Gender differences in substance abuse severity and psychiatric severity: Can personality dimensions and personality disorders explain them?

Personality and gender have frequently been shown to explain the same underlying substance abuse and psychiatric symptomatology. Therefore, many researchers have speculated that gender and personality may be confounded in ways which obfuscate findings. Few attempts have been made to examine the independent roles these variables play in determining symptom severity. Most research studies in this area have used only one of these variables, making it impossible to know whether personality and gender provide overlapping or unique contributions to the severity of symptoms. The goal of this study was to control for the potential confounds among gender, personality characteristics and measures of psychiatric and substance abuse severity by examining them simultaneously. Including all of these variables in the same set of analyses allows one to determine whether they contribute independently to symptom severity or whether they are partially or completely overlapping. The sample included a heterogenous substance-abusing population (n = 339) with approximately equal male and female representation. Structural equation modeling was used to examine the relationships among gender, personality characteristics, and the two outcome variables, psychiatric severity and substance abuse severity. Substance abuse severity was best predicted by the personality traits, Negative Affectivity and Behavioral Disinhibition without any of the direct or indirect effects of gender. Psychiatric severity was best predicted by both level of Negative Affectivity and gender, without any of the direct or indirect effects of Behavioral Disinhibition. More specifically, the relationship between gender and psychiatric severity was shown to be mediated by the personality trait Negative Affectivity. These analyses suggest that more attention should be given to personality pathology in conceptualizing and planning targeted treatments for substance use disorders. Using personality traits as matching variables, rather than gender, may improve the efficacy of treatment interventions in this population which has extremely high rates of personality pathology