A6stract. Objective: To determine the rate at which a group of women visiting the ED for reasons other than intimate violence return to the ED at a later time for intimate-violence-related injuries. Methods: Retrospective cohort study of a group of women with intimate-violence-related injuries on an index visit and a matched comparison group. Return visit rates to the ED for intimate violence injuries over the next 5 years were then compared. Results: The 95 women in each group were followed an average of 57 months. M the ED for reasons other than battering injuries. One ED-based study estimated that the incidence of acute (c1 month) domestic violence among women with a current male partner is 11.7%. Only about one fourth of these women presented for care because of trauma, and most did not have abuse documented in the medical record.' In other words, most women in violent relationships go to the ED for reasons other than acute injury and there is no documentation in the medical record of their abuse This realization h a s moved the concept of identifying violent relationships from asking only women with suspicious injuries, to asking all women. Many organizations have called for universal screening for intimate violence in all female p a t i e n t P ; however, the mechanism to do this in a busy ED, with limited available staff, has yet to be answered. Ideally, screening for any "disease" should be inexpensive, noninvasive, and safe. It should have a high sensitivity, but a low false-positive rate. Positive screens should result in a n intervention that improves outcomes.s If one goal of screening "asymptomatic" patients for violent relationships is to prevent further injury, it would be important to know whether "asymptomatic" women in the ED ever return to the ED for intimate-violence-related injuries. The purpose of this study was to determine t h e rate at which women visiting t h e E D for reasons other than intimate violence injury return to the ED at a later time with intimate violence injuries.
METHODSStudy Design. This was a retrospective cohort study of women with intimate-violence-related injuries on an ED index visit a n d a matched comparison group. The study was approved by the Truman Medical Center (TMC)/University of Missouri-Kansas City (UMKC) institutional review board.Study Setting and Population. The study took place at TMC, a n inner-city county hospital that serves a s the primary teaching hospital for the UMKC School of Medicine. It currently has a patient volume of >45,000 visits of patients aged 1 1 5 years. There was no formal screening protocol for intimate violence in place during the active study period. Study Protocol. We first identified retrospectively all women aged 2 18 years diagnosed as having intimate-violence-related injuries between January 1, 1992, and J u n e 30, 1992. The women in this case group were identified by the presence of the ICD-9 code6 diagnosis of 995.81 (adult maltreatment syndrome) in one of 3 diagnostic code