Saving the Medical Literature from Sponsorship Bias

Doucet and Sismondo, of the Philosophy Department of Queen's University in Ontario, recently wrote about the problem of sponsorship bias in the medical literature (http://jme.bmj.com/cgi/search?andorexactfulltext=and&resourcetype=1&disp_type=&sortspec=relevance&author1=doucet&fulltext=&volume=34&firstpage=; access requires subscription)

Doucet and Sismondo cite their fellow Canadian philosopher, Arthur Schafer, who wrote about his "sequestration thesis" in the same journal. In turn, I relied on Schafer's work to develop my own "divestment strategy" in HOOKED.

In summary, the present authors review the extensive evidence showing undue sponsorship bias, and the specific ways that medical journal reports of pharmaceutical trials may introduce inappropriate commercial bias. They then look at three commonly proposed solutions--increased financial disclosure; the CONSORT standards; and trial registries--and show precisely how each is able to address only a small portion of the different causes of inappropriate bias. They conclude that only radical changes, removing industry financing largely from drug trials, will be effective in the end (while acknowledging the political difficulty of implementing the radical changes).

Doucet M, Sismondo S. Evaluating solutions to sponsorship bias. J Med Ethics 34:627-30, 2008.