We wish to respond specifically to Mark Voysey's point that industry bias in reporting trial results is probable but that such bias has been refuted by an employee of a multinational drug company.
In our study, we found that industry-funded trials are associated with positive results in both medical and surgical randomized trials.1 Hirsch,2 an industry representative, argues against that finding. He states that Merck trial protocols undergo extensive scrutiny both internally and by external regulatory bodies and committees. Indeed, our review of trials found that drug trials achieved significantly higher mean quality scores than surgical trials (15.7 v. 13.4 out of a maximum of 21 points, p < 0.01). Companies such as Merck should be commended for their policy to publish the results of all “hypothesis-testing” clinical trials regardless of their outcomes. However, available evidence suggests that this policy is not universally applied by all drug companies. As reported in our study,1 there is a large body of literature (1572 randomized trials) for which there is a greater than 2-fold odds of a positive study result in industry-sponsored trials compared with non-industry-sponsored trials (odds ratio 2.3). We do not believe that this is, as Hirsch characterizes it, a “relation … of modest degree.” Ultimately, readers must critically examine all published trials with specific focus on the nature of industry involvement.
Mohit Bhandari Victor M. Montori P.J. Devereaux Department of Clinical Epidemiology and Biostatistics McMaster University Hamilton, Ont.
Footnotes
-
Competing interests: None declared.