A comparison of 2 protocols to detect intimate partner violence

Intimate partner violence (IPV) frequently results in maxillofacial injuries. The purpose of this study is to compare 2 protocols to identify women who self-report IPV-related injuries.

Using a cross-sectional study design, we enrolled a sample of injured women who presented to the emergency department (ED) for evaluation and treatment. The predictor variable was the protocol used to identify injury etiology, that is, an innovative diagnostic protocol (DP) or the ED's standard operating procedure (SOP). The outcome variable was self-reported injury etiology categorized as IPV or other. Descriptive and bivariate statistics were computed. Sensitivity, specificity, positive predictive value, negative predictive value, and odds ratios were computed (with P < or = .05 being considered statistically significant).

The sample was composed of 286 women with a mean age of 45.8 years. The self-reported prevalence of IPV-related injuries was 11.5% and 5% for the DP and SOP, respectively (P = .03). The diagnostic statistics for the DP and SOP, respectively, were as follows: sensitivities, 94% and 50%; specificities, 76% and 95%; positive predictive values, 34% and 50%; and negative predictive values, 98% and 95%. In the adjusted model the women evaluated with the DP had a 38-fold increased likelihood of reporting IPV (P < .01) when compared with the SOP.

These results suggest that the proposed DP is associated with an increased frequency of self-reported IPV-related injuries when compared with the ED's SOP.