Utility of Hospital Emergency Department Data for Studying Intimate Partner Violence
Saltzman, Linda E., Mahendra, Reshma R., Ikeda, Robin M., Ingram, Eben M., Journal of Marriage and Family
This article is in memory of Linda E. Saltzman, a Senior Scientist at the Centers for Disease Control and Prevention's Division of Violence Prevention, National Center for Injury Prevention and Control. As a well recognized expert in violence prevention, Dr. Saltzman provided consultation to federal and state health officials across the country and to researchers and violence practitioners, both nationally and internationally on topics such as public health surveillance of violence against women. She was a visionary in championing violence prevention efforts, and she will be greatly missed.
The authors examine 12 months of emergency department visit data (N = 2,521) from the National Electronic Injury Surveillance System All Injury Program and explore its utility for measuring and studying intimate partner violence. Given the dearth of national data on intimate partner violence-related injury and its potential value for public health surveillance and prevention, the data set appears promising for estimating national rates of emergency department visits. Missing perpetrator-patient relationship data limit estimation of intimate partner-related hospital visits for injury and pain, which precludes national rate estimation at this time, but the data are still useful for describing documented intimate partner violence cases and may be helpful in designing prevention strategies.
Key Words: emergency department, hospital, injuries, intimate partner violence, surveillance, violence against women.
Family researchers have been aware of the importance of intimate partner violence as an issue for many years (e.g., Finkelhor, Gelles, Hotaling, & Straus, 1983; Gelles, 1976), but the topic's importance as a public health issue has also been affirmed recently both in the United States and internationally (Department of Health and Human Services [DHHS], 1986; Heise & Garcia-Moreno, 2002; Saltzman, Green, Marks, & Thacker, 2000; Velzeboer, Ellsberg, Clavel Areas, & Garcia-Moreno, 2003). Instead of simply reacting to problems once they have occurred, a public health approach emphasizes prevention and focuses on behavior and its consequences that can be prevented; by preventing problems before they occur, we enable efficient use of extremely limited resources (Dahlberg & Krug, 2002). Because of the physical and mental health consequences (Coker, Smith, Bethea, King, & McKeown, 2000; Kernic, Wolf, & Holt, 2000) and disproportionate use of health care services (Miller, Cohen, & Rossman, 1993) associated with it, intimate partner violence has been a specific focus within several recently published federal research agendas (National Center for Injury Prevention and Control, 2002; National Research Council, 2004).
In October 1998, the U.S. Department of Health and Human Services and the U.S. Department of Justice cosponsored the Workshop on Building Data Systems for Monitoring and Responding to Violence Against Women, which brought together an invited group of researchers and practitioners from the public health and criminal justice fields. Participants addressed opportunities and challenges associated with public health surveillance-the ongoing and systematic collection, analysis, and interpretation of information (Centers for Disease Control and Prevention [CDC], 2000). Public health surveillance systems are typically developed with the capacity to collect, analyze, and disseminate data in a timely fashion for use in prevention and control activities; the data are essential for defining priorities quantitatively and for evaluating program effectiveness (Thacker, 2000). The needs for priority setting and program evaluation related to preventing intimate partner violence are key reasons for its inclusion in developing agendas. As with other preventable health-related conditions, surveillance for intimate partner violence is conducted for purposes much broader than estimating the magnitude of the problem. …