Skip to content
Journal article

Gaining more flexibility in Cox proportional hazards regression models with cubic spline functions

Heinzl H, Kaider A...(+2 more)

Computer Methods and Programs in Biomedicine, vol. 54, issue 3 (1997) pp. 201-208 Published by Elsevier Sci Ireland Ltd

  • 17


    Mendeley users who have this article in their library.
  • Citations

    Citations of this article.
  • Views

    ScienceDirect users who have downloaded this article.
Sign in to save reference


The Cox proportional hazards model is the most popular model for the analysis of survival data. The use of cubic spline functions allows investigation of non-linear effects of continuous covariates and flexible assessment of time-by-covariate interactions. Two main advantages are provided-no particular functional form has to be specified and standard computer software packages like SAS or BMDP can be used. A SAS macro which implements the method is presented.

Author-supplied keywords

  • Binary time-dependent covariate
  • Non-linear functional relationship
  • Proportional hazards regression model
  • Restricted cubic spline function
  • SAS macro
  • Time-by-covariate interaction

Find this document

Get full text

Cite this document

Choose a citation style from the tabs below