Testing the equality of two survival functions with right truncated data

Area of Science:

• Biostatistics
• Survival Analysis
• Statistical Methods

Background:

• Comparing survival functions with right-truncated data presents analytical challenges.
• Existing methods include weighted logrank tests and semi-parametric Mann-Whitney tests.
• Both existing methods have limitations regarding weight function dependency and sensitivity to truncation time distribution assumptions.

Purpose of the Study:

• To propose a novel non-parametric test statistic for comparing survival functions using right-truncated data.
• To evaluate the performance of the proposed test against existing methods through simulation.
• To demonstrate the practical application of these statistical methods on real-world data.

Main Methods:

• Developed a non-parametric test statistic based on the integrated weighted difference of estimated survival functions.
• Conducted a simulation study to compare the proposed test with methods by Lagakos et al. and Bilker and Wang.
• Applied the methods to a real dataset to illustrate their implementation.

Main Results:

• The proposed non-parametric test offers an alternative approach for survival function comparison.
• Simulation results provide comparative performance insights into the different statistical tests.
• The study demonstrates the practical utility of these survival analysis techniques.

Conclusions:

• The novel non-parametric test provides a valuable tool for analyzing right-truncated survival data.
• The research contributes to the field of survival analysis by offering a robust statistical method.
• Comparative analyses highlight the strengths and weaknesses of different approaches to survival data.