Testing equality of cause-specific hazard rates corresponding to m competing risks among K groups

Area of Science:

• Biostatistics
• Survival Analysis
• Epidemiology

Background:

• Comparing cause-specific hazard rates is crucial in analyzing complex failure data.
• Existing methods may require assumptions about the dependence of competing risks.
• Handling right-censored data in multi-group, multi-risk scenarios presents statistical challenges.

Purpose of the Study:

• To develop and validate a novel class of statistical tests for simultaneously comparing cause-specific hazard rates.
• To assess these tests in scenarios involving multiple competing risks and multiple groups.
• To provide a flexible testing framework applicable to censored survival data.

Main Methods:

• Development of tests based on differences in weighted averages of cause-specific hazard rates.
• Asymptotic analysis showing the test statistic follows a chi-squared distribution.
• Simulation studies using multivariate Gumbel distribution to evaluate weight functions.

Main Results:

• The proposed tests effectively compare cause-specific hazard rates across K groups for m competing risks.
• The test statistic demonstrates asymptotic chi-squared distribution.
• The study identifies an optimal weight function and proposes a practical alternative.

Conclusions:

• The developed statistical tests offer a robust method for analyzing competing risks in multi-group settings.
• The tests are applicable to right-censored failure time data without assuming risk dependence.
• The methodology is demonstrated through simulations and a real-world application in medical device data analysis.