State Point-of-Sale Comprehensive Background Check Policies and County Firearm Homicide Rates.

Importance: Background checks aim to prevent illegal firearm transfers. Existing research has shown little evidence for the effect of point-of-sale comprehensive background check (CBC) policies on firearm-related violence at the state level, but state-level estimates may mask local variation. Objective: To estimate the association between state-level point-of-sale CBC policies and county-level firearm homicide rates in the US. Design, Setting, and Participants: This cross-sectional study used a difference-in-differences analysis with synthetic controls. Six US states that implemented a point-of-sale CBC policy between 2013 and 2019 and did not have a permit-to-purchase policy from 2000 to 2021 (Colorado, Delaware, New Mexico, Oregon, Vermont, and Washington) and 8 eligible control states that did not have a CBC or permit-to-purchase policy from 2000 to 2021 and were similar to treated states in terms of state partisanship (Alabama, Arkansas, Kentucky, Louisiana, Maine, Minnesota, Mississippi, and West Virginia) were included. The unit of analysis was county-years from January 1, 2000, to December 31, 2021. Analyses were conducted from April 1, 2023, to December 15, 2025. Exposure: Implementation of point-of-sale CBC policies (that applied minimally to handguns) between July 1, 2013, and July 1, 2019. Main Outcomes and Measures: Annual county-level firearm homicide rates per 100 000 population were directly standardized to the age (<5 years, 10-year age bands from 5-84 years, and ≥85 years) and sex (female or male) distribution of the US population in 2000. The synthetic control method was used to estimate counterfactual posttreatment means. Results: The study population included 750 counties: 59 from Colorado, 3 from Delaware, 33 from New Mexico, 36 from Oregon, 14 from Vermont, 39 from Washington, and 566 from states that did not have a CBC policy (ie, eligible controls). In unadjusted models, state-level CBC implementation was associated with a mean of -2.13 (95% CI, -3.19 to -1.04) firearm homicide deaths per 100 000 population across counties; after adjusting for covariates, a mean of 0.13 (95% CI, -0.88 to 1.13) deaths per 100 000 population were found. While there was some heterogeneity in county-specific estimates in adjusted models (with point estimates ranging from -13.2 to 20.5 deaths per 100 000 population), nearly all 95% CIs crossed the null. Conclusions and Relevance: This cross-sectional study of the association between state-level CBC policies and county-level firearm homicide rates assessed county-level heterogeneity that may have been masked in prior state-level studies. As in prior state-level studies, no associations were found for adjusted results. Findings suggest opportunities for further research on the design, implementation, and enforcement of CBC policies.