Computer-based screening may encourage discussions about domestic violence

May 23 2006

Computer screening may increase the odds that a woman at risk for domestic violence will talk to a health care professional in the emergency department about the topic but does not guarantee that domestic violence would be addressed, according to an article in the May 22 issue of Archives of Internal Medicine.

Reluctance to raise sensitive issues prevents many physicians from identifying victims of domestic violence in health care settings, according to background information in the article. The emergency department (ED), where patients who have experienced domestic violence often seek care, presents additional challenges, including time pressures and urgent medical needs. Previous studies have shown that patients are more likely to disclose sensitive information, including experiences with domestic violence, on computer-based screenings than on paper surveys or in personal interviews.

Karin V. Rhodes, M.D., then at the University of Chicago and now at University of Pennsylvania Hospital, Philadelphia, and colleagues tested a computer screening program in two emergency departments, one urban and one suburban. A total of 903 women who visited the emergency department between June 2001 and December 2002 participated in the study and were randomly selected to either complete the computer-based risk assessment or receive usual care. The computer program asked questions about a variety of health risks, including eight that addressed domestic violence; if a woman responded positively to any of the eight questions, an alert advising the physician to assess her for domestic violence appeared on a printout that was then stapled to the patient's chart. Participants were audiotaped during their interactions with physicians and completed an exit questionnaire, which contained the same questions about domestic violence as the computer screening, before leaving the ED. Researchers reviewed the tapes to determine if domestic violence was discussed, disclosed or treated during each encounter.

Of the 903 women who completed the exit questionnaire, 26 percent at the urban ED and 21 percent at the suburban ED indicated they were at risk for domestic violence. A total of 871 women (331 from the suburban location and 540 from the urban site) were successfully audiotaped; among women who were audiotaped and disclosed domestic violence on the exit questionnaire, those who completed the computer screening were more likely to talk to a physician or nurse practitioner about domestic violence and twice as likely to disclose domestic violence during the ED visit than those who received standard care. In the urban ED, women who completed the computer survey were more likely than those who received routine care to discuss domestic violence with their clinician (56 percent vs. 45 percent), disclose their own domestic violence situations (14 percent vs. 8 percent) or receive care or referrals for domestic violence (8 percent vs. 4 percent, or 57 percent vs. 43 percent of the women who disclosed domestic violence). Women at the suburban site were much less likely to discuss (11 percent in the computer screening group and 9 percent in the standard group) or disclose domestic violence (5 percent in both groups).

At both sites, only 48 percent of the women (17 percent in the suburban ED and 61 percent in the urban ED) whose computer surveys generated a domestic violence prompt for the physician had a discussion about domestic violence during their visit. "Domestic violence is a highly prevalent condition, but detection in the ED remains elusive," the authors conclude. "We found that female patients will disclose their domestic violence risk to a computer. Our study both supports the potential for computer screening to increase identification and referral for domestic violence and raises the concern that ED physicians, particularly suburban physicians, may need additional training to adequately recognize and respond to chronic, complex psychosocial issues."