The fragility index in randomized clinical trials supporting clinical practice guidelines for acute coronary syndrome: measuring robustness from a different perspective

Abstract

Aims: In randomized clinical trials (RCTs) rejecting the null hypothesis, the fragility index (FI) yields the minimum number of participants who would need to have had a different outcome for the results of the trial to become non-significant. We evaluated the robustness of RCTs supporting American College of Cardiology/American Heart Association (ACC/AHA) and European Society of Cardiology (ESC) clinical practice guidelines (CPGs) for ST-elevation myocardial infarction (STEMI) and non-ST-elevation acute coronary syndrome (NSTE-ACS) using the FI. Methods and results: There were 407 RCTs among the 2128 studies cited in the 2013 and 2014 ACC/AHA and 2017 and 2020 ESC CPGs for STEMI and NSTE-ACS, respectively. The FI could be calculated in 132 RCTs (32.4%) meeting the needed criteria for its estimation (two-arm RCT, 1:1 allocation, binary outcome, P < 0.05). The median FI was 12 (interquartile range: 4-29). Hence, a change in the outcome status of 12 patients would be needed to reverse the statistical significance of the primary endpoint in 50% of the RCTs. The FI was ≤1% than their sample size in 55.7% RCTs, whereas in 47% of RCTs, the FI was lower than the number of patients lost to follow-up. Some study design features were associated with a higher FI (international, multicentre, private funding; all P < 0.05), whilst baseline patient characteristics were not substantially different by FI (e.g. age, female sex, white study participants; all P > 0.05), except for geographic enrolment (P = 0.042). Conclusion: The FI might be useful to evaluate the robustness of those RCTs with statistically significant findings for the primary endpoint that have an impact on key guideline recommendations.