Health conditions associated with sexual assault in a large hospital population

Objective: To develop a clinical informatics approach to identify patients with a history of sexual assault and to characterize the clinical risk factors and comorbidities of this population in a sex-stratified manner. Methods: We developed and applied a keyword-based approach to clinical notes to identify patients with a history of sexual assault in the Vanderbilt University Medical Center (VUMC) electronic health record from 1989 to 2021. Using a phenome-wide association study (PheWAS), we then examined diagnoses that co-occurred with evidence of sexual assault. We also examined whether sex assigned at birth modified any of these associations. Results: Our keyword-based algorithm achieved a positive predictive value of 90.4%, as confirmed by manual patient chart review. Out of 1,703 diagnoses tested across all subgroup analyses, we identified a total of 465 associated with sexual assault, many of which have been previously observed in the literature. Interaction analysis revealed 55 sex-differential phenotypic associations. Conclusions: In a large hospital setting, disclosures of sexual assault were associated with increased rates of hundreds of health conditions.