Introduction: A growing epidemic of atrial fibrillation (AF) in the United States (US) has been predicted, though no data on the clinical or community burden of AF has been reported for the US since 2010. The objectives of this study were to describe community-based trends in AF incidence, prevalence, and post-diagnosis survival from 2004 to 2016 within a large health care system, and (2) extrapolate observed prevalence rates to the entire US population to estimate the national AF burden.
Methods: This retrospective cohort study incorporated the patient population and electronic medical record of the Geisinger Health System (Geisinger), an integrated health care delivery system serving central and northeast Pennsylvania. The study population included patients receiving primary care and other health care services through Geisinger for at least two years. Incidence rates per 1000 person-years were calculated by calendar year, and point prevalence rates estimated on July 1st of the respective years from 2004 to 2016. Standardized incidence and prevalence rates were age- and sex-adjusted to the 2010 US Census. Rate ratios (RR) were estimated from Poisson regression and are interpreted as the annual relative change over time.
Results: A total of 464,363 patients met study inclusion criteria with a total person-time experience of 5,257,938 person-years (median 12.7 years, IQR: 7.7, 15.4). Age- and sex-adjusted AF incidence increased steadily over the study period with incidence rates of 4.7, 5.0, 5.8, and 6.2 in 2004, 2008, 2012, and 2016, respectively. The overall test for linear trend over time was statistically significant with RR=1.03 (95% CI: 1.02, 1.03) per year. Age- and sex-adjusted prevalence rates increased consistently over time from 2.7%, 3.0%, 3.4%, to 4.1% in 2004, 2008, 2012, and 2016, respectively (RR=1.04, 95% CI: 1.03, 1.04). In 2004, an estimated 6.1 million Americans had diagnosed AF, increasing to 6.7, 7.8, and 9.3 million in 2008, 2012, and 2016, respectively. Post-diagnosis survival has not improved in recent years.
Conclusion: AF incidence and prevalence have increased steadily since 2004, while post-diagnosis survival has not improved. This community-based study strongly supports a growing burden of recognized AF in the clinic and community.