This review of Evidence-Based Medicine: How to practice and teach EBM by David Sharp appeared in The Lancet, 9th November 1996; 348: 1297. Reproduced by kind permission.
David L Sackett, W Scott Richardson, William Rosenberg, R Brian Haynes. Edinburgh: Churchill Livingstone. 1996. 250pp. ISBN 0-443-05686-2.
There may be some people who are opposed to motherhood or who dislike apple pies, but not many. Evidence-based medicine (EBM) is another of these icons. Who could be against it? Even so, EBM has not been received with universal acclaim. Unavoidably, EBM the book will be in part judged by how well it counters criticism of EBM the discipline. One criticism has been that its criteria are so restrictive that clinical judgement is undervalued. Those who think that would do well to open this book at section 4. On treatment, for instance, clinicians are told to integrate the evidence unearthed with their clinical skills by estimating the 'impact of the treatment on patients just like yours', a reference to the fact that the patient did not take part in the trial that established efficacy; and then there comes a balancing of 'the values and preferences of your patient with the regimen and its consequences'. Sackett and colleagues are here reiterating (they often say 'We always said...') that clinical judgement has to remain in play as systematically gathered evidence is applied. And you can even call upon the basic sciences when controlled trial evidence is not available.
Agnostics or atheists who take a caring view of their fellow human beings are often told, annoyingly, 'That proves that you are a Christian (or whatever) after all'. There is an element of this in EBM too. Sackett and colleagues point to evidence, from general medicine, surgery, and psychiatry (and there is some around now from general practice too) that clinicians already practice EBM to some extent. This contrasts with older claims that only a derisory proportion of medicine was evidence based. From audit and other studies it is not always clear that clinical management was based on evidence sifted exactly in the way that this book defines; the accord between management and the EBM 'bible' was sometimes uncovered retrospectively. Anyway, it would be nice to think that wisdom acquired without the aid of the modern tools of medical informatics and then shared with trainees is not always wrong.
I am ill-qualified to question the ease with which information is to be had from electronic and other sources. A generation gap I suspect, but even with nimble-fingered young enthusiasts at hand I can barely scratch the surface of what is available. Tailored EBM looks fine when the patient can be called back and you have time to dig.
Sackett and colleagues, replying the charge that EBM could be abused by rationers, note that applying EBM 'may raise rather than lower' the cost of care. Dare I suggest that there will always be far greater eagerness to stop what is unproven than there will be to provide true extra funding for EBM-supported new services?
Thus EBM the discipline is vigorously defended. EBM the book is very good, good for those who wish to incorporate EBM into their clinical work and for those who want to teach it to others. It is certainly practical. The means of access to sources is especially well described. The American College of Physicians' Journal Club is a favourite of these authors; editors may be less fond of it because the club's security staff throw out 98% of what journals print. There are plenty of simple case examples from the real world. This joint McMaster/Oxford/New York book is well-written too, even with some self-deprecating humour., However, parenthetic clauses make some sentences complex; and the footnote device is overused.
This is a highly structured text, as you might expect. The shape is 5 x 5 with icons (modern usage) as signposts - the question, the search, appraising the evidence, applying it, and seeing how well you are doing, all directed at diagnosis, prognosis, treatment, harm, and, the ultimate, care of the highest quality.
Evidence-based Medicine is important for many reasons. It will not kill off the standard medical text just yet though book publishers would do well to recognise the trend and adjust their thinking. More confirmation of the ease and success with which EBM techniques can be incorporated into the undergraduate curriculum will be required before students can set fire to their copies of 'Harrison'. The book is a manual of EBM if that is what the reader seeks and an excellent text on clinical epidemiology and on the critical reading of information if it is not. It has another strength. These proponents of EBM, now given more space than previously, recognise that EBM may be a 'gold standard' but it is not an altar. That can only win them friends.
David Sharp
The Lancet, London, UK