Reader Comments

Post a new comment on this article

Tackling publication bias

Posted by plosmedicine on 31 Mar 2009 at 00:20 GMT

Author: Liz Wager
Position: Publications Consultant
Institution: Sideview
E-mail: liz@sideview.demon.co.uk
Submitted Date: February 26, 2008
Published Date: February 26, 2008
This comment was originally posted as a “Reader Response” on the publication date indicated above. All Reader Responses are now available as comments.

This interesting study highlights the serious problem of publication bias and the fact that clinical decisions are often based on selective information. If details of all clinical trials are entered onto publicly accessible registers before trials start, attempts to suppress negative or unfavourable findings will be easier to identify. In the last few years, several journals (including PLoS Medicine) have given a big boost to trial registration by refusing to publish studies unless they have been properly registered.

In the USA, the recent FDA Amendment Act (passed in late 2007) will make results reporting compulsory -- at least for future US studies.

Such initiatives cannot be retrospective so clinicians and consumers should continue to lobby to ensure that the results of clinical trials for all marketed medicinal products are publicly available, even if the studies were conducted before the legislation took effect.

I would be interested to learn the authors' views about how trial registration and compulsory results reporting might reduce publication bias in the future.

Competing interests declared: I am a member of the WHO Scientific Advisory Group on trial registration and involved with various initiatives to improve trial results reporting