From: The “RCT augmentation”: a novel simulation method to add patient heterogeneity into phase III trials
Re-included subpopulations | Natural augmentation (when opening the trial to the specific “real-world population” subgroup) | Prediction bias with natural augmentation | Mean squared error (MSE) of prediction with natural augmentation | |
---|---|---|---|---|
Relaxed eligibility criteria | Illness duration between 1 and 3 years | 56 | 0.004 | 0.733 |
1 past suicide attempt | 73 | 0.013 | 0.734 | |
Private practice | 69 | 0.003 | 0.737 | |
Alcohol abuse | 42 | −0.008 | 0.749 | |
Drug abuse | 22 | −0.004 | 0.756 | |
Illness duration between 1 and 3 years + 1 past suicide attempt | 121 | 0.033 | 0.716 | |
Illness duration between 1 and 3 years + private practice | 121 | 0.009 | 0.719 | |
Illness duration between 1 and 3 years + alcohol abuse | 90 | 0.011 | 0.728 | |
RCT population | not applicable | −0.016 | 0.777 | |
SOHO “real-world population” | not applicable | 0.000 | 0.000 |