A Multilevel Analysis of County-Level Sociodemographic Factors and State-Level Firearm Policies Associated with Suicide Mortality Rates in the United States

Background and Purpose: Suicide mortality varies substantially across geographic areas, reflecting spatial disparities in underlying social, demographic, and policy contexts. Although prior research has identified a range of individual and structural correlates, relatively little is known about how county-level sociodemographic conditions and state-level firearm policies simultaneously shape suicide mortality within a multilevel framework. This study examines the combined influence of county- and state-level factors on county-level suicide mortality rates in the United States.

Methods: The analytic sample included 932 counties nested within 35 states. County-level suicide death counts in 2023 were obtained from the Centers for Disease Control and Prevention (CDC). County-level predictors were drawn from the American Community Survey (ACS) 5-year estimates and included divorce rate, uninsured rate, poverty rate, educational attainment (percentage of adults aged 25 and older with a bachelor’s degree or higher), unemployment rate among individuals aged 20–64, median age, proportion of males in the population, and the Rural-Urban Continuum Code (RUCC) as a continuous measure of rurality. State-level firearm policy variables were obtained from the RAND Corporation database and included three policies associated with suicide prevention: waiting period laws, minimum age requirements for firearm purchase, and child access prevention (CAP) laws, each coded as 1 if in effect in 2023 and 0 otherwise. A multilevel negative binomial regression model was estimated to account for overdispersion in suicide death counts and clustering of counties within states. The natural logarithm of county population size was included as an offset term, such that the model estimates suicide mortality rates and coefficients can be interpreted as incidence rate ratios (IRRs).

Results: At the county level, higher divorce rates (IRR = 1.06, 95% CI [1.05, 1.07]), older median age (IRR = 1.02, 95% CI [1.01, 1.02]), a higher proportion of males (IRR = 1.01, 95% CI [1.00, 1.01]), and greater rurality (IRR = 1.06, 95% CI [1.05, 1.08]) were associated with higher county-level suicide mortality rates. Higher educational attainment was associated with lower county-level suicide mortality rates (IRR = 0.99, 95% CI [0.99, 1.00]). A higher uninsured rate was also associated with lower county-level suicide mortality rates (IRR = 0.98, 95% CI [0.98, 0.99]). At the state level, counties located in states with waiting period laws had lower county-level suicide mortality rates (IRR = 0.77, 95% CI [0.67, 0.90]) than counties in states without such laws, corresponding to an approximate 23% reduction. Minimum age requirements and CAP laws were not significantly associated with county-level suicide mortality rates.

Conclusions and Implications: These findings underscore the importance of county-level demographic and geographic factors in shaping suicide mortality outcomes. The association between waiting period laws and lower suicide mortality rates suggests that delaying access to lethal means may reduce suicide risk at the population level. Together, the results highlight the need for integrated, multilevel prevention strategies that address local contextual conditions alongside state-level firearm policies to reduce suicide mortality.