Are orthopaedic randomized controlled trials as statistically fragile as portrayed? A call for improved interpretation of the statistical fragility index

The fragility index (FI) was first proposed by Walsh et al. in 2014 as a statistical tool to augment p values and 95% confidence intervals and to facilitate the identifica- tion of trials with less robust results [10]. Specifically, in the context of increased criticism around threshold p‐ values as an overly simple concept to determine whether a treatment effect is truly present, the FI determines the minimum number of patients in a trial who would need to experience an alternative or‘flipped’ outcome (such as a patient experiencing treatment success actually experiencing treatment failure) to reverse the significance of that trial. Calculation of the FI entails simultaneously adding and subtracting out- come events (i.e., treatment successes) and nonevents (i.e., treatment failures) in a sequential manner until the p‐value of that trial rises above the commonly applied 0.05 threshold level, thus providing a measure of the trials'‘fragility’ or ‘robustness’ to alternate patient outcomes.