Using CBT to Reduce Risk Factors Linked to Recidivism in Sex Offenders

Background and Purpose: Recidivism remains the primary outcome variable used to measure the effectiveness of treatment for sexual offenders. Embedded in this process is the assumption that what occurs in treatment has a direct link to changes in recidivism. Yet, it is unclear as to what occurs in treatment that might have a direct impact on the identified risk factors associated with recidivism ( also referred to as criminogenic needs). Simply, there is no clear understanding as to what mechanisms within the treatment process influence reductions in recidivism. This study is an attempt to contribute to this area of the literature by presenting the results of a pilot study aimed at understanding whether changes in risk factors occur and using attachment theory as an explanatory theory to explain any reported changes.

Methods: The purpose of this study was twofold: 1) to test whether criminogenic needs changed over the course of an intense residential sexual offending CBT-based treatment program for 34 incarcerated adult male sex offenders (treatment n = 16, waitlist n= 18); and 2) whether changes in attachment styles influenced changes in criminogenic needs in this population. The study used a quasi-experimental design to test these two research questions. The treatment participants completed questionnaires at the start of the 5 month treatment program and at its completion. The waitlist group completed the same questionnaires using the same time frame.

Attachment styles were measured using the Relationship Scales Questionnaire (RSQ) and the Adult Attachment Scale (AAS). Criminogenic needs were measured using the Social Self-Esteem Inventory (SSEI), the UCLA Loneliness Scale (Version 3)(LS), , the Dysfunctional Emotional Regulation Scale (DERS), the Child Molester Empathy Measure (CMEM), and the Rapist Empathy Measure (RMEM).

Results:  The treatment group showed significant increases in emotional awareness (t = 6.93; p =.00) and emotional clarity (t = 2.35; p = .03) from pre to post treatment, and there was a significant decrease in subjective feelings of loneliness between participants in the treatment and waitlist groups (F = 8.99, p < .001). Most importantly, changes in attachment styles showed statistically significant changes in risk factors in the treatment group in comparison to the waitlist group.  ANCOVA runs used pre/post test score differences in attachment styles as the covariate and pre/post test score differences in criminogenic need scales as the dependent variable. Increase in the AAS close attachment style were directly associated with decreased loneliness (F = 4.84; p < .05), total dysfunctional regulation (F = 4.55; p < .05), and difficulties in engaging in goal directed behaviors (F =6.27; P < .05). Decrease in AAS anxious were directly associated with decreased difficulties in engaging in goal directed behaviors (F = 4.76; p< .05).

Implications: Although this study had a small sample, the findings support the thesis that intensive cognitive behavioral programs targeting criminogenic needs impact general attachment styles and risk factors associated with recidivism. The specific factors within the treatment process should be explored further to understand their impact on attachment and criminogenic needs.