4.9. Survival analysis (Cox‐type random effects regression model)

To investigate the extent of a survival difference between the family F3 case and control group, we use a Cox‐type random effects model:

where t_ij is the age at death for person j in family i. λ₀(t_ij) refers to the baseline hazard, which is left unspecified in a Cox‐type model. β is the vector of regression coefficients for the main effects of interest (Z). γ is a vector of regression coefficients for the effects of covariates and possible confounders (X). u_i > 0 refers to an unobserved random effect (frailty). In all Cox models, we adjust for sibship size, birth year, and sex.