Predicting treatment outcome for child sexual abusers

A total of 122 child molesters were studied to investigate the extent to which Alford's sex offender taxonomy and Finkelhor's Multifactor Theory would predict treatment response. The taxonomy was constructed incorporating two dimensions: perpetrator type and primary motivation to commit abuse. The results indicated that regressed offenders whose primary motivation to commit child abuse was to meet non-sexual needs made more favorable progress and had a higher ratio of therapy successes than comparison groups. A factor analysis of variables constructed from Finkelhor's theory produced seven factors. Three of these factors predicted treatment progress and one predicted termination outcome. Statistical analyses indicated that perpetrators 1) whose family of origin modelled child abuse, and 2) who were rejected by adult peers but, 3) who did not have any behavioral deficits such as an impulse control problem, responded more favorably in therapy than perpetrators who did not share these characteristics. Perpetrators with impulse control problems and/or severe emotional immaturity tended to be therapy failures. Although the Alford sex offender taxonomy and Finkelhor factors were significant predictors of treatment effects, they accounted for considerably less variance than number of treatment sessions attended and therapists' ratings of attitudes and behavior during treatment.