Regression standardization and attributable fraction estimation with between-within frailty models for clustered...

Area of Science:

• Biostatistics
• Epidemiology
• Survival Analysis

Background:

• Clustered survival data presents challenges for standard analysis methods.
• Unmeasured cluster-constant confounding can bias results in ordinary frailty models.
• The between-within frailty model addresses these limitations, offering more consistent and efficient estimates.

Purpose of the Study:

• To develop novel estimation techniques for regression standardization within between-within frailty models.
• To extend the application of between-within frailty models for estimating the attributable fraction function.
• To demonstrate the utility of these methods using a real-world cohort study.

Main Methods:

• Derivation of new estimation techniques for regression standardization.
• Application of between-within frailty models to estimate the attributable fraction function.
• Analysis of a large cohort study on preterm birth and attention deficit hyperactivity disorder.

Main Results:

• Novel regression standardization methods for between-within frailty models were successfully derived.
• The attributable fraction function was effectively estimated using these models.
• The methods were validated through analysis of a significant cohort.

Conclusions:

• Between-within frailty models provide a robust framework for analyzing clustered survival data.
• The developed methods enhance the utility of these models for causal inference and risk assessment.
• The study provides practical tools (R code) for wider adoption of these advanced statistical techniques.