Following a Firearm Injury: Recidivism and Behavior Change

Vivian Lyons | 2019

Advisor: Ali Rowhani-Rahbar

Research Area(s): Public Health Practice, Social Determinants of Health


Firearm injuries remain an important and understudied public health problem. Besides the high cost of injury care, loss of work, potential need for long term care and emotional trauma for the individual injured, firearm injuries pose unique harm to the larger community as well. Individuals with a history of firearm injury are at substantially higher risk of subsequent rehospitalization for another firearm or assault-related injury, arrest for a firearm-related or violent crime, non-firearm-related nonviolent crime, or firearm-related death in the five years after discharge from the hospital. A number of hospitals across the country have created violence intervention programs to specifically help patients who sustain violent injuries. There exist only a few studies of the effectiveness of these programs, some of which had small sample sizes and produced mixed findings. We conducted the Helping Individuals with Firearm Injuries (HiFi) to address this gap. HiFi was a randomized controlled trial of an intervention program that combined a hospital-based intervention, structured community outreach program, and multi-agency attention. Using a Cox proportional hazards model, the Prentice Williams and Peterson multiple failure survival model and G-computation, we calculated the hazard ratios and survival ratios of the risk of arrest and subsequent injury following an incident firearm injury for the intent-to-treat and per-protocol effect estimates, using administrative data linkages to ascertain outcomes and censoring. We did not find an effect of the intervention assignment on either arrest or subsequent injury, although participation in the study was associated with a decreased risk of arrest. Next, we conducted a fidelity assessment of the HiFi intervention to ensure that our null findings were not due to poor intervention implementation, but also illustrates how future interventions in this population could be tailored to improve patient engagement. Finally, using survey data on firearm-related behaviors, we sought to describe changes in reported firearm-related behaviors among GSW patients following their injury. Our results suggest that patients with a GSW, especially firearm owners, may change their firearm-related behaviors following injury, some by increasing firearm-related safety and others by increasing frequency of behaviors that may place them at increased risk of subsequent injury. Overall, we found that patients with firearm injuries are potentially open to interventions that support behavior change. However, the CTI model is likely not intensive enough to change the high risk of subsequent arrest and injury in this population. Future studies aimed at reducing arrest and injury should use more intensive approaches.