It's difficult now to recapture the excitement of that time, the feeling that a whole tradition of authoritarian practice was about to be swept away by a flood of evidence-based rationalism. I remember vividly the way I struggled to understand basic epidemiological concepts, and how they transformed my view of the implications of research evidence for a given clinical situation. Because that was really the point of EBM — not the evidence itself, but its relationship with the complex world of patients' multiple problems and varying, sometimes contradictory, wants and needs. You couldn't really grasp the essence of EBM unless you were a clinician, because it was founded on the idea that it was the ability to identify and define a specific lack of knowledge about how to tackle a problem arising in clinical practice that should set off the whole process of evidence-finding and interpreting. You didn't have to be a card carrying member of the David Sackett fan club to feel at least a little seduced by this re-alignment of the power balance in favour of clinicians.
How this was stood on its head and turned into a model of clinical practice in which evidence is both worshipped and withheld from ordinary clinicians (by which, of course, I usually mean GPs) will one day not doubt be studied by sociologists and historians of the NHS. They might draw an analogy with the Medieval church's reluctance to have the Bible translated into English — much safer for its full text to be restricted to an educated elite that understood Latin. I can only give my own opinion, and no-one should take this as objective, or even correct. From my point of view, what I see is that a revolutionary idea that I expected to improve GPs' knowledge and skills for the benefit of their patients and (not a minor point) enhance their autonomy, has been deformed into a straightjacket that has reduced GPs' autonomy and often rendered their knowledge and skills irrelevant.
It is easy to see how this arose. If I, as an enthusiast (a.k.a. early adopter), found it difficult to acquire EBM skills, then it was reasonable to doubt that the unenthusiastic majority (a.k.a. late adopters) would find it worth their while to grapple with likelihood ratios and research methodology. Nor is it feasible, even for an enthusiast, to critically appraise a paper before each clinical decision. So, since the object of the exercise was to get evidence into practice, the sensible thing to do was to set up committees of experts to appraise the evidence, and produce guidelines with recommendations based on the strength of the evidence. This is so self-evidently rational that it is rarely challenged. So here is a challenge.
The first problem with guidelines as currently experienced is that they are opaque. I recently had the bright idea of reviewing the NICE guideline on depression1 in order to understand the recommendations, and subsequently to organise an educational event in which GPs could compare the recommendations with the evidence. While the guideline does link recommendations to the strength of the evidence, the evidence itself is very difficult to find and is not cited in the guideline itself. This so-called ‘evidence-based’ document contains many recommendations at level C (expert opinion only) but has the character of an ex cathedra statement of absolute truth. It is, in fact, a statement of the considered wisdom of an expert group, and is taken in that spirit by most GPs. The evidence itself and the findings of the critical appraisals that the committee must have considered are not made easily accessible.
The second problem is that people who sit in these expert groups (and I have discussed these matters with a number of them) appear to find nothing wrong with their methodology. They tend to fulminate about GPs' lamentable ignorance of the content of most guidelines and wonder how GPs can be encouraged to follow them. The answer is that if you treat people like idiots who cannot be trusted to make up their own minds about the evidence, they will behave like idiots who can't find any evidence to make up their minds about anyway, and will (unless appointed as a clinical lead by their PCT) do only what is necessary to do to make up their QOF points and avoid getting into trouble.
So what is the solution? The key issue is that the central message of the EBM movement — the need for clinicians to understand appropriate evidence when making a clinical decision — has been ignored in order to circumvent the perceived (and, one must confess, real) lack of EBM skills among clinicians. Rather than put resources into developing critical appraisal, searching, and clinical epidemiology skills among clinicians, resources have been put into providing pre-packaged recommendations in order, by hook or by crook, to ‘get the findings of research into practice’. Unfortunately, rather like convenience food, these products are both indigestible and unhealthy. Can you imagine someone being paid to prepare food but not being expected to know how to cook?
The more paranoid among us might think that this is some kind of sinister plot to destroy professional autonomy. We might rant against ‘lickspittle committees’ whose sole aim is to dictate clinical practice and control our every action (and thank you George Galloway MP for your indefatigability in resurrecting so expressive a word as lickspittle). I am, of course, neither paranoid nor a ranter. I want simply to raise the question of whether in the long run it might be more sensible to invest in effective EBM training and improved online resources so that GPs and their patients can make informed choices about the consequences of research findings for clinical decisions, rather than rely on the blunt tool of guideline recommendations as presently formulated.
I would like to see NICE guidelines presented online as HTML documents as well as in PDF format so that enthusiasts like me can, within a consultation, quickly find the appropriate recommendation and, by clicking on a link, examine a précis of the evidence and its appraisal upon which the recommendation is based. This would enable me to help my patients to make informed decisions rather than merely having to take NICE's word for it.
I would also like the RCGP to consider the place of EBM in GP vocational training and continued professional development. I am fully aware of the difficulties of teaching EBM (how could I not be after the experience of so many workshops?). It is unfortunately true that if you ask most GPs ‘would you like to join a critical appraisal group?’ they either stare at you in horror or simply say ‘no’, and rapidly find someone else to talk to. Nevertheless, EBM skills seem to me to as essential in modern health care as the taking of a history and performing a clinical examination.
By analogy, just as you rarely, if ever, perform a full clinical examination in general practice, so you will rarely do a full search and critical appraisal. However, knowing how to do it, and the internalising of the epidemiological concepts involved in applying research findings to individual clinical problems change the way clinical dilemmas are perceived. Even if the decision turns out to be the same (for example, not prescribing antibiotics for a sore throat), the process of negotiating it in a transparent way informed by an understanding of the consequences of the evidence seems preferable to me than simply saying to patients ‘antibiotics don't work for sore throats’ or ‘we don't prescribe antibiotics for sore throats any more’. What's more, in the culture of continuing learning and professional appraisal, there is nothing more immediate than addressing problems identified directly from clinical practice.
So, while it may appear that EBM is dead, it has never been more necessary. If GPs are to remain autonomous clinicians (making context-specific, patient-centred decisions) rather than government-directed disease managers, they cannot afford to ignore the tools that enable them to use the information that informs clinical practice. Just as knowledge is power, lack of knowledge represents impotence. The profession cannot afford to allow control of the knowledge that guides clinical practice to remain entirely someone else's affair, however expert or well intentioned they may be. I would go so far as to say that lack of EBM skills is a greater threat to professional autonomy than the Shipman Inquiry, revalidation and QOF all rolled into one. Ever the optimist, I live in hope that sooner or later the penny will drop, and the profession will at last take seriously the need for systematic and universal training in EBM. Until that day arrives, all I can do is continue to plug away at the rather intriguing task of using evidence in my own consultations. If nothing else, it makes life a bit more interesting.