Evolution of the American College of Cardiology/American Heart Association Clinical Guidelines

Background The American College of Cardiology (ACC) and the American Heart Association (AHA) have been developing clinical guidelines to assist practicing clinicians. Objectives The goal of this study was to evaluate changes in ACC/AHA guideline recommendations between 2008 and 2014. Methods The previous and current ACC/AHA guideline documents that were updated between 2008 and June 2014 were compared to determine changes in Class of Recommendation (COR) and Level of Evidence (LOE). Each recommendation was classified as new, dropped, revised, or unchanged, and the changes in evidence were examined. Results During the study period, 11 guideline documents (9 disease based and 2 interventional procedure based) were updated. The total number of recommendations decreased from 2,067 to 1,869 (321 fewer recommendations in disease-based guidelines and 123 additional recommendations in interventional procedure-based guidelines). The recommendation class distribution of the updated guidelines was 50.1% Class I (previously 50.8%), 39.4% Class II (previously 35.4%), and 10.4% Class III (previously 13.8%) (p = 0.001). The LOE distribution among updated versions was 15.0% for LOE: A (previously 13.3%), 50.8% for LOE: B (previously 41.4%), and 34.2% for LOE C (previously 45.3%) (p < 0.001). Among all guidelines, 859 recommendations were new, 1,339 were dropped, 881 were unchanged in COR and LOE, and 129 were revised. Of the revised guidelines, 75 recommendations had an increase in LOE (the majority from LOE: C to LOE: B); 34 recommendations had a decrease in LOE; and 20 recommendations had class changes. LOE increases were justified by introduction of new randomized controlled trials, new studies, and new meta-analyses. Conclusions The ACC/AHA guideline recommendations are undergoing significant changes, becoming more evidence based and scientifically robust with a tendency to exclude recommendations with insufficient scientific evidence.