Firearm Injuries in Maryland, 2005-2014: Trends, Recidivism, and Costs
AdvisorJohantgen, Mary E.
MetadataShow full item record
AbstractBackground: Violent injuries related to firearms are common in the U.S., whether accidental or intentional. Restrictions on use of Federal dollars for research on injury prevention involving firearms has limited our knowledge of how firearm injury impacts the health care system. The objectives of this study are to characterize firearm injuries (FI) in Maryland, quantify recidivism, and to describe hospital treatment and their associated costs for Maryland. Methods: ED and inpatient hospital records utilizing E codes consistent with FI were linked across visits to create unique cases from 2005-2014. Recidivism was defined as any subsequent ED visit or hospitalization for FI. The relationship of social determinants of health derived from US Census data to the rate of FI hospitalization by zip code were examined with generalized linear models, as were FI associated hospital costs. Results: Those with a FI are primarily single young black males, with overall hospitalizations decreasing over the time period. While 9% died in their initial FI, recidivism occurred in 3% of the individuals. Personal Disadvantaged (IRR = 1.13) and Working Disadvantaged (IRR = 1.04) factors were associated with increased rates of FI within zip codes. Hospital costs were significantly predicted by being self-pay/charitable and injury severity, with estimated mean costs for one FI of $47,364. In 2013, FI hospitalizations totaled $14m, of which 25% (n=129) accounted for over $10m. Discussion and Implications: FI hospitalizations are decreasing and are increasingly linked to social determinants of health, which require multifaceted interventions with short term goals of interrupting ongoing violence and long-term goals of preventing future violence. The states are absorbing much of the health cost burden. Further research is needed, which should include developing a registry linking hospitalizations, deaths, and crime data that can be used to evaluate trends and effectiveness of interventions.
University of Maryland, Baltimore