During the past 40 years, heart disease mortality has declined in the U.S., although there has been a shift in the concentration of high-rate counties from the Northeast to the South, according to an analysis based on a fully Bayesian spatiotemporal model.

Lead researcher Michele Casper, PhD, of the Centers for Disease Control and Prevention in Atlanta, and colleagues published their results online in Circulation on March 21.

For this analysis, the researchers used the model to document changing geographic patterns of heart disease mortality over time, geographic variation in rates of declining heart disease mortality and the changing magnitude of disparity in heart disease mortality among counties.

They obtained data on the annual number of heart disease deaths per county from 1973 to 2010 from the National Vital Statistics System of the National Center for Health Statistics. They calculated heart disease death rates per 100,000 and standardized the rates to the 2000 U.S. population. They also used a set of 3,099 counties from the contiguous 48 states and aggregated county-level, age-standardized counts and populations into two-year intervals.

The average percentage decline in heart disease death rates for the U.S. was 61.6 percent, but it ranged from 9.2 percent to 83.4 percent among counties. The counties with the slowest declines were primarily in Alabama, Mississippi, Louisiana, Arkansas, Oklahoma and Texas. The counties with the fastest declines were mostly in the northern part of the country as well as in Florida and South Carolina.

During the beginning of the study, the largest concentrations of high-rate counties were from the Northeast through Appalachia and the Midwest and along the coastal areas of North Carolina, South Carolina and Georgia, according to the researchers. By the end of the study, however, the largest concentrations of high-rate counties were in the South and southern Appalachia.

“Whereas most of the high-rate clusters were in the Northeast
and Midwest at the beginning of the study period, by 2009 to 2010, the vast majority of high-rate clusters were south of the Mason-Dixon line,” the researchers wrote. “Geographic changes in the low-rate clusters also occurred, with the disappearance of many low-rate clusters from Texas and New Mexico and the appearance of low-rate clusters in New England. Low-rate clusters in Florida
 and parts of the Pacific Northwest were maintained during the study period.”

The researchers cited a few limitations of the study, including that it relied on death certificates, which may have the potential for misclassification and inaccurate causes of death. They also could not determine if spatiotemporal trends in heart disease subtypes such as coronary heart disease and heart failure contributed differentially to the changing geographic patterns of heart disease mortality.

“Comprehensively documenting spatiotemporal patterns of heart disease mortality among communities in the United States, and exploring new hypotheses regarding the correlates of the changing geographic patterns, will enhance our ability to ensure that all communities experience optimal declines in heart disease mortality and that we narrow the existing geographic disparities in heart disease mortality,” the researchers wrote.