Violence Perpetration Among Patients Hospitalized for Unintentional and Assault-Related Firearm Injury: A Case-Control Study and a Cohort Study

Ann Intern Med. 2016 Dec 20;165(12):841-847. doi: 10.7326/M16-1596. Epub 2016 Oct 18.


Background: Hospital-based violence intervention programs typically focus on patients whose firearm injury occurred through interpersonal violence (assault). Knowledge of violence perpetration by victims of unintentional (accidental) firearm injury is limited.

Objective: To examine violence perpetration before and after a patient becomes hospitalized for firearm injury according to injury intent (intentional [assault] or unintentional [accidental]).

Design: A case-control study and a retrospective cohort study.

Setting: Hospitals in Washington.

Patients: Persons aged 15 years or older hospitalized for a firearm injury, other injuries, or a noninjury reason from 2006 to 2007.

Measurements: In the case-control study, the odds of violence-related arrest from 2001 through hospitalization by injury intent among 3 groups were compared. In the cohort study, the rates of violence-related arrest from hospitalization through 2011 by injury intent among 3 groups were compared.

Results: Patients with unintentional firearm injuries (n = 180) were more likely than those with other unintentional injuries (n = 62 795; odds ratio [OR], 2.01 [95% CI, 1.31 to 3.09]) and no injuries (n = 172 830; OR, 3.43 [CI, 2.22 to 5.32]) to have been arrested for a violent crime before hospitalization. Prior violence-related arrest did not differ between patients with assault-related firearm injuries (n = 339) and those with other assault-related injuries (n = 2342; OR, 1.10 [CI, 0.84 to 1.46]). During follow-up, the cumulative incidence of violence-related arrest for patients with unintentional and assault-related firearm injuries was 10% and 15% (subhazard ratio, 1.88 [CI, 1.11 to 3.17] and 1.61 [CI, 1.08 to 2.44]), respectively, compared with 1% for those without injuries.

Limitation: Exclusion of self-inflicted injuries, misclassification of intent, and ascertainment bias.

Conclusion: Some firearm injuries classified as accidental may indicate involvement in the cycle of violence and present an opportunity for intervention.

Primary funding source: City of Seattle and the University of Washington Royalty Research Fund.

MeSH terms

  • Accidents* / statistics & numerical data
  • Adolescent
  • Adult
  • Case-Control Studies
  • Female
  • Hospitalization
  • Humans
  • Male
  • Middle Aged
  • Retrospective Studies
  • Violence / statistics & numerical data*
  • Washington / epidemiology
  • Wounds, Gunshot / epidemiology*
  • Wounds, Gunshot / etiology
  • Wounds, Gunshot / psychology*
  • Young Adult