Inference for competing risks based on area between curves statistics

TL;DR

This paper introduces a new statistical test based on the area between cumulative incidence functions for competing risks, using a wild bootstrap approach to handle crossing functions and improve inference accuracy.

Contribution

It proposes a novel area-based test statistic with a bootstrap method for valid inference in competing risks models with crossing cumulative incidence functions.

Findings

The proposed method performs well in simulations compared to existing tests.

It effectively handles crossing cumulative incidence functions.

The approach is asymptotically valid and practical for real data analysis.

Abstract

In competing risks models, cumulative incidence functions are commonly compared to infer differences between groups. Many existing inference methods, however, struggle when these functions cross during the time frame of interest. To address this problem, we investigate a test statistic based on the area between cumulative incidence functions. As the corresponding limiting distribution depends on quantities that are typically unknown, we propose a wild bootstrap approach to obtain a feasible and asymptotically valid two-sample test. The finite sample performance of the proposed method, in comparison with existing methods, is examined in an extensive simulation study.