The differences in power between the Cox model and the logistic regression models were relatively similar across the RAFs for each of the combinations of cumulative disease incidence and censoring strategy (Supplementary Figures S5–S7). In the secondary simulations, where the event time distributions were kept consistent across the censoring strategies, we observed that there was less divergence between the HRs and ORs as the rate of censoring increased (Supplementary Table S6). There was also a small increase in the difference in power between the two models when there was censoring (Supplementary Table S6). This occurred because, unlike when there was no censoring, the logistic regression model was now more precise than the Cox model (Supplementary Table S6).
