Healthcare cost distribution generally presents a high level of skewness, with a relatively small number of subjects accounting for a large portion of health care expenditures. Information on factors that predict high expenditures is of interest in health care planning. The aim of this paper was to inspect the behaviour of extreme regression models (ER) in special cases of heterogeneity and strong asymmetry of the outcome variable and to discuss the application of ER models to the analysis of three datasets of diabetes, lung cancer and myocardial infarction patients. A simple simulation study based on the LogNormal distribution was also performed. The ER applied showed to be able to cope fairly well with skewed distribution but under the condition that all observations have strictly positive costs.
Extreme Regression Models for Characterizing High Cost Patients
GREGORI, Dario;BO, Simona;MERLETTI, Franco;
2009-01-01
Abstract
Healthcare cost distribution generally presents a high level of skewness, with a relatively small number of subjects accounting for a large portion of health care expenditures. Information on factors that predict high expenditures is of interest in health care planning. The aim of this paper was to inspect the behaviour of extreme regression models (ER) in special cases of heterogeneity and strong asymmetry of the outcome variable and to discuss the application of ER models to the analysis of three datasets of diabetes, lung cancer and myocardial infarction patients. A simple simulation study based on the LogNormal distribution was also performed. The ER applied showed to be able to cope fairly well with skewed distribution but under the condition that all observations have strictly positive costs.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.