Two Ways of Modelling Multilevel Survival Data

Multilevel survival data with nested structures can be modelled by random-effect models such as multilevel mixed linear models and multilevel frailty models. We show how to analyze the well-known chronic granulomatous disease data, which comprise recurrent infection times of patients from different hospitals, using both types of multilevel survival models and discuss their relative merits. Inferences are based upon the hierarchical likelihood, which provides a simple unified framework and a numerically efficient fitting algorithm for various random-effect models.