Quality of evidence underlying the American Heart Association/American College of Cardiology/Heart Rhythm Society guidelines on the management of atrial fibrillation

IMPORTANCE: The joint American College of Cardiology (ACC), American Heart Association (AHA), and Heart Rhythm Society (HRS) guidelines on the management of atrial fibrillation (AF) are used extensively to guide patient care. OBJECTIVE: To describe the evidence base and changes over time in the AHA/ACC/HRS guidelines on AF with respect to the distribution of recommendations across classes of recommendations and levels of evidence. DATA SOURCES: Data from the AHA/ACC/HRS guidelines on AF from 2001, 2006, 2011, and 2014 were abstracted. A total of 437 recommendations were included. DATA EXTRACTION AND SYNTHESIS: The number of recommendations and distribution of classes of recommendation (I, II, and III) and levels of evidence (A, B, and C) were determined for each guideline edition. Changes in recommendation class and level of evidence were analyzed using the 2001 and 2014 guidelines. RESULTS: From 2001 to 2014, the total number of AF recommendations increased from 95 to 113. Numerically, there was a nonsignificant increase in the use of level of evidence B (30.5% to 39.8%; P = .17) and a nonsignificant decrease in the use of level of evidence C (60.0% to 51.3%; P = .21), with limited changes in the use of level A evidence (8.4% to 8.8%; P = .92). In the 2014 guideline document, 10 of 113 (8.8%) recommendations were supported by level of evidence A, whereas 58 of 113 (51.3%) were supported by level of evidence C. Most recommendations were equally split among class I (49/113; 43.4%) and class IIa/IIb (49/113; 43.4%), with the minority (15/113; 13.3%) assigned as class III. Most class I recommendations were supported by level of evidence C (29/49; 59.2%), whereas only 6 of 49 (12.2%) were supported by level of evidence A. No rate control category recommendations were supported by level of evidence A. CONCLUSIONS AND RELEVANCE: Some aspects of the quality of evidence underlying AHA/ACC/HRS AF guidelines have improved over time. However, the use of level of evidence A remains low and has not increased since 2001. These findings highlight the need for focused and pragmatic randomized studies on the clinical management of AF.