Background: In population-based studies, the quality of care delivered to injured patients is commonly judged by hospital survival rates. Evidence suggests injured patients surviving hospitalization remain at risk for death from their injuries after discharge. Patient characteristics associated with higher risk of late death are not completely defined.
Methods: The National Death Index is a government-maintained database composed of death certificate records from all decedents in the United States. Patients in a trauma registry were cross-linked to decedents in National Death Index on the basis of Social Security number or other unique identifiers. Decedents' time from injury to death was calculated. Logistic regression models were fit to those who died at hospital discharge and those who died in the first year after injury.
Results: Among 4293 hospitalized injured patients recorded in a trauma registry, 157 died during hospitalization. Among the 4136 discharged alive, 91 patients were linked to death certificate records filed in the 365 days after discharge. Patients over the age of 65 had a 15-fold greater odds of death than younger patients.
Conclusion: Trauma registry data cross-linked to vital statistics records is practicable. Patients who die in the year after injury differ from the traditional population used to evaluate quality of trauma care, and new standards are needed that evaluate long-term survival.