A friend of my husband’s, a physiotherapist and yoga instructor in Queensland, is staying with us this week. Last night, he said to me that EBM is a medical model and can’t be applied to everything (I had mentioned a literature search I had completed late last year for evidence about yoga for treatment of pain). His comment about yoga and pain was that it would be very difficult to construct a trial of yoga for pain because: therapist expressions and attitudes during therapy (patients might guess something wasn’t right), the question of sham yoga, and whether it was ethical to give people in pain fake therapy. It is true that there has been some debate about the health of EBM in the literature lately (for example the opinion piece written by Des Spence titled Evidence Based Medicine is Broken) but is it the model itself broken or is it something else? It is true that RCTs are the gold standard in trial design, but it is also true that RCTs cannot be applied to all types of research questions. Carl Heneghan, Director of CEBM, wrote a BMJ blog post recently about the health of EBM. It is true that published evidence is suffering from many ills including poor trial design, peer review sheannigans, and research that doesn’t address patient needs, but this is the about quality of the evidence, not the quality of EBM as a model. EBM teaches you to critically appraise the literature and ask questions such as – does this apply to my patient? Evidence is only a third of the EBM equation – the others, which are equally important and are being ignored in this argument, is clinicial judgement and patient values. Evidence is important, but it only makes up part of the picture.