (94%; n= 16) stated that, in their opinion, communities have become a safer place as the result of Stan Daniels. When staff members were asked what this increased safety was attributed to, respondents suggested enhanced programs, better preparation for release, and that the Centre provided a place for growth and was a generally supportive environment. As one respondent stated, “they have opportunities to learn and practice new skills, which make for a safer community.”

Employment in the Community

An examination of employment needs in the community^[55] did not reveal any significant between group differences for the matched sample (59.0% of residents had employment needs, and 53.5 % of non-residents were assessed as having these needs, Χ²(1, n=838)=2.564, p=0.11). This finding was replicated within the Aboriginal demographic of the matched sample, as 63% of both Aboriginal residents and Aboriginal non-residents were rated as having employment needs, (Χ²(1, n=485)=0.006, p=.93).

Results from interviews with key sources showed that a field placement or community service work with community agencies are essential to the reintegration of Stan Daniel's residents into the community. The vast majority of residents (75%; n= 28) had access or were referred to a community agency through the Stan Daniels Healing Centre. Of those interviewed who had access to the agencies, all but one confirmed that the placement had been helpful (20 of 21).

---

^[41] See Selection Criteria in this report.

^[42] Institutional Adjustment risk and Security Risk are two subscales that comprise the Custody Rating Scale (CRS). The CRS is a research based tool that assists in determining the most appropriate level of security for the initial penitentiary placement of the offender. The sub-items that comprise the Institutional Adjustment Risk scale are : History of involvement in institutional incidents,- Age at time of sentencing, Length of current sentence, Street stability, and Alcohol and drug use. The sub-items that comprise the Security Risk scale are: Escape history, Most serious outstanding charge, and Previous periods on parole or statutory release.

^[43] Only Aboriginal Offenders were considered for the analyses, while the actual selection criteria do not limit the candidate pool as such.

^[44] Only those within 6 months of a discretionary release (day parole or full parole release) were considered for the analyses, while the actual selection criteria also include those within 6 months of a statutory release.

^[45] Source: National Unit Costs, from Community Reintegration Operations

^[46] Source: Resource Allocation Exercise, from Financial Management

^[47] Recall that the matched cohort was drawn from male offenders either incarcerated in a minimum security institution or supervised in the Prairie region between May 1^st, 1999 and June 1^st, 2005. Matching criteria were sentence length, risk level and age at time of admission. These matching criteria introduced controls for time, opportunity and tendency in the research design. The random sample was also stratified on ethnicity groupings (Aboriginal and non-Aboriginal) within offender status groupings (inmate, parole, statutory release). This was done to ensure that comparisons reflected exposure to the Stan Daniels Healing Centre rather than characteristics associated with a particular ethnic demographic of the offender population.

^[48] For inmates, 'discretionary release' could be either day parole or full parole, while for residents already on day parole, a successful full parole grant constituted the discretionary release.

^[49] Ibid.

^[50] A technical revocation is a revocation for the violation of a condition of the parole where no criminal incident occurred but the offender's risk to public safety was assessed to have required a return to penitentiary.

^[51] For the matched sample, failure rates (return with a technical revocation prior to the end of the survival period) were 33.4% and 21.6% for the resident and non-resident counterparts respectively. For the Aboriginal demographic of the sample, these failure rates were 35.8% and 25.6% for the resident and non-resident groups respectively.

^[52] Reconviction is defined as a return with a new offence.

^[53] For the matched sample, failure rates (return with a new offence) were 14.3% and 17.6% for the resident and non-resident counterparts respectively. For the Aboriginal demographic of the sample, these failure rates were 13.7% and 18.0% for the resident and non-resident groups respectively.

^[54] Specifically, the analyses require that residents are also matched on the proportion of their sentence served, the duration of correctional programs that were taken, and the amount of time available in the resident's sentence in order to complete the program.

^[55] Assessments are completed by the community to provide complete, accurate and quality information to assist in every activity related to the offender's assessment in the correctional process. For employment, a rating of “FACTOR SEEN AS AN ASSET TO COMMUNITY ADJUSTMENT” indicates that employment has been stable and has played an important role for the offender. A rating of “NO IMMEDIATE NEED FOR IMPROVEMENT” indicates that employment, under employment, sporadic employment, or chronic unemployment have not interfered with daily functioning. An offender receives a rating of “SOME NEED FOR IMPROVEMENT” if any of the aforementioned have caused minor adjustment problems while in the community and “CONSIDERABLE NEED FOR IMPROVEMENT” if the employment situation has caused serious adjustment problems (CSC Standard Operating Practice #700-03, Assessments Completed by the Community).

TABLE of CONTENTS

NEXT