A Flexible Test for Comparing Two Hazard Rate Functions With Arbitrary Differences
Statistics in Medicine · 2026-11 · 45 (25-27)
DOI: 10.1002/sim.70831 ↗At a glance
Journal article in Statistics in Medicine, published online on 2026-11-19. Written by Xiaoxi Zhang and 2 co-authors. 28 references.
Abstract (original language)
In medical studies, comparison of two hazard rate functions is fundamentally important when analyzing survival data and evaluating treatment effects. Traditional comparison methods, such as the log‐rank test, work well for comparing two proportional hazard rates, but could fail when the two hazard rates cross each other or demonstrate a treatment time‐lag effect. In practice, however, crossing hazard rates and treatment time‐lag effects are fairly common. To study the related research problems with such difference patterns in the hazard rates, some statistical methods have been developed using either parametric survival modeling or nonparametric testing. But, these methods cannot detect both types of difference patterns effectively. In addition, it is usually difficult to justify their parametric model assumptions or derive the null distributions of their nonparametric test statistics. Thus, new methods are needed to compare two hazard rates with an arbitrary difference pattern. In this paper, a flexible omnibus testing procedure is developed, which integrates the traditional log‐rank test, a weighted log‐rank test, and a specific version of the Fleming‐Harrington test, which is good in comparing two hazard rates with a proportional or nearly proportional difference, crossing difference, and difference due to a treatment time‐lag effect, respectively. Numerical studies show that it works well in various cases considered.
Cite
APA
Zhang, X., Datta, S., & Qiu, P. (2026). A Flexible Test for Comparing Two Hazard Rate Functions With Arbitrary Differences. Statistics in Medicine, 45(25-27). https://doi.org/10.1002/sim.70831
Vancouver
Zhang X, Datta S, Qiu P. A Flexible Test for Comparing Two Hazard Rate Functions With Arbitrary Differences. Statistics in Medicine. 2026;45(25-27). doi:10.1002/sim.70831
BibTeX
@article{Zhang2026,
author = {Zhang, Xiaoxi and Datta, Somnath and Qiu, Peihua},
title = {A Flexible Test for Comparing Two Hazard Rate Functions With Arbitrary Differences},
journal = {Statistics in Medicine},
year = {2026},
volume = {45},
number = {25-27},
doi = {10.1002/sim.70831}
}
EUVIMED