Answers are grounded in this paper and cite their sources.
Background Increasing randomized clinical trials evaluating novel therapies for inflammatory bowel disease necessitated the scrutiny of statistical robustness. This study aimed to quantify their fragility and identify factors associated with robustness. Methods This cross-sectional analysis included randomized clinical trials studying biologics, small-molecule inhibitors, fecal microbiota transplantation (FMT), and stem cell therapy (SCT), and then the calculated fragility index (FI) and continuous fragility index (CFI) for binary and continuous outcomes, respectively. Factors affecting robustness were analysed through correlation analysis and multiple linear regression. Results Among 129 trials from 53 studies, the median FI and CFI were 6 and 14.8, respectively. The FI varied significantly by the treatment type, trial phase, outcome type, and P values. The FI was positively correlated with the sample size (ρ = 0.734, P < 0.001), discontinuations (ρ = 0.479, P < 0.001), publication year (ρ = 0.253, P = 0.017), impact factor (ρ = 0.368, P < 0.001), and events percentage (ρ = 0.299, P = 0.005). The CFI was influenced by the outcome type and was strongly correlated with the sample size. After adjustment for other characteristics, biologics/small-molecule drugs displayed enhanced robustness relative to FMT/SCT (correlation coefficient B with the natural logarithm of FI: B = 0.283, P = 0.011). The primary or co-primary outcome exhibited greater robustness than did the other outcomes (FI: B = 0.288, P = 0.013; CFI: B = 0.459, P = 0.024). The sample size was positively correlated with both the FI (B = 0.001, P = 0.021) and the CFI (B = 0.001, P = 0.019), whereas discontinuation did not significantly affect the robustness. Conclusion Randomized clinical trials of novel inflammatory bowel disease therapies exhibit varying robustness and are influenced by multiple study characteristics. Trials of biologics or small molecules, those with positive primary outcomes, and larger studies demonstrated greater robustness, supporting that robustness should be considered in future research when interpreting the efficacy.
No takes yet. Share an insight, caveat, or question.
Zheng et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: