Background: Trauma system planners use patient home address as a proxy for injury location, although this proxy has not been validated. We sought to determine the precision of this proxy by evaluating the relationship between the location of injury death and the location of residence.
Methods: This national descriptive analysis used the Multiple Cause of Death data files from 1999 to 2006 to determine the proportion of subjects in which county of residence (RC) matched county of death for all US injury deaths. Subgroup analyses were completed by age and injury intentionality using two sample tests of proportions. χ(2) tests were used to evaluate differences in concordance over time and by size of the RC.
Results: Analysis included 3,141 US counties and 1,255,881 subjects. A total of 73.4% of subjects died in the RC and 87.7% died in the RC or a contiguous county. Intentional injury deaths were more likely than unintentional to happen within a decedent's RC (85.1% vs. 68.1%, p < 0.001) and within the RC or contiguous county (93.4% vs. 85.2%, p < 0.001). Adult injury deaths were more likely than pediatric to happen within a decedent's RC (73.6% vs. 68.4%, p < 0.001) and within the RC or contiguous county (87.9% vs. 84.2%, p < 0.001). Subjects from larger counties were more likely to die within the RC or a contiguous county (same p < 0.001, same or adjacent p < 0.001).
Conclusion: The preponderance of fatal injury deaths occur close to home. This supports the practice of trauma system's planning using home location available in administrative data to proxy injury location.