Skip to main content
Fig. 2 | Trials

Fig. 2

From: Estimands for clinical endpoints in tuberculosis treatment randomized controlled trials: a retrospective application in a completed trial

Fig. 2

Point-range plot of risk difference estimates according to each estimand/estimation method. Each row corresponds with unique analyses within a given estimand. The vertical dotted line represents the non-inferiority margin of 6%. The results are shown as a point estimate of a risk difference and a corresponding 97.5% confidence interval. Results in orange are estimated with the Kaplan-Meier estimator (KM), results in green are estimated with the Cochrane Mantel Haenszel method (CMH), and results in blue are estimated according to the Principal Stratum method (PS). Point estimates represented by a square have implemented Inverse Probability of Censoring Weighting (IPCW) to handle certain intercurrent events. Point estimates represented by an asterisk have implemented Multiple Imputation (MI) to handle certain intercurrent events. Points estimates represented by a triangle have implemented the Principal Stratum (PS) method to handle certain intercurrent events. The remaining point estimates are represented by a circle meaning that no special statistical methods were used to handle intercurrent events

Back to article page