Exploring the longitudinal offending pathways of child sexual abuse victims: A preliminary analysis using latent variable modeling

Very little research has been conducted to show the way in which criminal behavior unfolds over the life-course in children who have been sexually abused, and whether it differs from the 'age-crime' patterns consistently documented in the criminology literature. This study investigated the temporal pathways of criminal offending between the ages of 10-25 years among medically confirmed cases of child sexual abuse (CSA), and considered whether abuse variables, offense variables, and the presence of other adverse outcomes, were associated with heterogeneity in offending pathways among CSA survivors. This study utilized data gathered as part of a large-scale study involving the linkage of forensic examinations on 2759 cases of medically ascertained CSA between 1964 and 1995, to criminal justice and public psychiatric databases 13-44 years following abuse, together with a matched comparison sample of 2677 individuals. We used the subsample of 283 offending individuals (191 victims: 92 comparisons) for whom complete offending data were available. We compared the aggregate age-crime curves for CSA victims and comparisons, and applied longitudinal latent class analysis to identify distinct subgroups of offending pathways between ages 10-25 years within the abuse sample. Four latent pathways emerged among sexually abused offenders, labeled: Early-Onset/High-Risk/Adolescence-Limited; Intermediate-Onset/Low-Risk/Adolescence-Limited; Late-Onset/Low-Risk/Slow-Declining; and Early-Onset/High-Risk/Persistent offenders. Age at abuse, the nature and frequency of offending, and mental health problems, were associated with the offending pathway followed by CSA victims. Consistent with criminological literature, findings indicate considerable heterogeneity in the longitudinal offending patterns of offenders exposed to CSA. Implications for clinical practice and directions for research are highlighted. (C) 2017 Elsevier Ltd. All rights reserved.