Comparing the importance of prognostic factors in Cox and logistic regression using SAS Georg Heinze *, Michael Schemper Department of Medical Computer Sciences, University of Vienna, A-1090 Vienna, Spitalgasse 23, Austria Received 20 December 2001; accepted 25 April 2002 Abstract Two SAS macro programs are presented that evaluate the relative importance of prognostic factors in the proportional hazards regression model and in the logistic regression model. The importance of a prognostic factor is quantified by the proportion of variation in the outcome attributable to this factor. For proportional hazards regression, the program %RELIMPCR uses the recently proposed measure V to calculate the proportion of explained variation (PEV). For the logistic model, the R 2 measure based on squared raw residuals is used by the program %RELIMPLR. Both programs are able to compute marginal and partial PEV, to compare PEVs of factors, of groups of factors, and even to compare PEVs of different models. The programs use a bootstrap resampling scheme to test differences of the PEVs of different factors. Confidence limits for P -values are provided. The programs further allow to base the computation of PEV on models with shrinked or bias-corrected parameter estimates. The SAS macros are freely available at www.akh-wien.ac.at/imc/biometrie/relimp. # 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Bias correction; Explained variation; Monotone likelihood; Proportional hazards; Relative importance; Separation; Shrinkage 1. Introduction Statistical analysis of a prognostic factor study should involve the estimation of marginal and partial effects. Results of such an analysis are suitably presented in a table containing estimated effects, confidence limits and P -values. These standard requirements of a proper statistical analysis have been suggested by Armitage and Gehan [1] and Gehan and Walker [2]. Schemper [3] pointed out that apart from reporting marginal and partial effects estimates, it can be necessary to comment on the relative importance of factors in a model by computing the proportion of total variation in the outcome variable that can be explained by each factor. The concept of relative importance allows a descriptive ranking of prognostic factors accord- ing to their statistically determined importance. Using bootstrap techniques [3] it is also possible to compare the importance of different prognostic factors statistically. * Corresponding author. Tel.: /43-1-40400-6684; fax: /43- 1-40400-6687 E-mail address: georg.heinze@akh-wien.ac.at (G. Heinze). Computer Methods and Programs in Biomedicine 71 (2003) 155 /163 www.elsevier.com/locate/cmpb 0169-2607/02/$ - see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved. PII:S0169-2607(02)00077-9