Fig. 5From: Using Bayesian statistics to estimate the likelihood a new trial will demonstrate the efficacy of a new treatmentForest plot of the risk difference in failure rates between the control and experimental groups in a non-cumulative frequentist meta-analysis (I2Â =Â 45%, QÂ =Â 0.04) (a) and in a cumulative frequentist meta-analysis (b). Negative estimates are in favour of the experimental groupBack to article page