The Evolution of Community Supervision: Research on Training and Implementing Evidence-Based Practices Guy Bourgon, Ph.D., C. Psych. Los Angles, CA July 2015

The supervision officer as a foot soldier Like the soldier who has marched into the battlefield and, tired, weary, and burdened with the ever-changing orders from the generals, he must make crucial decisions all alone, the community supervision officer “follows” the orders of management and works tirelessly behind closed doors to promote change in clients characterized as antisocial, lacking motivation or in downright denial, resistant, defensive, aggressive, and criminal.

Community Supervision & Safety What does the research tell us about reducing reoffending?

Community Supervision Effectiveness? Meta-Analytic Findings Outcome Φ k N General Recidivism .02 26 53,930 Violent Recidivism .00 8 28,523 Bonta et al. (2008) k = number of effect sizes Minimal impact on recidivism

Risk – Need – Responsivity Model of Effective Corrections However… Risk – Need – Responsivity Model of Effective Corrections

Three Principles of Effective Correctional Intervention 1. Risk Principle Intervention intensity is proportional to risk levels; higher risk = more time, energy and resources. 2. Need Principle Target the criminogenic needs of moderate and high risk offenders (i.e., the dynamic risk factors that, when changed, will change the probability of an offender’s recidivism). 3. Responsivity Principle Create an optimal learning environment by: General: Utilize cognitive-behavioral techniques. Specific: Matched to the individual abilities, learning style, culture and personality, create an environment via the skills, language, and activities that enhance client engagement and facilitates learning. 6

Adherence to RNR Principles Treatment Effect Sizes Andrews, D. A., & Bonta, J. (2010). The psychology of criminal conduct (5th ed.). New Providence, NJ: LexisNexis Matthew Bender. (page 367). Dowden, C. & Andrews, D.A. (2000). Effective correctional treatment and violent reoffending: A meta-analysis. Canadian Journal of Criminology. 449-467. Dowden, C. & Andrews, D.A. (2003). Does family interventions work for delinquents? Results of a meta-analysis. Canadian Journal of Criminology & Criminal Justice. 327-342. Dowden, C. & Andrews, D.A. (1999). What works for female offenders: A meta-analytic review. Crime & Delinquency, 45, 438-452. Hanson, R.K., Bourgon, G., Helmus, L. & Hodgson, S (2009). The principles of effective correctional treatment also apply to sex offenders: a meta-analysis. Criminal Justice & Behavior, 36, 865-891. (Sexual recidivism). Gutierrez, L. & Bourgon, G. (2012). Drug Treatment Courts: A quantitative review of study and treatment quality. Justice Research and Policy. 14, 47-77. Unpublished. Data from Bonta, J., Rugge, T., Scott, T., Bourgon, G., & Yessine, A. (2008). Exploring the black box of community supervision. Journal of Offender Rehabilitation, 47, 248-270.  

What about the RNR principles? Does adherence make a difference? Community Supervision policies adherence with Risk-Need-Responsivity RNR Adherence Φ k N RNR = 0 .017 7 47,885 RNR = 1 -.009 12 2,716 RNR = 2 .078 5 2,415 RNR = 3 .092 2 914 RNR: Trends congruent with past literature are noticed in the results. Generally, as adherence to the RNR principles increased in the studies included in the Bonta and colleagues meta-analysis, the odds ratios decreased in magnitude (indicating more effective treatment). For example, studies adhering to none of the RNR principles demonstrated the weakest treatment effect (OR = .88), while the odds ratio taken from studies adhering to all three RNR principles (.67) demonstrates the strongest treatment effect. A significant linear trend exists between adherence to the RNR principles and treatment effectiveness. In other words, this trend indicates that when adherence to the RNR principles increases, a corresponding decrease in recidivism is observed (t = 2.997, p < .01) It should be noted that the confidence intervals of the odds ratios belonging to the “zero RNR principles adhered to” are slightly over-lapping with the confidence intervals of the “three RNR principles adhered to” prohibiting any definitive conclusions regarding differences. However, only two of the 26 effect sizes were taken from studies that adhered to all three of the RNR principles. Increase in RNR adherence = Decrease in recidivism

It “Works” But… Demands of RNR are high… Assessment Criminal risk factors and needs (changeable) Strengths and learning styles (responsivity) Services that target appropriate needs Requires appropriate intensity and targets Commitment to reduce re-offending Services are tailored for “offenders” learning styles Use cognitive-behavioural social learning strategies Officers use EBP skills and techniques Optimal learning environment = respect and collaboration

How tough? Tough enough for formal treatment programs… Adherence to principle % of studies Risk 75% Need 45% Responsivity 21% All 3 Principles 16%

Can we improve the work done “behind closed doors” to more effectively reduce reoffending? Implementing “Evidence-Based Practices”

Evidence Based Practices New initiatives…. Core Correctional Practices and the new training initiatives of STICS, STARR, EPICS & PCS… Each of these emphasizes a change in work… Evidence Based Practices = CHANGE AGENT APPROACH

Evidence of EBP Community Supervision Training Efforts: Approximately 11% difference in recidivism Source: Chadwick, N., Smeth, A.H. & Serin, R.C. (in press). Effectively training community supervision officers: A meta-analytic review of the impact on offender outcome. Criminal Justice and Behavior.

Lessons of Implementing… EBP requires continuous education! At every level Continuous education critical! ??? Invest in your leaders! Quality over Speed Demands change in Individuals & Organization! Vision, goals, roles, policies, practices & behavior

Implementation Lesson 1 Continuous Education

Control vs. Low Support vs Control vs. Low Support vs. High Support Skills in Sessions > 9 months post-training Anova shows Structrure F,2,39) = 3.081; p = .057; Sig diff for High vs. Contl Relationships F (2,39)=1.014 p = .372; ns BEHTECH F(2,39)=1.634; p=.208; ns COGTECH F92,39)=14.909; p=.000; signif diff all: contl vs. Low vs High. STICSTOT F(2,39) = 4.710; p=.015; sign diff contl & low vs. high; contl = low.

Control vs. Low Support vs Control vs. Low Support vs. High Support Cox Regression (control age & LSI-R criminal history) Low Support: Exp(B) = .864 95%CI = .433 - 1.724 Δ -4% Hi Support: Exp(B) = .500 95%CI = .222 - 1.127 Δ -19% Control n=43 Low Support n = 59 High Support n = 41 Exp(B) = .864 for Low (~4% difference) and .500 for High (~18.8% difference) 2 Year Unadjusted Recidivism Control = 40.5% STICS Low Supt = 27.7% STICS High Supt = 21.4%

Implementation Lesson 2 Invest in Leaders

“Expertise” is the critical Do not have sufficient Initial “Expertise” Developing it takes time! Trainers Coaches Officers Expertise

Skill levels of Coach vs Officer

Skill levels of Coach vs Officer

Skill levels of Coach vs Officer

Responsivity Principle… Implementing EBP requires creating an environment that facilitates change, supports it, and helps sustain it… Sound familiar??? Responsivity Principle…

Implementation Lesson 3 Change Takes Time

Change over time: How officers work.

Implementation Lesson 3 Organizational Change

The Vision… Community Supervision’s Identity Crisis…

Probation Officers?

Or Probation Officer?

Probation Work?

Or probation work?

Other Organizational Challenges Implementation Integrity Policy/Procedure/Staff “Conflicts” with EBP (what are the existing barriers? What can change?) Quality assurance/evaluate uptake Create & Resource “Structures” HR plans, workloads, logistics, etc. Build Capacity & Sustainability Everything new so NO existing “expertise” Capacity to develop new staff, coaches, trainers etc.

350 Officers and 3 years later Where are we now?

Full scale vs. original pilot % of Time During Session Topic Discussed

Full scale vs. original pilot % of Sessions in which Attitudes (cognitions) Discussed % of Sessions in which Cognitive Techniques were used STICS: Focus on Attitudes & Use Cognitive Techniques

“Change Agent” Skills: Skills and intervention techniques used by PO Uptake looking positive so far…

Going Forward… Research and evaluation efforts continue What factors enhance officer skill and use? Link key skills/techniques to reoffending? What to do with newly hired staff? Training Model…new sustainability approach 18 month training with 4 separate courses Utilize existing resources & structures Rely on and build more internal capacity/expertise

Although there are many hurdles to quality implementation, both at an organizational and an individual level, these barriers can be overcome. For each individual officer who faces the prospect of change, it is a daunting task. Organizations and trainers can provide the opportunities and supports, but ultimately, when the door closes and the officer and client meet face-to-face, the officer must decide what he or she will do with the time allotted to their clients. Like the lonely foot soldier, laden with a heavy burden alone in the trenches, officers must draw upon their own resources and decide for themselves whether to march forward or stay where they are. The power to choose is theirs.

Available on the Public Safety Canada Website Further Information Available on the Public Safety Canada Website www.publicsafety.gc.ca Contact Guy.Bourgon@ps.gc.ca 340 Laurier Ave West Ottawa, ON Canada K1A 0P8