Universal screening for interpersonal violence: inability to prove universal screening improves provision of services

Universal screening for interpersonal violence is recommended despite a lack of confirmed efficacy. We hypothesized that the detection of violence via universal screening would result in high intervention rates for victims.

Women aged 18-65 years presenting to an emergency department were screened using a standard protocol. Medical and social work records were reviewed for positively screened patients. Outcomes included whether victims received counseling/referral services. Secondary outcomes were the documentation of services offered and safety assessment performed.

A total of 1732 patients were evaluated; 615 (35.5%) responded positively to at least one query. Patients had a mean age of 34.7+/-12 years, 79% were non-white, 19% were married, and 76% had completed high school. Twenty-five out of 606 victims (4%) had documentation of violence. Residents were more likely than faculty or nurses to document domestic violence [3.3% (95% confidence interval 1.8-4.8%) versus 2.1 (0.9-3.4) versus 0.7 (0.0-1.4)]. The documentation of police contact, suicide/homicide risk, weapon presence, safety assessment and outside resource referrals occurred in less than 2% of charts. Only two victims were referred to social work (0.3%; 0-0.9%).

Even in an institution with a heavy emphasis and training on interpersonal violence and alternative mechanisms for universal screening we could not prove that the identification of victims resulted in counseling/referral being offered in the emergency department.