County-level social, preventive care, and behavioral correlates of colorectal cancer mortality in the United States: a nationwide ecological analysis.

BACKGROUND: Geographic disparities in colorectal cancer (CRC) outcomes remain substantial across U.S. counties. Associations of social conditions, preventive care, behavioral risk environments, and rurality with CRC mortality remain incompletely characterized in analyses addressing state-level clustering and spatial dependence. METHODS: We linked CDC PLACES, American Community Survey, United States Cancer Statistics, and 2023 National Center for Health Statistics rurality data in a nationwide county-level ecological study. Outcomes were age-adjusted CRC mortality and all-cancer mortality during 2019-2023 and CRC incidence during 2018-2022. Population-weighted core-adjusted linear models evaluated each exposure separately, adjusting for the county proportion aged ≥65 years, nonmetropolitan status where applicable, and state fixed effects. Standard errors were clustered by state using CR2 estimation with Satterthwaite degrees of freedom. RESULTS: The ACS-complete dataset included 3,142 counties; 3,134 remained after linkage to the 2023 NCHS rurality classification. Final fixed samples included 2,126 counties for CRC mortality, 3,005 for all-cancer mortality, and 2,628 for CRC incidence. In core-adjusted CRC mortality models, greater broadband deprivation (β = 1.82, 95% CI 1.44 to 2.20), higher current smoking prevalence (β =2.08, 95% CI 1.74 to 2.42), and nonmetropolitan status (β=2.73, 95% CI 2.35 to 3.11) were associated with higher mortality. Higher CRC screening uptake (β = -1.64, 95% CI -1.97 to -1.31) and recent dental visit prevalence (β = -1.84, 95% CI -2.27 to -1.40) were associated with lower mortality. Key associations were directionally consistent in log-rate and matched spatial error model sensitivity analyses, with similar patterns for all-cancer mortality and CRC incidence. CONCLUSIONS: CRC mortality was associated with adverse social and behavioral profiles, lower preventive care engagement, and nonmetropolitan residence. Overlap across CRC mortality, all-cancer mortality, and CRC incidence suggests that these indicators may reflect broader geographic cancer disadvantage rather than CRC-specific mechanisms. Integrated structural, preventive care, behavioral, and rural health strategies may help reduce county-level disparities in CRC outcomes.