Abstract:
|
This paper advocates the landmarking approach that dynamically adjusts predictive models for survival data during the follow-up. This updating is achieved by directly fitting models for the individuals still at risk at the landmark time-point. Using this approach, simple proportional hazards models are able to catch the development over time for models with time-varying effects of covariates and/or data with time-dependent covariates (biomarkers). Robustness of the prediction can be enhanced by ignoring information beyond the prediction horizon (Stopped Cox regression). To smooth the effect of the landmarking, sequences of models are considered with parametric effects of the landmark time-point and fitted by maximizing appropriate pseudo log-likelihoods that extend the partial log-likelihood to cover the landmarking approach. The methodology will be demonstrated on real life clinical data.
|