Domestic Violence in a Forensic Mental Health Sample

BACKGROUND AND PURPOSE. Both men and women perpetrate and are victims of domestic violence (DV), leading some to propose a gender symmetry model of DV based on conflict theory (Straus, 2006). Others argue that a naïve gender symmetry model ignores the context and impact of violence (Johnson, 2006) or fails to account for motivational differences by gender (Swan & Snow, 2006). The gendered nature of DV is an important issue which requires more attention, and with a focus on different populations. When we look at marginalized populations with greater cumulative risk factors for DV, such as the mentally ill or those involved in the criminal justice system, we may see different predictors of DV. The current study adds to this debate by examining DV in a sample of incarcerated men and women in treatment for mental illness and in a matched comparison sample. METHODS. Using data collected in individual interviews using the WMH-CIDI, 75 incarcerated men and women on a mental health unit were matched to a sample of 75 men and women drawn from the National Comorbidity Survey Replication (Kessler, 2004) using propensity scores. Binary regression equations test the gender symmetry hypothesis and model severe and moderate DV victimization with family and teen dating violence, marital power, psychological distress, DSM diagnoses, and other CID variables. RESULTS. Annual prevalence of DV victimization in the research sample is 30.8% for women and 41.2% for men (p<.05). Bivariate analysis found both moderate and severe DV victimization correlated with family of origin violence, teen dating violence, and DV perpetration. Moderate DV is also predicted by diagnoses of substance use disorder and an absence of major depressive disorder. Neither gender nor sample predicted DV, with the only predictor of either moderate or severe DV victimization being DV perpetration. IMPLICATIONS. These data support a naïve gender symmetry hypothesis, a lifelong pattern of violence experience, and an argument that violence is often mutual in a forensic mental health population. These results suggest one reason both incarceration and batterer programs have limited specific DV deterrence effects is not only that some offenders lack a “stake in conformity” (Feder & Dugan, 2002; Sherman, et al., 1992), but some offenders are stimulated to violence by their own victimization. Gender symmetry in a forensic mental health population supports emergent DV typologies and theories of women's DV which account for the context of their violence. Primarily African American and lower income, these women may choose to fight back rather than call the police or use social services due to negative experiences with these systems. In practice, those who work in psychiatric and correctional settings are ill advised to incorporate the same approaches to DV intervention that are commonly practiced in traditional settings. Our data suggest the importance of affective and trauma disorders in men's and women's DV in this population. DV approaches for women and men marginalized by psychological distress, social class, and race should be tailored to those realities.