Methods: Latent Profile Analysis (LPA)in Mplus 5.12 was used to investigate patterns of adult depression, anxiety, and substance abuse problems. Multinomial logistic regression models were then used to compare the derived subgroups (based on these patterns)according to prospectively measured physical and sexual abuse, after controlling for gender and childhood SES.
Results: A three-class model was found to be best fitting and represented the most parsimonious conceptual interpretation of the observed data patterns. The resulting three subgroups based on these profiles were: (1) a no disorder group(n=276, 82% of the analysis sample);(2) a mild to moderate, comorbid group (mild-moderate depression, mild anxiety, and moderate-high substance abuse, n=25, 7.4%) ; and (3) a moderate to severe depression and anxiety group (moderate-severe depression, severe anxiety, and low substance abuse, n=35, 10.4%). Multinomial logistic regression analyses showed that childhood sexual abuse was associated with an increased likelihood of being classified in the moderate to severe depression and anxiety group compared to no disorder group(OR=3.9, p<.01). After controlling for gender and SES, the association remained marginally significant (OR=2.4, p<0.1). However, childhood physical abuse did not help differentiate among the groups derived from the first-step LPA.
Conclusions and Implications: Analyses found that childhood sexual abuse is a precursor to adult mental health and substance use problems. Findings suggest the need for focused analyses of adulthood sexual abuse sequalae. This study also supports a call for preventive intervention programs focused on reducing sexual abuse and on enhancing protection from the ill effects of abuse experiences.