(a) Predicted survival probabilities for four dynamic treatment regimens (DTRs) in Design II shown in Figure 1: Joint modeling (solid lines, 3.3) compared with the inverse probability of treatment weighting estimator (IPWE, black dashed lines, 3.1) and the weighted risk set estimator (WRSE, black dotted lines, 3.2). (b) Predicted survival rates for four dynamic treatment regimens (DTRs) in Design II when a covariate |$V=1$|. The left plot shows the survival curves from our proposed joint model and the model from Tang and Wahed (2015). Each pair of the overlapping curves of the DTRs is labeled with {|$A_1B_1C_1$|}, {|$A_1B_2C_1$|} and {|$A_2B_1C_1$|}, respectively. The non-overlapping curves of {|$A_2B_2C_1$|} are labeled with arrows. The right plot shows the survival curves from our proposed joint model and the subset analysis using WRSE and IPWE (i.e. subset data to |$V=1$| before model fit). The labeling is similar to the previous plot.
This PDF is available to Subscribers Only
View Article Abstract & Purchase OptionsFor full access to this pdf, sign in to an existing account, or purchase an annual subscription.