Reinventing the wheel: Testing a new CoSA-based sex offender intervention

John Vaccaro

In 1994, a small-town event catalyzed the creation of a new method of supervising sexual offenders in the community. At that time, the residents of Ontario, Canada became aware that a high risk sexual offender, who had molested multiple children, was being released with no supervisory restrictions (Canada’s equivalent of “maxing out”). This led to immense upheaval in the local community. The situation seemed irresolvable until a Mennonite pastor of a local congregation offered to supervise the offender and keep him accountable through regular meetings, checkups, and guidance. After the situation turned into a success story – the offender was never rearrested – the Mennonite Central Committee of Ontario and the Correctional Service of Canada partnered to create what is now called Circles of Support and Accountability (CoSA).

Based on the principles of restorative justice, CoSA does not promise to prevent recidivism, but instead provides an offender with a support system as well as supervision. Each CoSA consists of two concentric circles (Wilson & McWhinnie, 2010). The inner circle is comprised of the Core Member (the ex-offender receiving the intervention) and four to six briefly trained community member volunteers who meet with the Core Member regularly and act as a support network. The outer circle consists of corrections and mental health professionals who support the inner circle.

In terms of the effectiveness of CoSA, several notable evaluation studies of varying empirical rigor have found reductions in sexual and general recidivism for Core Members when compared to control groups (Wilson, Picheca & Prinzo, 2005; Wilson, Wilson, Cortoni & McWhinnie, 2009; Duwe, 2013). Using three studies that reported statistically significant reductions in the odds of sexual recidivism, Wilson, McWhinnie, Picheca, Prinzo, & Cortoni (2007) reported a total reduction of 77%.

However, Elliott, Zajac and Meyer (2013) critiqued the research base supporting CoSA evaluations. Their analysis found a) methodological issues, including nonequivalent experimental and control groups as well as short follow-up times, b) statistical inconsistencies – some tests were a poor fit for the data, and c) a lack of statistically significant results in some articles. Due to these potential limitations in the current literature base supporting CoSA, Elliott and colleagues (2013) suggest the current evaluations be considered an estimate of CoSA’s potential effectiveness.

In March of 2013 John Jay College of Criminal Justice Psychology professors Elizabeth Jeglic and Cynthia Calkins came up with a research idea to test a program similar to CoSA but make several key changes. The professors left the task of designing and implementing a pilot study to four graduate students in their Sex Offender Research Lab (SORL) (SORL Twitter). The changes hinged on three main concerns with the original CoSA design:

1. CoSA uses briefly trained lay volunteers to supervise high risk sexual offenders

2. CoSA does not use an empirically validated structured assessment of criminogenic needs (e.g., stability in housing, employment, and social relationships) and

3. CoSA focuses more on social support than the offenders’ other reintegration needs (e.g., applying for governmental assistance, obtaining housing and employment, navigating sex offender specific restrictions)

The use of volunteers makes CoSA extremely cost effective to implement. However, briefly trained volunteers may not be aware of the many issues pertaining to working with sexual offenders, such as difficulty managing deviant sexual interests and helping offenders adhere to numerous laws and restrictions. This is consistent with Wilson et al., (2005) who surveyed core volunteers and found that over half thought they should have received more training.

Additionally, CoSA does not comprehensively assess each offender to determine what their needs are nor create a structured plan on how to help them meet those needs. While CoSA naturally excels at bolstering the offender’s social support, it may be more beneficial to target the logistical aspects of the reintegration process that are particularly prevalent shortly after the offender’s release.

Finally, previous evaluations of CoSA have relied on recidivism data as an outcome measure. This can be misleading due to short follow up times and the low base rate of recidivism (Elliott et al., 2013). Our intervention addresses this limitation by measuring dynamic risk factors as outcomes. These include increased satisfaction with one’s life and attainment of goals, perceived preparedness to reenter the community, and participation in prosocial activities that encourage the continuation of a crime free life.

Current study

The Social Support and Case Management Intervention pilot study recruited a convenience sample of sexual offenders from an out-patient sex offender treatment center. Participants were completing out-patient treatment at the center as part of their probation or parole requirements. The intervention implements the following components over a six month period:

Support Intervention Team (SIT)

A Support Intervention Team (SIT) is comprised of two graduate students with extensive training and experience in sexual offender treatment and reintegration. The SITs meet with the participants weekly and contact the participant by phone at least once a week. If the offender requires more intensive interactions additional phone sessions can be arranged. The SIT works with the offender to complete logistical tasks associated with reintegration and provides social support.

Structured Needs Assessment

A needs assessment was created for this project that assesses criminogenic and logistical needs of offenders reentering the community. The semi-structured interview evaluates need domains identified by prior research. Each domain is rated as either being primary (most pressing), secondary (not as pressing), or not a need. This is used to plan which needs will be focused on first in the intervention.

Informational Packets

The study designers created informational packets targeting specific need domains. The packets were designed to provide comprehensive information to help the offender meet specific needs. Current packets include information to aid with finding employment, housing, support groups, applying for government assistance, etc. The SIT works through the packet with the offenders and aids with utilizing the packet information.

Bi-monthly data collection

Several self-report surveys are given to participants bi-monthly to assess their general life satisfaction as well as their satisfaction with the intervention and their need domains. This data is essential for monitoring the program’s effectiveness and will be used in revision of the study procedures and methodology.


Throughout the study SIT staff provide feedback in the form of suggested changes to informational packets, suggestions for new packets, a log of additional information used in participant sessions, as well as suggestions given by participants. This data is used for honing study methods and resolving logistical problems.

The graphic below illustrates the intervention:


Future Directions

Feedback from this pilot study will be used to employ a larger Randomized Controlled Trial (RCT) design to assess if the addition of a Social Support and Case Management Intervention with regular out-patient treatment can improve offenders’ chances at successfully reintegrating into the community and help them sustain crime free lives.


The Social Support Intervention Research Team consists of four John Jay College of Criminal Justice graduate students: John Vaccaro, Julia Campregher, Michelle Rosselli, and Kyle Meditz. Each involved in their own personal thesis research, they collaborated to make the Social Support and Case Management Pilot Study a reality. Collectively, they share a common group of interest in researching and working clinically with sexual offenders, specifically developing an innovative treatment and supervision program that can be broadly implemented. Their shared mission is to prevent sexual violence in the community by facilitating a smooth reintegration into the community for convicted sexual offenders, and they work very hard to make sure their dreams are realized.

Suggested citation:
Vaccaro, J. (2015, Mar 22). Reinventing the wheel: Testing a new CoSA-based sex offender intervention. [Weblog post]. Retrieved from

Want to submit your own post? Click here to find out how!

Email us:
Follow us: @nextgenforensic

One thought on “Reinventing the wheel: Testing a new CoSA-based sex offender intervention”

Comments are closed.

%d bloggers like this: