Title: Combined tests based on restricted mean time lost for competing risks data

Abstract: Competing risks data are common in medical studies, and the sub-distribution hazard (SDH) ratio is considered an appropriate measure. However, because the limitations of hazard itself are not easy to interpret clinically and because the SDH ratio is valid only under the proportional SDH assumption, this article introduced an alternative index under competing risks, named restricted mean time lost (RMTL). Several test procedures were also constructed based on RMTL. First, we introduced the definition and estimation of RMTL based on Aalen-Johansen cumulative incidence functions. Then, we considered several combined tests based on the SDH and the RMTL difference (RMTLd). The statistical properties of the methods are evaluated using simulations and are applied to two examples. The type I errors of combined tests are close to the nominal level. All combined tests show acceptable power in all situations. In conclusion, RMTL can meaningfully summarize treatment effects for clinical decision making, and three combined tests have robust power under various conditions, which can be considered for statistical inference in real data analysis.