Comprehensive Cohort Study: An Alternative to Randomized Consent Design in a Breast Preservation Trial

Summary If in a clinical trial only few patients consent to randomization, all those patients meeting the clinical inclusion criteria should be considered for analysis. Prerandomization is shown to be efficient only if patients can be persuaded to accept the prescribed therapy and do not introduce a self-selection bias. Under full informed consent conditions, as, for example, in a breast preservation trial, the first requirement must be questioned for ethical, the second one for methodological reasons. Conventionally randomized trials with the inclusion of patients having therapeutic preferences appear to be preferable. We briefly discuss the way of analysing the data from such a cohort study.