Fig. 5From: A comparison of methods to adjust for continuous covariates in the analysis of randomised trialsSimulation results for binary outcomes (power)Back to article page