OBJECTIVES: Articles describing the epidemiology of intimate partner homicide (IPH) have often been positioned as one-time research projects utilizing a single data source for case identification. However, researchers without access to centralized repositories for data pertinent to IPH can ascertain cases by using multiple extant data sources. The authors describe a surveillance system that can serve as a model for state health departments and others seeking to quantify and characterize violent deaths related to intimate partner relationships on an ongoing basis. METHODS: Retrospective surveillance was conducted to identify and characterize deaths related to intimate partnerships by linking the following data sources: death certificates, newspaper articles, law enforcement reports, and medical examiners' records. RESULTS: The authors identified at least 34% more IPHs using multiple data sources than would have been recognized solely using Supplementary Homicide Reports--the most frequently cited data source for IPH case ascertainment--and 22% more cases than newspaper data alone would have allowed for. CONCLUSIONS: While it was discovered that at least 181 IPHs--and at least 128 other deaths related to intimate partnerships--occurred in Michigan from 1999-2001, this frequency of occurrences is probably conservative. This limitation is due, in part, to the researchers not having access to information from all possible data source contributors. However, in the absence of statewide data systems, the authors demonstrate the utility of using multiple data sources for violent death surveillance to address incomplete case ascertainment.
OBJECTIVES: Articles describing the epidemiology of intimate partner homicide (IPH) have often been positioned as one-time research projects utilizing a single data source for case identification. However, researchers without access to centralized repositories for data pertinent to IPH can ascertain cases by using multiple extant data sources. The authors describe a surveillance system that can serve as a model for state health departments and others seeking to quantify and characterize violent deaths related to intimate partner relationships on an ongoing basis. METHODS: Retrospective surveillance was conducted to identify and characterize deaths related to intimate partnerships by linking the following data sources: death certificates, newspaper articles, law enforcement reports, and medical examiners' records. RESULTS: The authors identified at least 34% more IPHs using multiple data sources than would have been recognized solely using Supplementary Homicide Reports--the most frequently cited data source for IPH case ascertainment--and 22% more cases than newspaper data alone would have allowed for. CONCLUSIONS: While it was discovered that at least 181 IPHs--and at least 128 other deaths related to intimate partnerships--occurred in Michigan from 1999-2001, this frequency of occurrences is probably conservative. This limitation is due, in part, to the researchers not having access to information from all possible data source contributors. However, in the absence of statewide data systems, the authors demonstrate the utility of using multiple data sources for violent death surveillance to address incomplete case ascertainment.