A shared frailty model for case-cohort samples: parent and offspring relations in an adoption study

Area of Science:

• Biostatistics
• Epidemiology
• Genetics

Background:

• The Danish adoption register provides a unique dataset of 12,301 nonfamilial adoptions (1924-1947).
• Understanding genetic and environmental influences on mortality is crucial for public health.

Purpose of the Study:

• To develop and evaluate statistical methods for analyzing survival data in a case-cohort study using shared frailty models.
• To investigate the association between adoptee survival and their biological or adoptive parents' survival, assessing genetic and environmental factors.

Main Methods:

• A case-cohort sample was drawn from the Danish adoption register.
• Inverse probability weighting was applied to a conditional, shared frailty Poisson model.
• A robust variance estimator was utilized, and a simulation study assessed performance with different covariate types.

Main Results:

• The estimation of the frailty parameter was largely unbiased, with minor bias for rare causes of death.
• Variance estimation was unbiased for individual and within-family covariates, but overestimated for between-family covariates.
• A moderate genetic influence on mortality was observed, especially for all-cause and vascular deaths; no familial environment effect was detected.

Conclusions:

• The proposed statistical methodology is suitable for analyzing case-cohort data within shared frailty models.
• Genetic factors play a moderate role in mortality risk, while the shared familial environment appears to have minimal impact.