Skip to main content
Fig. 1 | BMC Medical Research Methodology

Fig. 1

From: Including non-concurrent control patients in the analysis of platform trials: is it worth it?

Fig. 1

Left: power curves when the individual Z-test (red lines) and WLSs2 (blue lines) are used respectively for testing H02, both using stage two control responses. Right: reduction in \(var(\hat {\delta }_{2})\) when stage one control responses is used relative to not using stage one control responses. Line types correspond to different values of n22 when the new arm is added to the on-going trial. The timing of adding the new arm is reflected by n01/n0 on the x-axis

Back to article page