Analyzing differences between restricted mean survival time curves using pseudo-values

Hazard ratios are ubiquitously used in time to event analysis to quantify treatment effects. Although hazard ratios are invaluable for hypothesis testing, other measures of association, both relative and absolute, may be used to fully elucidate study results. Restricted mean survival time differences between groups have been advocated as useful measures of association. Recent work focused on model-free estimates of the difference in restricted mean survival for all follow-up times instead of focusing on a single time horizon. In this work a model-based alternative is proposed with estimation using pseudo-values. A simple approach is proposed easily implementable with available software. It is also possible to compute a confidence region for the curve. As a by-product, the parameter 'time until treatment equipoise' (TUTE) is also studied. Examples with crossing survival curves will be used to illustrate the different methods together with some simulations.
Comment: 19 pages, 4 figures, 2 tables