Incidence and economic burden of suspected adverse events and adverse event monitoring during AF therapy

Objective: Rhythm-and rate-control therapies are an essential part of atrial fibrillation (AF) management; however, the use of existing agents is often limited by the occurrence of adverse events. The aim of this study was to evaluate suspected adverse events and adverse event monitoring, and associated medical costs, in patients receiving AF rhythm-control and/or rate-control therapy. Research design and methods: This retrospective cohort study used claims data from the Integrated Healthcare Information Systems National Managed Care Benchmark Database from 2002-2006. Patients hospitalized for AF (primary diagnosis), and who had at least 365 days' enrollment before and after the initial (index) AF hospitalization, were included in the analysis. Suspected AF therapy-related adverse events and function tests for adverse event monitoring were identified according to pre-specified diagnosis codes/procedures, and examined over the 12 months following discharge from the index hospitalization. Events/function tests had to have occurred within 90 days of a claim for AF therapy to be considered a suspected adverse event/adverse event monitoring. Results: Of 4174 AF patients meeting the study criteria, 3323 received AF drugs; 428 received rhythm-control only (12.9%), 2130 rate-control only (64.1%), and 765 combined rhythm/rate-control therapy (23.0%). Overall, 50.1% of treated patients had a suspected adverse event and/or function test for adverse event monitoring (45.5% with rate-control, 53.5% with rhythm-control, and 61.2% with combined rhythm/rate-control). Suspected cardiovascular adverse events were the most common events (occurring in 36.1% of patients), followed by pulmonary (6.1%), and endocrine events (5.9%). Overall, suspected adverse events/function tests were associated with mean annual per-patient costs of $3089 ($1750 with rhythm-control, $2041 with rate control, and $6755 with combined rhythm/rate-control). Limitations: As a retrospective analysis, the study is subject to potential selection bias, while its reliance on diagnostic codes for identification of AF and suspected adverse events is a source of potential investigator error. A direct cause-effect relationship between suspected adverse events/function tests and AF therapy cannot be confirmed based on the claims data available. Conclusions: The incidence of suspected adverse events and adverse event monitoring during AF rhythm-control and/or rate-control therapy is high. Costs associated with adverse events and adverse event monitoring are likely to add considerably to the overall burden of AF management.