A summary of the Inventory paper describing new approaches to survival analysis in BMT
The CLINT project aimed at supporting the EBMT to develop its infrastructure for the conduct of trans-European clinical trials in accordance with the EU Clinical Trials Directive, and to facilitate International prospective clinical trials in stem cell transplantation.
The initial task is to create an inventory of the existing biostatistical literature on new approaches to survival analyses that are not currently widely utilised. These approaches will be made available to the statistical community within the field of stem cell transplantation by placing them on the CLINT portal.
The estimation of survival endpoints is introduced, with an emphasis on recent developments which complements standard analysis. The issues raised are new regression models that allow the estimation of time dependent effect for cause specific hazard, cumulative incidence and more generally mean response.
New development in multi state model, with notably, recent regression models that assess the influence of covariates directly on transition probabilities are detailed. Some recent test for comparing cumulative incidence function across treatment arm are introduced. The estimation of centre effect in multi centric studies is also documented. Sample size calculation in the presence of competing risks are then presented.
We close with the inventory of available packages and macro in R and SAS that implement the previous survival models.
The introduction of this paper invent.intro.pdf