Abstract 206: Projected Estimates of Prevalence and Annual Growth Rate of Atrial Fibrillation in the United States from a Dynamic Age-Period Progression Model

Abstract

Background: Estimates from epidemiologic studies projecting the prevalence of atrial fibrillation (AF) in the US have varied widely in the literature, ranging from 2.7 million cases to 6.1 million cases for 2010. While the prevalence of AF is increasing due to an aging population, the extended survival of the elderly population and increases in other AF risk factors in the US population have not been factored into the current published AF prevalence projections. The large range of prevalence estimates for AF may be due to several factors, including study design, time period when study participants were ascertained, increases in AF over time, changes in AF diagnostic tools, and aging demographics. Methods: In an effort to quantify the population size for AF, a yearly annual incidence rate was estimated over a 7-year period (2002-2009) using data from a large US health insurance claims database (i3, n=45,871). Prevalence estimates were modeled for each year based on US population age and gender distribution and mortality rates. The US AF prevalence was computed (upweighted to age-standardized US census data) in a dynamic age-period cohort progression (Markov chain) model, which back-tracked prevalence to 1990 from incidence rates and projected trends forward to 2020. Results: The number of prevalent chronic plus current year incident cases of AF for 2010 was estimated at 5.2 million cases and was projected to increase to 8.4 million by 2020 (Figure). The estimated AF incidence was projected to double from approximately 200 cases/100,000 in 2010 to over 450 cases/100,000 in 2020, equating to an annual growth rate of 4.7%. Conclusion: These estimates, derived from the largest longitudinal study sample reported thus far, with a dynamic age-period cohort progression model, suggest that the diagnosed AF population in the US in 2010 was approximately 5.2 million cases, which falls within the wide range previously reported in the literature and may represent a more precise estimate.