Spatiotemporal Trends in Heart Disease Mortality Rates by Age, Race, and Sex

• Activity Number: 294 - Epidemiologic Methods for the Re-Use of Existing Data
• Type: Topic Contributed
• Date/Time: Tuesday, July 31, 2018 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Harrison Quick* and Adam Vaughan and Michele Casper and Linda Schieb and Michael Kramer
• Companies: Drexel University and Centers for Disease Control and Prevention and Centers for Disease Control and Prevention and Centers for Disease Control and Prevention and Emory University

Abstract

Previous work has demonstrated declines and changing geographic patterns in heart disease mortality rates by race and sex in the United States. In this talk, we present results of an assessment of these spatiotemporal trends in heart disease mortality by age, race, and sex over the 43-year period from 1973 to 2015. In addition to the computational burden associated with modeling over 3.2 million data-points from 3,115 counties, 43 time-periods, and 24 age/race/sex combinations, over 83% of these data-points are based on fewer than 10 deaths. As a result, we rely on a multivariate spatiotemporal Bayesian model to obtain reliable rate estimates. Based on this work, we identify unexpected trends in heart disease mortality among those aged 35-44, in which a large percent of counties experience an inversion in rates for black women and white men - i.e., that rates for black women are higher than rates for white men - suggesting that in these counties the impact of socioeconomic factors may be overwhelming the female biological protectiveness for heart disease mortality in younger age groups.