Incidence and Prevalence of AFib and Associated Mortality Among Medicare Beneficiaries

Previous epidemiological studies have established demographic and clinical risk factors for the development of atrial fibrillation (AF) as well as a clear association linking the presence of AF to increased mortality. Given that AF incidence is age-related and that risk factors for developing AF are increasing in the population, previous studies have also predicted dramatic growth in the US prevalence of AF over the next few decades. In order to further explore recent trends in the incidence and prevalence of AF, as well as the mortality and associated medical conditions of AF patients, Piccini and colleagues conducted a comprehensive analysis using roughly 15 years of Medicare claims data.(1)

For this study the authors used standard analytic files containing both inpatient and outpatient claims from a representative 5% sample of patients enrolled in Medicare fee-for-service plans. The analysis was limited to a minimum age of 65. Incident AF was defined by the presence of a diagnosis code for AF in one inpatient or 2 outpatient claims in a patient not having visits coded with an AF diagnosis in the previous 2 years, while prevalent AF was calculated as all beneficiaries with an AF diagnosis recorded within a calendar year and alive at the end of that year, divided by the total number of Medicare beneficiaries alive within the 5% sample at any time during that year.

Figure 1: Incidence and Prevalence of Atrial Fibrillation and Associated Mortality among Medicare Beneficiaries, 1993–2007The authors found little change in the AF incidence rate, which overall was 27.3 per 1000 beneficiary life-years in 1993 and 28.3 per 1000 in 2007. Previously described epidemiological findings were confirmed: AF incidence was higher in men than women and in whites than non-whites, and increased uniformly with age. Over time, incident AF cases were more frequently associated with hypertension (84% in 2007), diabetes (34%), peripheral vascular disease (30%), renal disease (16%) and COPD (36%). Incident AF was also frequently associated with coronary artery disease (51%) and heart failure (36%), but these rates did not change over time.

Mortality rates following incident AF diagnosis in the Medicare population were notably high: 11% at 30 days and 25% at one year in 2007. These rates were just slightly lower than in 1993. The 3-year mortality rate in 2005, the last year available, was 42%. The 1-year mortality rates in beneficiaries with incident AF were ~3.5 times higher than expected after adjustment for age and sex. Three-year mortality rates were roughly twice as high as age- and sex-adjusted rates.

In contrast to the relatively constant incidence and mortality rates, the prevalence of AF was noted to rise markedly over time, from 41 cases per 1000 beneficiaries in 1993 to 85 cases per 1000 beneficiaries in 2007. Prevalence rates were highest in men (103 cases per 1000), whites (91 cases per 1000) and patients over 80 (>130 cases per 1000). The observation of a dramatic (an average of 5% per year, or >100% over 15 years) rise in AF prevalence despite little change in incidence is somewhat surprising. This may be partly explained by improved survival of AF patients, however, 3-year mortality improved only slightly from 45% in 1993 to 42% in 2005. The greatly increased prevalence therefore also appears to be related to changes in the frequency of coding AF as an outpatient diagnosis – when AF prevalence was defined solely using inpatient claims, the 2007 AF prevalence was only 35 per 1000, vs. 25 per 1000 in 1993. Whether this reflects greater attention to AF in the outpatient setting or simply improved documentation in medical claims is not clear. Given the overall size of the Medicare population, the authors estimate that 2.2 million Medicare beneficiaries had AF in 2007.

Figure:2 Incidence and Prevalence of Atrial Fibrillation and Associated Mortality among Medicare Beneficiaries, 1993–2007In addition to confirming some of the previously reported epidemiological and clinical features of AF in the US, this study highlights the importance of this condition to the Medicare program and to the treatment of older patients in general. Based on these data it appears that in 2007, over 10% of men and 7% of women aged 65 and older in the Medicare program had prevalent AF. Furthermore, over a 15 year time span, declines in 30-day, 1-year and 3-year mortality among AF patients were modest. Despite these sobering data, there have been relatively few national efforts to improve the efficiency or quality of care or the outcomes of AF patients, particularly when AF is compared with other common cardiovascular conditions such as myocardial infarction or heart failure. Abundant literature, for example, continues to show underutilization of anticoagulants in patients with AF and identifiable stroke risk factors.

Analysis of the past 15 years of Medicare data on AF patients also clearly suggests, as others have previously predicted, that aging of the population and continued improvements in the survival of patients with cardiovascular disease will lead to continued increases in the prevalence of AF in the coming years. Improved strategies for the prevention and management of AF will be needed to stem the rising tide of AF cases and their associated adverse outcomes.


  1. Piccini JP, Hammill BG, Sinner MF, Jensen PN, Hernandez AF, Heckbert SR, Benjamin EJ, Curtis LH. Incidence and prevalence of atrial fibrillation and associated mortality among Medicare beneficiaries, 1993-2007. Circ Cardiovasc Qual Outcomes 2012; 5:85-93

Keywords: Atrial Fibrillation, Coronary Artery Disease, Diabetes Mellitus, Risk Factors

< Back to Listings