Health Services Utilization and Medical Costs Among Medicare AF Patients

Interview by Jodie Elrod
Interview by Jodie Elrod
In this interview we speak with Erin Sullivan, MPH, PhD, Vice President of Health Economics and Outcomes Services at Avalere Health, about a new report entitled “Health Services Utilization and Medical Costs Among Medicare Atrial Fibrillation Patients,” which shows an increased use of healthcare services among Medicare patients with atrial fibrillation (AF). The report was written by Avalere Health and funded as an outcome of AF Stat™: A Call to Action for Atrial Fibrillation, which is collaboration of healthcare leaders and organizations working to improve the health and well-being of people affected by atrial fibrillation. It is an initiative sponsored by sanofi-aventis U.S. LLC. First, tell us about the work that you do with AF Stat, and how this series of reports came about. Avalere Health is a healthcare research, consulting and health policy firm based in Washington D.C., and we have been working with the AF Stat team for the last couple of years. Their report on the costs associated with atrial fibrillation (AF) is the second report that we’ve done for them. Their first report, “Medicare and Atrial Fibrillation: Consequences in Cost and Care,” was based on an overall review of general quality initiatives that are ongoing in AF; we were approached by AF Stat to work on that paper about 2 years ago, and this was a follow-on report. It seemed like a natural follow-up to try to better understand the economic burden of AF in the Medicare patient population and try to obtain some real cost data and put some numbers into the equation so people can better understand the costs. We looked at the annual costs associated with treating AF patients. What can you tell us about the study group population? For the analysis, we looked at patients who were included in Medicare’s 5% standard analytic files (or “SAFs”), which is actually a nationally representative subset of the Medicare patient population. These are publicly available database files. We looked specifically at patients who had their first acute hospitalization for AF between 2005 and 2007, as indicated by the diagnosis codes on the Medicare inpatient hospital claims. We required patients in our population to be enrolled in Medicare for one year prior to, and one year after this first hospitalization (called the “index hospitalization”). We identified over 14,000 patients with a primary diagnosis of AF during their index hospitalizations who met our enrollment criteria. Our AF patient population was similar to those reported in other published AF studies. The average age was 76; 60% were female; and over 90% of the patients were Caucasian. Patients included in this analysis had several comorbid conditions in addition to AF, including hypertension, cardiovascular conditions, and diabetes. Tell us about the average medical cost for AF patients during the follow-up period. How did the numbers compare to costs for other cardiovascular conditions? We looked at inpatient hospital costs, and specifically at hospital readmissions, outpatient hospital costs, and physician costs. It is important to note we did not have drug costs available in our database. It is also important to note that due to patient privacy regulations, we did not have the actual dates of service for any health care event. Medicare only discloses the quarter of service in the 5 percent limited data set SAFs. As such, the follow-up period includes the quarter of the index hospitalization and the four quarters after the index hospitalization quarter. When we looked at that overall amount of costs during these 5 quarters, we found that the total average medical cost for treating AF was nearly $24,000 during that time. The average cost was $23,899, and you’ll see in the report that about 62% of costs over that period were associated with inpatient services. Of those, 63% of the inpatient service costs were associated with readmission costs that occurred after that index hospitalization. The costs observed in our analysis are similar to those reported for other major cardiovascular conditions, including heart failure and acute coronary syndrome. These findings highlight the specific economic burden of AF among Medicare beneficiaries and emphasize the need to identify ways to improve the quality of care and reduce the high utilization and medical costs among AF patients, especially as the prevalence of AF increases and the Medicare patient population continues to grow. Where do most of the costs associated with AF and its complications occur according to the report? According to our report, the majority of costs were incurred in the inpatient setting. A little over half, about 52% of patients, were readmitted to the hospital at least once during the follow-up period. Interestingly, about 12% of patients had 3 or more readmissions during the follow-up period. To me, that was surprising. There were also a high number of physician encounters observed during the follow-up period. On average, patients had 67 physician encounters, and it’s important to keep in mind that that includes encounters in all different types of settings, so for example, if a patient goes to the doctor and has their blood drawn, that would count as one encounter with a physician. When we looked specifically at how many visits patients actually had to a physician’s office, they had 30 visits. So this is still a relatively high number looking at a 5-quarter follow-up period. The physician services ended up as the second highest cost, with about $6,300 average spent over the follow-up period, followed by the hospital outpatient services. One last thing worth mentioning is that 61% of AF patients in our study visited the emergency department (ED) at least once, and went to the ED 3 times on average during the follow-up period. This pattern of extensive ED use may warrant additional research to try to identify opportunities to reduce the need for emergency care among this patient population. Why is it important to examine the burden of AF on the Medicare population? How can we use this information to improve AF management? Medicare is the primary payer of AF services across all settings of care and as such, absorbs the majority of the clinical and economic burden of caring for AF patients. The economic burden of AF on Medicare is not currently well understood. Our analysis helps to characterize AF costs among Medicare patients and is based on more recent data (2004–2008) than what is available in the published literature. As far as using this information to improve AF management, one of the key conclusions of our study is that AF patients have high rates of health services utilization, particularly hospital readmissions, and high medical costs. One of the interesting findings in this report was that nearly half of the rehospitalizations occurred within either the quarter of or the quarter immediately following the index hospitalization, so within the first 3–6 months of that initial ‘index’ hospitalization. Seeing that the timing of the rehospitalization is occurring in a relatively short amount of time after that initial hospitalization, I definitely think that this research should cause folks to start asking questions about why the costs are so high and what types of services might we be able to prevent. Are we treating and coordinating care for AF patients as well as we could be, and are there things that we could do to maybe reduce the likelihood of rehospitalization? We need to understand more about why patients are being rehospitalized. Our analysis allows us to examine the diagnoses reported at these rehospitalizations but does not enable us to fully understand the factors driving rehospitalization and the high rates of utilization in the physician office and in the hospital outpatient setting. Identifying ways to reduce readmissions and improve care management may present real opportunities for providers and payers to reduce overall medical costs among Medicare AF patients.

Top findings from the report:

  • Patients with atrial fibrillation had multiple comorbidities, including hypertension, other cardiovascular conditions, and diabetes.
  • AF patients in this study frequently used hospital outpatient (90 percent had at least 1 outpatient visit) and physician health services (98 percent visited their physician at least once).
  • A high percentage (61 percent) of AF patients visited the ER at least once during the follow-up period.
  • Fifty-two percent of AF patients were readmitted to an inpatient hospital at least once during follow-up. In addition, 24 percent of patients were readmitted multiple times.
  • Almost half of hospital readmissions in these AF patients were due to cardiovascular-related conditions.
  • AF was more often reported as a secondary diagnosis to other conditions, which could influence healthcare utilization patterns and costs of care.
For more information, please visit:
In addition, the full report is available at


  1. Sullivan E, et al. Health Services Utilization and Medical Costs Among Medicare Atrial Fibrillation Patients. Avalere Health. 2010:4:1-58.