SEARCH

SEARCH BY CITATION

Abstract

Data from a controlled clinical trial in liver cirrhosis are used to illustrate that multi-state models may be a useful tool in the analysis of data where survival is the ultimate outcome of interest but where intermediate, transient states are identified. We compare models for the marginal survival time distribution with models including transient states, both with respect to their clinical interpretation and with respect to the precision of survival probability estimates obtained from the various models. Copyright © 2000 John Wiley & Sons, Ltd.