Background: Domestic violence (DV) is prevalent across all racial and socioeconomic classes in the United States. Little is known about whether physicians differentially screen based on a patient's race or socioeconomic status (SES) or about resident physician screening attitudes and practices.
Objective: To assess the importance of patient race and SES and resident and clinical characteristics in resident physician DV screening practices.
Design, participants: One-hundred and sixty-seven of 309 (response rate: 54%) residents from 6 specialties at a large academic medical center responded to a randomly assigned online survey that included 1 of 4 clinical vignettes and questions on attitudes and practices regarding DV screening.
Measurements: We measured patient, resident, and clinical practice characteristics and used bivariate and multivariate methods to assess their association with the importance residents place on DV screening and if they would definitely screen for DV in the clinical vignette.
Results: Residents screened the African-American and the Caucasian woman (51% vs 57%, P = .40) and the woman of low SES and high SES (49% vs 58%, P = .26) at similar rates. Thirty-seven percent of residents incorrectly reported rates of DV are higher among African Americans than Caucasians, and 66% incorrectly reported rates are higher among women of lower than of higher SES. In multivariate analyses, residents who knew where to refer DV victims (adjusted odds ratio [AOR] = 3.54, 95% confidence interval [CI]: 1.43 to 8.73) and whose mentors advised them to screen (AOR = 3.46, 95% CI: 1.42 to 8.42) were more likely to screen for DV.
Conclusion: Although residents have incorrect knowledge about the epidemiology of DV, they showed no racial or SES preferences in screening for DV. Improvement of mentoring and educating residents about referral resources may be promising strategies to increase resident DV screening.