APPENDIX 4.2
PSYCHOPATHOLOGY AND SEXUAL OFFENDING
Mental health problems such as depression are negatively associated with future offending among adolescent offenders (Katsiyannis, Zhang, Barrett, & Flaska, 2004; Vermeiren, Schwab-Stone, Ruchkin, De Clippele, & Deboutte, 2002), which suggests psychopathology may, in fact, have a protective effect. However, problems such as depression could still play a role in the onset of sexual offending if, for example, they led to the use of sex as a coping strategy and if using sex as a coping strategy increased the likelihood of having sexual contact with a child (Cortoni & Marshall, 2001).
Leue, Borchard, and Hoyer (2004) reported on the psychiatric morbidity of 55 adult sex offenders, 21 of whom had offended against children. The prevalences were high and probably reflected the fact that the sample was recruited from a forensic psychiatric setting; for example, 69% met diagnostic criteria for any anxiety disorder during their lifetime, and 56% met diagnostic criteria for any mood disorder during their lifetime. Dunsieth et al. (2004) also reported high prevalence rates in a sample of 113 male sex offenders seen at a residential treatment facility; for example, 58% met diagnostic criteria for a mood disorder, and 23% met diagnostic criteria for an anxiety disorder. The 84 paraphilic sex offenders, one half of whom were pedophiles, were more likely to have any mood disorder, anxiety disorder, or impulse control disorder than the 26 nonparaphilic sex offenders (3 men were not classified as either paraphilic or nonparaphilic). Långström, Sjöstedt, and Grann (2004) reported on the psychiatric comorbidity of all adult male sex offenders released from prison in Sweden between 1993 and 1997 based on hospital admissions (N = 1,215 sex offenders; 522 had offended against children). Of those who offended against children, relatively few (4.8%) had any psychiatric disorder, and the most common diagnoses for both groups of sex offenders were substance abuse, substance dependence, and personality disorders. Långström et al. suggested that an explanation for the relatively low prevalences found in this study was that the diagnoses were based on hospital admissions, and only some diagnoses were the focus of the inpatient assessments. Finally, N. C. Raymond, Coleman, Ohlerking, Christenson, and Miner (1999) reported on the psychiatric diagnoses of 45 pedophilic sex offenders; a majority had a diagnosable mood disorder (67%) or anxiety disorder (64%) at some point in their lives, and one half (53%) met diagnostic criteria for another paraphilia.
All of these studies indicate there is a great deal of psychiatric comorbidity among adult sex offenders against children. What they do not indicate is whether sex offenders against children differ from other offenders. Only studies that directly compare adult sex offenders and other offenders, using the same measures and recruitment methods, would reveal if psychopathology might be a unique factor. In addition, studies would need to account for additional factors that might explain the comorbidity of psychopathology and sexual offending against children and establish that psychopathology was present before the onset of sexual offending against children (e.g., by reviewing prior mental health records).
1 The logic and available inference of these types of specificity design studies—studies that compare two or more groups on theoretically relevant variables—has been discussed by Garber and Hollon (1991). Briefly, these specificity design studies allow the researcher to determine if a feature of interest is unique (specific) to pedophiles or to sex offenders against children. Failing to show a difference between groups would rule out a simple causal model, that is, that the feature of interest is a necessary and sufficient cause of pedophilia or sexual offending against children. It does not rule out, however, causality in a more complex model that involves multiple factors or interactions between factors.
2 Because very different definitions and measures of childhood sexual abuse were used, these percentages should not be interpreted as prevalence estimates. However, because each study in Seto and Lalumière’s (2007) meta-analysis used the same definition and measure of childhood sexual abuse in comparing adolescent sex offenders and other adolescent offenders, the difference between groups can be meaningfully interpreted.