|
|
|
|
The basis of clinical practice is evidence that drugs are effective and safe. That is a binary justification for the choice to prescribe or not to prescribe. How do we reach such a black and white position from the evidence we actually have? In other words how do we weigh the evidence and tip the scales one way or the other to reach a decision? Are such decisions simply rational or are they, as neuroscience suggests, intrinsically emotional and maybe subjective?
A desire for evidence based practice has been the big idea in medicine for my working life time. Indeed the idea has proved contagious; politicians now solemnly advocate evidence based policing, for example. The spiritual home of evidence based practice is the Cochrane collaboration. It came into existence in 1993. It is named after a relatively obscure British epidemiologist called Archie Cochrane. Its brand fast acquired an iconic status. This surprised me personally because data synthesis (meta-analysis) was the preferred methodology and it was in no way new. But Cochranites multiplied fast, developed a rule book of do’s and don’ts and a slightly them and us mentality. The conclusions of a Cochrane review soon carried surprising status. In the case of Cochrane’s psychiatry wing, this was always somewhat at odds, for me, with its scruffy offices in Oxford. The scruffiness was largely imposed by uncertain funding but it also fitted the ethos of the 1990s. Cochranites had radical ideals to de-construct the medical hierarchy (eminence based medicine?) and replace it with a more democratic evidence base.
The Cochrane collaboration still supports good work, and no one can dispute the premise that evidence should underpin medical practice. But, the problem with evidence based medicine is that sometimes, actually pretty often, there isn’t much really good evidence. Practice cannot stop until there is. Worse, you can include in your Cochrane review stuff that really should not be included. If you insist on including dose finding trials when the doses were too low you can undermine evidence for efficacy, and if you include doses that were too high you magnify the burden of adverse effects. Finally if you insist on unrealizable trial standards for certainty, you end up doubting all the evidence you have. So Cochrane reviews can become orgies of uncertainty – deflated effects, inflated side effects, all evidence debunked as being of a low standard. You may need to be quite expert to know what is worth considering and what isn’t and expertise was something Cochrane set out to debunk as opposed to ‘the evidence’.
In my opinion, there have been times when Cochrane zealots have tarnished the idealistic image of ‘the collaboration’. I am tempted to say that it has become a bit like a failing political party – leaderless, easy to join and harbouring bewilderingly strong irrational opinions. The strongest opinions have inevitably been against the comic villain of schoolboy socialism, the drug companies. This has led to a number of ‘controversial’ reviews of the literature on antidepressants, Tamiflu, treatment for ADHD. Along with an intellectually challenged ally, the British Medical Journal, these ‘Cochrane reviews’ have helped to stoke uncertainty and controversy, at least where I live.
But let’s get back to decision making. Criticism of the Cochrane approach is ultimately respectful. What should be particularly applauded is its latest GRADE system (http://tech.cochrane.org/gradepro). Cochrane rules on the evidence hierarchy traditionally revered the RCT and the summary of RCTs that could form a meta-analysis. Observational studies were treated as lower down the food chain. This approach has now been modified by the introduction of a way for us to describe how we rank evidence. In the words of my wise colleague Andrea Cipriani: ‘‘the positive value of GRADE is that it allows people to do recommendations in a structured and transparent subjectivity‘‘. I love that because it actually models how we really make decisions.
May I end by wishing my friends around the world a very happy Christmas and a prosperous New Year. Long may subjectivity be transparent.
Guy Goodwin
ECNP President |
|
|
|
|
|
|
|
|
|
|
|
