Authors' choice of study was ill informed

EDITOR—McCormack and Greenhalgh's suggestion that those involved in running and reporting clinical trials might be able to engineer a worldwide “groupthink” spin on the results is an intriguing notion.1 But their choice of the United Kingdom prospective diabetes study (UKPDS) as an example to support their hypothesis is ill informed given the manner in which this study was reported.

We note with interest Greenhalgh's earlier commentary on an article by Horton concerning the “spin that authors place on their own work.”2 In this, she highlighted the “unjustified assumption that this spin is necessarily evil, insidious, and the last remaining bastion of caprice in the otherwise objective terrain of scientific publication,” and she challenged Horton to “produce a single, clinically important instance of scientific heads being turned by rhetoric and rhetoric alone.”

There was a complete embargo on all outcome data from the United Kingdom prospective diabetes study before their presentation at a meeting of the European Association for the Study of Diabetes on 12 September 1998. To avoid the usual scenario whereby conference reports are given wide publicity before peer reviewed manuscripts are available, the UKPDS Group worked closely with the editors of the Lancet and the BMJ to ensure that as many of the primary results as possible were published in five peer reviewed papers on the same day as our conference presentation. In addition, 100 slides illustrating the published data were made available on our website at midnight that day (www.dtu.ox.ac.uk/ukpds/).

We believe that the manuscripts …