A SAS macro for estimation of direct adjusted survival curves based on a stratified Cox regression model

Often in biomedical research the aim of a study is to compare the outcomes of several treatment arms while adjusting for multiple clinical prognostic factors. In this paper we focus on computation of the direct adjusted survival curves for different treatment groups based on an unstratified or a stratified Cox model. The estimators are constructed by taking the average of the individual predicted survival curves. The method of direct adjustment controls for possible confounders due to an imbalance of patient characteristics between treatment groups. This adjustment is especially useful for non-randomized studies. We have written a SAS macro to estimate and compare the direct adjusted survival curves. We illustrate the SAS macro through the examples analyzing stem cell transplant data and Ewing's sarcoma data.

[1]  R W Makuch,et al.  Adjusted survival curve estimation using covariates. , 1982, Journal of chronic diseases.

[2]  J. Klein,et al.  Survival Analysis: Techniques for Censored and Truncated Data , 1997 .

[3]  E. Kaplan,et al.  Nonparametric Estimation from Incomplete Observations , 1958 .

[4]  Ruta Bajorunaite,et al.  Comparison of autologous and allogeneic hematopoietic stem cell transplantation for follicular lymphoma. , 2003, Blood.

[5]  Mei-Jie Zhang,et al.  Confidence Bands for the Difference of Two Survival Curves Under Proportional Hazards Model , 2001, Lifetime data analysis.

[6]  R. Gill,et al.  Cox's regression model for counting processes: a large sample study : (preprint) , 1982 .

[7]  R. Brant,et al.  Comparison of 2 methods for calculating adjusted survival curves from proportional hazards models. , 2001, JAMA.

[8]  J. Coresh,et al.  Adjusting survival curves for confounders: a review and a new method. , 1996, American journal of epidemiology.

[9]  D G Altman,et al.  Use of a prognostic index in evaluation of liver transplantation for primary biliary cirrhosis. , 1986, Transplantation.

[10]  D G Altman,et al.  A note on the calculation of expected survival, illustrated by the survival of liver transplant patients. , 1991, Statistics in medicine.

[11]  James Lee,et al.  Covariance adjustment of survival curves based on Cox's proportional hazards regression model , 1992, Comput. Appl. Biosci..

[12]  M Pagano,et al.  Corrected group prognostic curves and summary statistics. , 1982, Journal of chronic diseases.

[13]  David R. Cox,et al.  Regression models and life tables (with discussion , 1972 .

[14]  J. Klein,et al.  Statistical Models Based On Counting Process , 1994 .

[15]  Ørnulf Borgan,et al.  A note on confidence intervals and bands for the survival function based on transformations , 1990 .

[16]  N. Breslow,et al.  Analysis of Survival Data under the Proportional Hazards Model , 1975 .

[17]  Stephen R Cole,et al.  Adjusted survival curves with inverse probability weights , 2004, Comput. Methods Programs Biomed..

[18]  M. H. Gail,et al.  Variance Calculations for Direct Adjusted Survival Curves, with Applications to Testing for No Treatment Effect , 1986 .