Fig. 4From: The “RCT augmentation”: a novel simulation method to add patient heterogeneity into phase III trialsComparative efficacy of virtual RCTs comparing drug D1 and drug D2, in two parallel study arms with 250 patients each. Source populations are displayed on the x-axis: RCT or augmented RCTs as a result of relaxing any of the eight eligibility criteria. Comparative effectiveness is reported for the real-world population in the full SOHO cohort. Each box plot represents the distribution of comparative efficacy values obtained for the 1000 sampling replicates (bootstrapping)Back to article page