Screening for Intimate Partner Violence in Health Care Settings
A Randomized Trial
- Harriet L. MacMillan, MD, MSc, FRCPC;
- C. Nadine Wathen, PhD;
- Ellen Jamieson, MEd;
- Michael H. Boyle, PhD;
- Harry S. Shannon, PhD;
- Marilyn Ford-Gilboe, RN, PhD;
- Andrew Worster, MD;
- Barbara Lent, MD;
- Jeffrey H. Coben, MD;
- Jacquelyn C. Campbell, PhD;
- Louise-Anne McNutt, PhD
- for the McMaster Violence Against Women Research Group
- Author Affiliations: Departments of Psychiatry and Behavioural Neurosciences (Drs MacMillan and Boyle and Ms Jamieson), Pediatrics (Dr MacMillan), Clinical Epidemiology and Biostatistics (Drs Boyle and Shannon), and Emergency Medicine (Dr Worster), McMaster University, Hamilton, Ontario, Canada; Faculty of Information and Media Studies (Dr Wathen), Arthur Labatt Family School of Nursing (Dr Ford-Gilboe), and Department of Family Medicine (Dr Lent), The University of Western Ontario, London, Canada; Department of Emergency Medicine and Community Medicine, West Virginia University School of Medicine, Morgantown (Dr Coben); School of Nursing, Johns Hopkins University, Baltimore, Maryland (Dr Campbell); and Department of Epidemiology, School of Public Health, University at Albany, State University of New York, Albany (Dr McNutt).
Abstract
Context Whether intimate partner violence (IPV) screening reduces violence or improves health outcomes for women is unknown.
Objective To determine the effectiveness of IPV screening and communication of positive results to clinicians.
Design, Setting, and Participants Randomized controlled trial conducted in 11 emergency departments, 12 family practices, and 3 obstetrics/gynecology clinics in Ontario, Canada, among 6743 English-speaking female patients aged 18 to 64 years who presented between July 2005 and December 2006, could be seen individually, and were well enough to participate.
Intervention Women in the screened group (n=3271) self-completed the Woman Abuse Screening Tool (WAST); if a woman screened positive, this information was given to her clinician before the health care visit. Subsequent discussions and/or referrals were at the discretion of the treating clinician. The nonscreened group (n=3472) self-completed the WAST and other measures after their visit.
Main Outcome Measures Women disclosing past-year IPV were interviewed at baseline and every 6 months until 18 months regarding IPV reexposure and quality of life (primary outcomes), as well as several health outcomes and potential harms of screening.
Results Participant loss to follow-up was high: 43% (148/347) of screened women and 41% (148/360) of nonscreened women. At 18 months (n = 411), observed recurrence of IPV among screened vs nonscreened women was 46% vs 53% (modeled odds ratio, 0.82; 95% confidence interval, 0.32-2.12). Screened vs nonscreened women exhibited about a 0.2-SD greater improvement in quality-of-life scores (modeled score difference at 18 months, 3.74; 95% confidence interval, 0.47-7.00). When multiple imputation was used to account for sample loss, differences between groups were reduced and quality-of-life differences were no longer significant. Screened women reported no harms of screening.
Conclusions Although sample attrition urges cautious interpretation, the results of this trial do not provide sufficient evidence to support IPV screening in health care settings. Evaluation of services for women after identification of IPV remains a priority.
Trial Registration clinicaltrials.gov Identifier: NCT00182468
