Abstract P267: Geographic Patterns of Cardiovascular Disease Risk Factor Prevalence in the Continental US
Objectives: Our understanding of geographic variation in cardiovascular disease (CVD) risk factors is based upon self-reported variables or geographically limited coverage. Our objective was to explore geographic variation in measured hypertension, measured diabetes, measured dyslipidemia, and self-reported current smoking prevalence. Methods: We used baseline data from the REasons for Geographic and Racial Differences in Stroke (REGARDS) cohort, whose community-dwelling participants were recruited nationally between 2003 and 2007. Participants underwent a telephone interview and in-home examination. Hypertension, diabetes, and dyslipidemia were based on physiologic measures or reported medication use. Current cigarette smoking was self-reported. Using participants’ residential latitude and longitude, we tested for clustering of each risk factor using the difference in Ripley’s K functions test and, when we found evidence of clustering, used thin plate regression splines (TPRS) in a logistic regression framework to create age- race-, and sex-adjusted maps of risk factor prevalence. Results: Risk factor status and location data were available for 27,787 of the 30,239 participants (92%). Mean (±SD) age of these participants was 65(±9) years, 41% were black, 55% were women, 59% had hypertension, 22% had diabetes, 54% had dyslipidemia, and 15% were current smokers. We found statistically significant geographic clustering of hypertension, diabetes, and smoking prevalence, but not dyslipidemia. The regions with the highest prevalence varied across risk factors (Figure 1). Conclusions: Louisiana and Mississippi might require the most intense management of CVD risk factors. These maps show variation across and within administrative units, providing an accurate representation of geographic variation in risk factor prevalence. High resolution maps could be put to use by healthcare organizations to justify requests for higher reimbursement rates based upon local population health.