Pulmonary arterial hypertension trials put to the test: Using the fragility index to assess trials robustness

Abstract

BackgroundRandomized clinical trials (RCTs) are considered the gold standard for evidence-based medicine. The Fragility Index (FI) is a tool to assess the robustness of RCT results. FI was validated for dichotomous outcomes and recent work expanded its use to continuous outcomes. Studying the robustness of RCTs in Pulmonary Arterial Hypertension (PAH) treatments is crucial due to the severity and mortality risks associated with this rare condition. ObjectivesAnalyze FI and Fragility quotient (FQ) of significant primary outcomes in PAH RCTs and study FI correlation with sample size and journal impact factor. MethodsFI and FQ calculation followed by Spearman correlation to assess the correlation between FI and sample size, and FI and impact factor. ResultsThe median sample size of the 21 trials was 202 patients (IQR 106–267), with 6 trials reporting primary outcomes as dichotomous and 15 reporting continuous primary outcomes. The median FI was 10 (IQR 3–20), and the median FQ was 0.044 (0.026–0.097). A moderate correlation was found between FI and sample size, with r = 0.56; P = 0.008 and FI and journal impact factor (r=0.50; P=0.019). The FI for continuous outcomes was similar to that for dichotomous outcomes. ConclusionsThis study represents the first analysis of the FI and FQ of PAH treatment RCTs, and expands the use of FI to continuous outcomes in this context. The moderate correlation between FI and sample size suggests that increasing sample size alone is partially correlated to a higher FI. The similarity between FI for continuous and dichotomous outcomes supports the broader use of FI in PAH RCTs.