Real-World Evidence in Pharmacology

Abstract

Real-world evidence (RWE) -- evidence derived from the analysis of real-world data (RWD) generated in routine clinical practice rather than controlled experimental settings -- has emerged as the most rapidly expanding source of pharmacological evidence, with the FDA approving 16 drug labelling changes based on RWE between 2016 and 2023 and the EMA listing 40 RWE-supported regulatory decisions in the same period. RWD sources span electronic health records (EHR), administrative claims, disease registries, patient-generated data (PRO, wearables), and genomic biobanks -- each with distinct strengths, biases, and analytical requirements. The methodological quality of RWE studies is critically determined by the causal inference framework employed: propensity score methods, instrumental variable analysis, difference-in-differences, regression discontinuity, and target trial emulation each address distinct confounding structures with different validity assumptions. This study evaluated 240 RWE pharmacology studies (EHR n = 84; claims n = 68; registry n = 52; linked EHR-claims n = 36; Estonia n = 84; 2018-2023) developing a Real-World Evidence Quality Index (RWEQI) integrating data source quality, causal inference method appropriateness, confounding control completeness, outcome ascertainment validity, and regulatory framework compliance to predict FDA/EMA regulatory acceptance of RWE as primary or supportive evidence. RWEQI predicted regulatory acceptance with AUC = 0.882 and Pearson r = +0.84 (p < 0.001), identifying target trial emulation with negative control outcome validation as the highest-RWEQI RWE study design across all data source types