Brief intervention for perpetration of intimate partner violence (IPV): simulation versus instruction alone.

OBJECTIVE: To compare two brief randomized computer-based interventions about perpetration of intimate partner violence (IPV). METHODS: The study was a prospective cross-sectional cohort study of patients during randomized 4-hour shifts in an urban hospital-based emergency department setting with an emergency medicine residency. The site is a level 1 trauma center emergency department with an annual census of 60,000 adults. A touch-screen computer program was developed in Visual Basic 2005 Studio with consecutive data entry screens. A series of questions to evaluative IPV knowledge, attitude, and practices (KAP) was given before and after the brief intervention. The subjects were randomized to one of two intervention groups. Both groups were shown a set of PowerPoint slides addressing IPV prevention, which was followed by either a control group with nothing further (control) or a five-minute simulation video depicting IPV perpetration in adults (SIM). The main outcome was the net improvement in KAP scores defined as number who improved minus number who did worse. Perpetrators and victims were assessed separately. Demographics were also obtained. For demographics, descriptive statistics and percentages were used. Wilcoxon signed rank test was used for pre-post test paired data. RESULTS: A total of 239 patients, presenting during 52 randomized four-hour shifts, completed the study; 118 were in the control group, 121 in SIM; 115 (48%) were male and 124 (51 %) female (9 unknown). Most participants were from 21 to 30 years old (38%); 77 (32%) were Hispanic and 100 (42%) were white. There was an overall improvement in responses to KAP questions with correct answers to all questions in 46% before vs 59% after the computer intervention (Diff 13% 95% confidence interval [CI] 4-22). The net improvement in KAP score in the control group was 8% and in the SIM group it was 22% (Diff = 15%, 95% CI = 6-24). Forty subjects (17%, 95% CI = 12-21) were perpetrators and 52 subjects (22%, 95% CI = 17-27) were victims. KAP scores were not significantly improved in the SIM group over the control for either perpetrators or victims. CONCLUSION: An IPV simulation video resulted in a greater percent of subjects showing improvement in knowledge, attitudes, and practices about IPV perpetration. A brief IPV simulation video intervention can be performed in a busy emergency department setting.

Adult intimate partner violence perpetrators are significantly more likely to have witnessed intimate partner violence as a child than nonperpetrators.

BACKGROUND: In a previous study, no association was found between intimate partner violence (IPV) victims and being an adult who witnessed IPV as a child (ACW). OBJECTIVE: The objective of the present study was to determine whether perpetrators of IPV (Perps) could be identified in a busy emergency department (ED) and whether Perps were more likely than non-Perps to be ACWs. The hypothesis was that Perps differed significantly from non-Perps in being ACWs, in being victims of IPV, and in demographics. METHODS: The design was a cross-sectional cohort of patients presenting to an academic ED during randomized 4-hour shifts. A choice of computer touch screen data vs paper format was offered. Data collected included demographics as well as scales to determine whether subjects were a Perp, victim, and/or ACW of IPV. Six validated scales were used to screen and confirm victims, Perps, and ACWs. Predictor variables were ACW, ongoing IPV, and demographics. RESULTS: Two hundred thirty-six subjects were entered, 207 had complete data sets. Forty-four (19%) were Perps. By univariate analysis, there was a significant correlation of Perps and ACW (P = .001 by single question) and between Perp and being victims (P = .001). No other univariate variables were significantly correlated. By regression analysis, significant predictors of Perp included ACW and spouse drug abuse. CONCLUSIONS: The Perps were identified in a busy ED setting. Perps were significantly more likely than non-Perps to be ACWs, but not more likely to be victims. Spouse drug abuse and ACW were the 2 significant predictors of Perp.