Editor's Choice Editor’s Choice

The confidence to doubt

BMJ 2008; 336 doi: https://doi.org/10.1136/bmj.39524.482535.47 (Published 20 March 2008) Cite this as: BMJ 2008;336:0
  1. Fiona Godlee, editor, BMJ
  1. fgodlee{at}bmj.com

    In the documentary The English Surgeon reviewed this week (doi: 10.1136/bmj.39511.711227.3A), English surgeon Henry Marsh debates with his Ukrainian colleague whether success in medicine is about dedicating your life to saving patients or accumulating academic honours. Why should it be one or the other, I found myself asking. So I’m intrigued to find this conflict between academic and clinical medicine as a subplot in our Head to Head article on metabolic syndrome (doi: 10.1136/bmj.39484.636586.94, doi: 10.1136/bmj.39477.500197.AD).

    Edwin Gale, an academic clinician, argues that we should dump the syndrome because it has no clinical value. In the syndrome’s defence, George Alberti and P Z Zimmet, both also academic clinicians, say its value lies in its ease of use in primary care. Gale says it has spawned a flourishing academic industry but is hardly used by practising clinicians. Alberti and Zimmet characterise the syndrome’s critics as living in a specialised academic world, out of touch with the “real world” of primary care.

    Is there anything edifying in this refined form of mud slinging? The BMJ invites the combatants into the pit (though we do ask them to bring their own mud), so it’s fair to assume we think there is, even though the 7th century Chinese sage Seng Ts’an said “The struggle between ‘for’ and ‘against’ is the mind’s worst disease.” I for one came away from reading these two well argued pieces with a conclusion (and you will no doubt draw others). Since there is uncertainty as to whether it’s clinically useful for GPs to measure people’s waists, we need a randomised trial. Alberti and Zimmet say that focusing on the syndrome results in better management of people with type 2 diabetes, but they give no reference so I assume no “real world” trial has yet been done.

    Nick Freemantle and Alar Irs make a strong case for randomised trials in another context—drug safety (doi: 10.1136/bmj.39491.493252.80). They warn us against relying on observational studies for pharmacovigilance by telling the cautionary tale of aprotinin. A prominently published observational study concluded that this antifibrinolytic substantially increased the long term risk of death after cardiac surgery. Despite the likelihood of confounding (the drug may have been given preferentially to patients at increased risk of bleeding, for example), these observational data probably contributed to the stopping of a large randomised trial of aprotinin and the drug’s subsequent withdrawal. Now we may never know whether aprotinin could save lives. A recent Cochrane review of randomised trials found no evidence of increased risks from its use.

    We must constantly doubt the products of academia—Robert Merton called this “organised scepticism.” And we must also develop the confidence to doubt our own practice, say PW Teunissen and Tim Dornan in the second article in our Competent Novice series (doi: 10.1136/bmj.39434.601690.AD). This is the essence of lifelong learning, without which it will be hard to survive in medicine, they say: “Lifelong learning means striking the right balance between confidence and doubt.” “We may and probably will kill patients,” says Henry Marsh in the documentary. What we must never do, whether in academia or clinical practice, is stop striving to make things better.

    View Abstract