Using evidence on yourselfBMJ 2000; 320 doi: https://doi.org/10.1136/bmj.320.7228.0 (Published 15 January 2000) Cite this as: BMJ 2000;320:0
Some argue that the appearance of evidence based medicine has transformed medicine, and a table on the BMJ's website compares and contrasts the old world and the new (www.bmj.com). The central change is perhaps that systematic analysis of the evidence becomes the main source of knowledge rather than expert opinion. Some doctors think the whole thing poppycock. But even they will probably be fascinated by Chris Del Mar's evidence based analysis of what to do about his own asymptomatic haematuria (p 165). It illustrates changed thinking.
It was discovered accidentally when he was teaching students. His first thoughts were traditional, remembering the possible sources of bleeding and disease processes. Next, to the textbook—Harrison's Principles of Internal Medicine. It suggested a swathe of investigations, including—most helpfully—checking blood relatives for urine abnormality.
Then to the general practitioner, who suggested playing for time and then starting some investigations. So there was time to examine the evidence, but what was the question to ask? This is one of the hardest parts of evidence based practice. Eventually Del Mar hit on the empirical question that mattered: “What is the chance of having a serious condition with asymptomatic haematuria?” Standard textbooks didn't help, and nor did the Cochrane Library. Medline produced 230 articles. A lot of reading produced two studies (one from the BMJ, I'm glad to see) that followed up large numbers of patients who had been screened and discovered to have asymptomatic haematuria. Tiny numbers had anything serious, and in one study those who didn't have any bleeding were just as likely to have something serious. “Expectant observation” is thus Del Mar's strategy.
One of the impediments to practising evidence based medicine is learning how to do it. Another learning problem is learning how to do surgical procedures at a time when the public is becoming intolerant of suboptimal results. The senate of surgery has said that “there should be no learning curve as far as patient safety is concerned,” immediately raising problems about how to learn to do something you've never done before. Three paediatric cardiac surgeons describe how they went about learning a new procedure for aortic valve replacement (p 171). They undertook a course in aortic root surgery, refined their technique on cadavers, did the first operation with an expert, and then assisted each other with the operation. The results were good. Ironically, in Britain it is impossible to practise on animals, and surgeons are not permitted to visit an experienced surgeon and do the operation. It's the expert who must travel.