To re-censor, or not to re-censor, that is the question: Critical considerations when applying statistical methods to adjust for treatment switching in clinical trials

To determine when re-censoring should be incorporated in statistical analyses undertaken to adjust for treatment switching in randomised controlled trials, and to demonstrate the utility of inverse probability weighting (IPW) as an alternative to re-censoring. Treatment switching often has a crucial impact on estimates of the effectiveness and cost-effectiveness of new oncology treatments (Figure 1). Switching adjustment methods such as rank preserving structural failure time models (RPSFTM) and two-stage estimation (TSE) estimate ‘counterfactual’ (i.e. in the absence of switching) survival times and incorporate re-censoring to guard against informative censoring in the counterfactual dataset. However, recensoring often involves a loss of longer term survival information which is problematic when estimates of longterm survival effects are required