The impact of hospital-based violence intervention programs on posthospitalization care among patients injured by gunfire: A review of a multicenter database.

BACKGROUND: Survivors of gunshot wounds frequently require structured posthospitalization services to prevent functional decline and complications, yet access to rehabilitation and home health care remains inequitable. Hospital-based violence intervention programs may influence postdischarge referral pathways, but prior evaluation has been limited to single-center experience. METHODS: We conducted a retrospective cohort study using the American College of Surgeons Committee on Trauma Firearm Study Research Data Set, including 12,134 adult firearm-injured patients treated at 121 participating trauma centers. Patients were stratified by receipt versus nonreceipt of hospital-based violence intervention program services. Primary outcomes were recommendations for rehabilitation services and home health care at hospital discharge. Because patients were clustered within centers and hospital-based violence intervention program-exposed patients were unevenly distributed across them, primary multivariable analyses were mixed-effects logistic regression models with a random intercept for treating center. RESULTS: Among 12,134 patients, 1,413 (11.6%) received hospital-based violence intervention program services. Rehabilitation services were recommended more often for hospital-based violence intervention program participants (19.4% vs 12.5%; P < .001), as was home health care (13.7% vs 8.1%; P < .001). After adjustment and accounting for center, hospital-based violence intervention program participation remained independently associated with rehabilitation recommendation (adjusted odds ratio, 1.34; 95% confidence interval, 1.09-1.65) and home health care recommendation (adjusted odds ratio, 1.35; 95% confidence interval, 1.06-1.73). Between-center variation was substantial (intraclass correlation, 0.193 and 0.292, respectively). After accounting for center and adjusting for ethnicity, employment status, and sex, race was not independently associated with rehabilitation recommendation (Black versus White, adjusted odds ratio, 0.96; 95% confidence interval, 0.81-1.14). CONCLUSION: Hospital-based violence intervention program participation was independently associated with higher odds of recommendations for posthospitalization services. Variation between treating centers accounted for a substantial share of the variance in whether these services were recommended, indicating that where a patient is treated is a major determinant of postdischarge referral and a potential target for standardization.