Evaluating the strength and quality of evidence in American Heart Association/American Stroke Association’s guidelines for primary stroke prevention

BACKGROUND: Stroke prevention depends on evidence-based guidelines. This study assessed the strength of recommendations and quality of evidence in AHA/ASA primary stroke prevention guidelines across four editions (2006, 2011, 2014, 2024). METHODS: Guidelines were reviewed for Class of Recommendation (COR) and Level of Evidence (LOE). Two reviewers independently extracted data using a set protocol; agreement was measured with Cohen's κ. Cochran-Armitage trend tests assessed changes in COR/LOE over time. RESULTS: Of 35 guideline categories, 8 were descriptive with no formal COR/LOE. The remaining 27 categories yielded 83 recommendations: 45% Class I, 19% Class IIa, 24% Class IIb, and 12% Class III. For evidence quality: 13% LOE A, 24% LOE B-R, 30% LOE B-NR, 5% LOE C-EO, and 28% LOE C-LD. LOE A recommendations fell significantly, from 41% (2006) to 13% (2024) (p < 0.001), while LOE B increased from 33% to 54% (p < 0.001). High-quality (LOE A) evidence was concentrated in pharmacological recommendations for major risk factors, while lifestyle and emerging risk domains relied mostly on LOE B or C evidence. CONCLUSION: Fewer than half of AHA/ASA primary stroke prevention recommendations carry the top COR, and only 13% rest on high-quality evidence. The drop in LOE A reflects both the guidelines' expansion into newer clinical areas and ongoing challenges in running large randomized controlled trials for those domains.