Generalized mean residual life models for case-cohort and nested case-control studies

Area of Science:

• Survival analysis
• Biostatistics
• Epidemiology

Background:

• Mean Residual Life (MRL) is crucial for survival analysis, often used instead of hazard functions.
• Traditional MRL models primarily use full-cohort studies, which can be costly.
• Case-cohort and nested case-control designs offer cost savings for large cohorts, especially for rare diseases and biomarker studies.

Purpose of the Study:

• To develop and evaluate generalized Mean Residual Life (MRL) models for case-cohort and nested case-control designs.
• To provide methods for estimating regression parameters and baseline MRL functions under these designs.
• To assess the performance of the proposed estimators through simulations and a real-world health study.

Main Methods:

• Utilized inverse probability weighting to construct estimating equations for MRL models.
• Developed methods for inference on regression parameters and the baseline MRL function.
• Employed extensive numerical simulations to evaluate finite-sample performance.

Main Results:

• Established asymptotic properties for the proposed MRL model estimators.
• Demonstrated the effectiveness of the weighted estimating equations.
• Validated the models and a diagnostic method using data from the New York University Women's Health Study.

Conclusions:

• Generalized MRL models are effective for case-cohort and nested case-control designs.
• The proposed methods provide valid and efficient inference for time-to-event data in resource-constrained settings.
• The study offers practical tools for analyzing survival data in epidemiological research.