Finite mixture regression models: assessing prognostic factors of survival in cirrhotic patients

Finite mixture regression models are an interesting and flexible model family, that have been successfully used in various kinds of tasks for modelling large heterogeneous populations. In this paper we applied this class of models with reference to liver cirrhosis. Cirrhosis is one of the principal causes of death in the world. It is a consequence of chronic liver disease, characterized by replacement of liver tissue by fibrous scar tissue as well as regenerative nodules, leading to progressive loss of liver function. We examined a sample of 271 cirrhotic patients followed in Medical Clinic of Messina University. The aim of this paper is to estimate a finite mixture regression model with concomitant variable in order to individualize the prognostic factors that influence survival times in patients affected by cirrhosis with different aetiology virus (viral or alcoholic hepatitis).