Abstract WP215: Population Attributable Risks For Potentially Modifiable Risk Factors Of Cerebrovascular And Cardiovascular-related Mortality In The National Health And Nutrition Examination Survey

Abstract

Introduction: Stroke and heart disease are leading causes of mortality in the United States. We studied population attributable risks (PARs) for potentially modifiable stroke and cardiovascular-related mortality risk factors in the National Health and Nutrition Examination Survey (NHANES)-National Death Index (NDI) linked dataset. Methods: Adult NHANES participants surveyed from 1999-2014 with linked NDI data were included in this sample-level analysis. The primary outcome was death secondary to stroke (ischemic or hemorrhagic) or cardiovascular disease as recorded in NDI. Exposures were behavioral, environmental, clinical, and laboratory variables dichotomized at the sample median collected in NHANES. Missing data were imputed. PARs were calculated using adjusted hazard ratios from Cox proportional models. Results: We included 47,356 participants (median age 46 years [IQR 30-64]; 51.8% female; 261 stroke- and 1,112 cardiac-related deaths, 2.9%). Mean follow-up time was 97.2 months (SD 54.4 months). Among men < 65 years, characteristics with the top 3 PARs were elevated cholesterol (PAR 42%), elevated systolic blood pressure (PAR 30%), and low platelet count (PAR 23%) (Table 1). Among women < 65 years, not using food label nutritional fact panels (PAR 29%), prescription medication use (PAR 34%), and an elevated white blood cell count (PAR 27%) were the top 3 risk factors by PAR. Among men ≥ 65 years, not using food label nutritional fact panels (PAR 32%), elevated systolic blood pressure (PAR 22%), and cigarette smoking (PAR 20%) were the top 3 factors by PAR. Among women ≥ 65 years, not using food label nutritional fact panels (17%), elevated cholesterol (29%), and elevated creatinine (22%) were the top 3 risk factors by PAR. Conclusions: We provide age- and sex-specific adjusted PARs of potentially modifiable risk factors of stroke and cardiovascular-related mortality. These findings may inform targeted stroke and cardiovascular risk prevention strategies.