Homicide as a medical outcome: racial disparity in deaths from assault in US Level I and II trauma centers.

BACKGROUND

Since 1900, thousands of medical journal articles have been published on the topic of racial disparities in health and medical outcomes in the United States, including overlapping disparities based on health insurance status. But research on the question of such disparities in the medical treatment of injury from assault-matters of public safety, considerable public expense, and policy debate-is lacking.

METHODS

To determine differences by race and insurance status on death from intentional injury by others on and after trauma center arrival, propensity score matching is used to estimate adjusted mortality risk ratios by race and medical insurance controlling for facility, case, and injury characteristics. Analysis is based on a nationally representative sample of 100 Level I and II US trauma centers (National Trauma Data Bank 2005-2008) and includes 137,618 black and white assault cases aged 15 years and older: 35% white, and 65% black, with 46% of the whites and 60% of the blacks coded as uninsured.

RESULTS

Black patients showed higher overall raw mortality rates from assault than whites (8.9% vs. 5.1%), but after propensity score matching, the black to white adjusted risk ratio for death from assault (homicide) dropped significantly across firearm, cutting/piercing, and blunt injuries. After adjustment, estimated black deaths were 29% in excess of white deaths for firearm injuries, 36% in excess for cutting/piercing injuries, and 61% in excess for blunt injuries. Uninsured blacks comprised 76% of all excess trauma center deaths from assault.

CONCLUSIONS

Along with insurance status, and after excluding on-scene deaths, among patients brought to the Level I and II trauma centers, race is a substantial independent predictor of who dies from assault. Blacks, especially the uninsured, have significantly worse outcomes overall, but there is some evidence that this pattern is minimized at higher levels of injury severity.

LEVEL OF EVIDENCE

I, prognostic study.