A proposal for the analysis of kidney graft survival.

A survival analysis using a model provides numerous advantages over other approaches. The effectiveness of concomitant variables in describing the data can be evaluated whether the variables are continuous or dichotomized. We can evaluate the effectiveness of such variables in describing either short- or long-term components of the risk. The expected life span of the graft may be computed with the model and complements the insight obtained from the more commonly displayed survival curves.