Dear all, I am interested in performing survival analyses with an area-level random effect (frailty term for each area of residence of individuals). I am using coxph, and include a frailty term in the model. Here is my model: mdcox<-coxph(Surv(time,censor)~ gender + age + frailty(area, dist='gauss'), data) The model provides the variance of the frailty term (between-area variance), along with a p-value. Maybe anyone knows if it is possible to obtain a standard error for the variance parameter, in order to construct a 95% confidence interval for this parameter? Second, I would like to obtain those shrunken estimates of the frailty term in each area. Do you know if there is a command allowing one to ask the frailty term for each group to be computed? I guess it should be possible to do it with predict(mdcox, type= ...), but I don't know exactly how to do. Thank you so much for your help. Best regards, Basile Chaix French National Institute of Health and Medical Research [[alternative HTML version deleted]]