Statistics, Department of

The R Journal
Date of this Version
12-2017
Document Type
Article
Citation
The R Journal (December 2017) 9(2); Editor: Roger Bivand
Abstract
Treatment switching in a randomised controlled trial occurs when participants change from their randomised treatment to the other trial treatment during the study. Failure to account for treatment switching in the analysis (i.e. by performing a standard intention-to-treat analysis) can lead to biased estimates of treatment efficacy. The rank preserving structural failure time model (RPSFTM) is a method used to adjust for treatment switching in trials with survival outcomes. The RPSFTM is due to Robins and Tsiatis (1991) and has been developed by White et al. (1997, 1999).
The method is randomisation based and uses only the randomised treatment group, observed event times, and treatment history in order to estimate a causal treatment effect. The treatment effect, , is estimated by balancing counter-factual event times (that would be observed if no treatment were received) between treatment groups. G-estimation is used to find the value of such that a test statistic Z ( ) = 0. This is usually the test statistic used in the intention-to-treat analysis, for example, the log rank test statistic.
We present an R package, rpsftm, that implements the method.
Included in
Numerical Analysis and Scientific Computing Commons, Programming Languages and Compilers Commons
Comments
Copyright 2017, The R Foundation. Open access material. License: CC BY 4.0