County Level Contributors to Geographic Variation in Medicare FFS Stroke Hospitalization Rates: A Cross-Sectional Study

BackgroundGranular regional stroke incidence data in the US is lacking. We sought to identify factors associated with county-level hospitalization rates and counties with hospitalization rates above or below expectation using publicly available data. MethodsThis cross-sectional study is based on the analysis of county-level three-year average stroke hospitalization rates (principal ICD-10 I63, I65-I66) per 100,000 population among Medicare fee-for-service (FFS) beneficiaries from 2018-2020 using data from the CDCs Interactive Atlas of Heart Disease and Stroke (ATLAS) and other sources. ATLAS provided reliable data on 3,198 (98.6%) counties and county-equivalents. Linear mixed models were fitted to investigate six sets of factors (Total of 61) associated with hospitalization rates in a serial additive stepwise fashion (i.e., demographics, overall population vascular risk factors, risk factor treatment, health delivery and access, environmental features, and socioeconomic status). We reported on the predicted hospitalization rates, marginal R2 of the fixed effects, the most impactful factors using average marginal effects, and characterized proportional difference between crude and predicted hospitalization rates. ResultsThe cohort of 3,198 counties and county-equivalents had a mean stroke hospitalization rate of 11.2 per 100,000 (SD= 2.6). Mean characteristics of included counties: 19.4% age [≥]65 years, 73% white, 7.6% coronary heart disease (CHD) prevalence, 38% hyperlipidemia prevalence, and 5.7 primary care physicians per 10,000. In the fully adjusted model, between-county unexplained variation remained moderately high (R2= 0.57). The most impactful factors associated with stroke hospitalization rates were prevalence of CHD, hypertension, smoking, nonadherence to antihypertensive medication, and the elevation of the county above sea level. Counties in the northwest United States generally had lower than expected hospitalization rates. ConclusionsConsiderable unexplained county-level variance in stroke hospitalization rates exists after accounting for a wide variety of known and potential predictors. Future work to clarify the mechanism of known predictors and explain variance may inform stroke mechanisms and interventions to improve systems of care.