Table 1 The maximum (median) absolute bias of the estimated difference in mean responses of the newly added treatment and the control arm when there is a trend. Values with -4 order of magnitude are set to zero
From: Including non-concurrent control patients in the analysis of platform trials: is it worth it?
λ | WLSall | WLSs2 | Ma1 | Ma2 | Mb1 | Mb2 |
|---|---|---|---|---|---|---|
Linear trend | ||||||
2% | 0.007 (0.005) | 0 (0) | 0.001 (0) | 0 (0) | 0 (0) | 0 (0) |
4% | 0.015 (0.010) | 0 (0) | 0.001 (0) | 0 (0) | 0 (0) | 0 (0) |
6% | 0.022 (0.015) | 0 (0) | 0.001 (0) | 0 (0) | 0 (0) | 0 (0) |
8% | 0.030 (0.020) | 0 (0) | 0.001 (0) | 0 (0) | 0 (0) | 0 (0) |
Step trend | ||||||
2% | 0.015 (0.010) | 0 (0) | 0.007 (0.004) | 0 (0) | 0.010 (0.005) | 0 (0) |
4% | 0.030 (0.020) | 0 (0) | 0.015 (0.008) | 0 (0) | 0.019 (0.010) | 0 (0) |
6% | 0.045 (0.030) | 0 (0) | 0.023 (0.012) | 0 (0) | 0.029 (0.015) | 0 (0) |
8% | 0.060 (0.040) | 0 (0) | 0.031 (0.016) | 0 (0) | 0.038 (0.019) | 0 (0) |