Traditionally, p-values have been used to determine statistical significance of research findings. However, statistical significance does not guarantee clinical relevance. The fragility index provides additional insight into robustness but still depends on p-values. In contrast, the relative risk index focuses on divergence from therapeutic equivalence, independent of statistical significance. A recent simulation found the relative risk index weakly correlates with p-values, suggesting distinct information. The relative risk index may better capture clinical relevance. Rather than solely relying on p-values and fragility indices, the relative risk index warrants consideration as a complementary metric when evaluating research to guide evidence-based practice.
Citation: Heston TF. Statistical significance versus clinical relevance: a head-to-head comparison of the fragility index and relative risk index. Cureus. 2023;15(10):e47741. DOI:10.7759/cureus.47741.
