a soft and warm approach to randomization for clinical comment on An approach to randomization into surgical clinical trials

An approach to randomization into surgical clinical trials. BJS 2017; 104: 11-12. Having read with interest the article of Treasure and Baum I agree about the known difficulties with randomization which is even more difficult in less advanced regions(1, 2). We have experienced severe difficulties with randomization after joining the PulMiCC trial in 2013 (3,4). The question that the patient poses and the authors pointed out, “Which operation do you think is better?” is common when there is uncertainty over the long-term oncologic result of one operation compared to another. Even if not proved I am quite sure that the patient will pose the same question to the surgeon even if they are seen after randomization. Nowadays, after the introduction of DRGs, surgeons and hospitals are not happy to endorse another surgeon or hospital, and therefore it becomes more difficult to conduct randomized trials. There must exist a different method of randomization, which maintains high standards of treatment without compromising patient’s wishes, surgeon’s competence and hospital income. When two treatments need to be compared, participants are allocated to groups A and B, based on which clinic they go to. Participants receive either treatment A or treatment B depending on which trial participating centre they attend which is determined by the expertise at each of the participating centres. All patients who arrive at Centre A undergo treatment 1. All patients who arrive at centre B undergo treatment 2. It is not necessary, from an ethical point of view, to inform patients of other options if that option is not proven to be superior as such information can promote unwanted patient anxiety. Finally, there must be no interaction between the centres throughout the study except at the end of the trial when the results are analysed.