Modelling time-dependent hazard ratios in relative survival: Application to colon cancer

The Cox model is widely used in the evaluation of prognostic factors in cli nical research. In population-based studies, however, which assess long-ter m survival of unselected populations, relative survival models are often co nsidered more appropriate. In both approaches, the validity of proportional hazard hypothesis should be evaluated. To explore the validity of the prop ortional hazard assumption in a population-based study of colon cancer, to propose non-proportional hazard relative survival models and to evaluate th eir utility. The use of a piecewise proportional hazard relative survival m odel in colon cancer has shown that the effects of most clinical prognostic factors such as age, period of diagnosis and stage are non-proportional. T he non-proportional hazard relative survival models developed in this artic le have been found to be efficient tools for better understanding the time- dependent aspect of prognostic factors. (C) 2001 Elsevier Science Inc. All rights reserved.