Editor's choice

Delusions and dupes

Here at the BMJ we like to think we're a bit different from other journals, and this week there's some evidence to support our delusion. Matthias Egger and colleagues studied randomised controlled trials published in five general journals and found that most of the trials randomised individuals and studied drug interventions (p 1253). Those in the BMJ, however, were less likely to test drug interventions and more likely to involve cluster randomisation. In their letter Egger and colleagues hypothesise that these trials may be less affected by the traditional biases, including commercial interests and lack of patient involvement.

Australians are getting fed up with commercial interests. The Australian Museum runs prestigious awards for science and science communication, but its decision to establish a new one for medical journalism sponsored by Pfizer has provoked opposition from journalists, academics, and doctors (p 1258). In describing the row Melissa Sweet says that in the past she has taken sponsored trips and entered journalism awards. "But no more. With compelling evidence to show that close ties with industry can influence doctors' behaviour, there's no reason to expect journalists would be any different." And it's not just journalists who are getting sceptical. Australian teachers have been staying away from workshops sponsored by Philip Morris on encouraging children to resist drugs, smoking, drinking, and bullying (p 1206).

David Carvel was also clearly annoyed by a manifestation of commercial interests. He opened a package delivered to his surgery marked "urgent"only to find an advert for an expensive drug (p 1259). "I was incensed by this shameless advertising gag and incensed that I had been duped by it." His complaints got nowhere, but in his personal view he encourages others to complain"not only are drug companies getting more powerful, they are now taking over the language."

One thing that puzzles researchers is why antithrombotic drugs are consistently underused in atrial fibrillation, despite clear evidence of benefit. On p 1218 P J Devereaux and colleagues add to an increasing body of understanding that doctors and patients differ in their attitudes to risks. The researchers presented patients at risk of atrial fibrillation and doctors with scenarios about strokes and bleeding and asked them to trade off the risks. They found that the minimum number of strokes that needed to be prevented for warfarin to be justified was lower (and the number of excess bleeds acceptable higher) for patients than for doctors. But, importantly, there was great variability in both groups. So the only answer is to ask patients what they value when deciding on treatment.

Footnotes

To receive Editor's choice by email each week subscribe via our website: www.bmj.com/cgi/customalert