The Cox proportional hazards model is the most popular model for the analysis of survival data. The use of cubic spline funtions 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. The SAS macro RCS (restricted cubic splines) which implements the method is available.

 

Referenzen:
Heinzl, H., Kaider, A. (1997): "Gaining more flexibility in Cox proportional hazards regression models with cubic spline functions", Computer Methods and Programs in Biomedicine 54, 201 - 208
Heinzl, H., Kaider, A., Zlabinger, G. (1996): "Assessing interactions of binary time-dependent covariates with time in Cox proportional hazards regression models using cubic spline functions", Statistics in Medicine 15, 2589 - 2601
Hess, K. R. (1994): "Assessing time-by-covariate interactions in proportional hazards regression models using cubic spline functions", Statistics in Medicine 13, 1045 - 1062
Durrleman, S., Simon, R. (1989): "Flexible regression models with cubic splines", Statistics in Medicine 8, 551 - 561

These macros are provided at the github repository:

https://github.com/HaraldHeinzl/RCS