Restricted survival benefit with right-censored data.

Area of Science:

• Biostatistics
• Clinical Trial Analysis
• Survival Analysis

Background:

• The hazard ratio (HR) is a common metric for treatment effects in survival analysis.
• HR interpretation can be challenging for patients and clinicians, particularly when it varies over time.
• Existing alternative measures like restricted mean survival time (RMST) differences have limitations.

Purpose of the Study:

• To introduce the Restricted Survival Benefit (RSB) as a novel, comprehensive measure of treatment effects.
• To provide a flexible framework for assessing treatment efficacy using multiple survival data aspects.
• To develop robust statistical methods for estimating and inferring RSB with censored survival data.

Main Methods:

• Proposed the Restricted Survival Benefit (RSB) as an integrated treatment effect measure.
• Developed estimation and inference procedures using inverse-probability-censoring-weighted (IPCW) methods.
• Employed jackknife empirical likelihood for robust statistical inference.
• Validated the method through extensive simulation studies.

Main Results:

• The proposed RSB method provides a comprehensive assessment of treatment effects.
• Simulation studies demonstrated the numerical performance and reliability of the RSB estimation procedures.
• The RSB method was successfully applied to analyze data from the SWOG S0777 Phase III clinical trial.

Conclusions:

• The Restricted Survival Benefit (RSB) offers a valuable and interpretable alternative to the hazard ratio.
• The developed statistical framework allows for a more complete evaluation of treatment benefits in survival studies.
• RSB enhances clinical trial data interpretation for both researchers and patients.