Tuesday, April 18, 2006

Evidence About How Commercial Firms Control the Research They "Sponsor"

Last week's Journal of the American Medical Association (JAMA) had a short, but very important article on how the commercial sponsors of clinical trials may manipulate how the trials' results are disseminated. [Full citation: Gotzsche PC et al. Constraints on publication rights in industry-initiated clinical trials. JAMA 2006; 295: 1645-46.]

The investigators reviewed the original trial protocols from all "industry-initiated randomized trials that were approved in 1994-1995 by the Scientific-Ethical Committees for Copenhagen and Frederiksberg in Denmark and that resulted in publication." The investigators also reveiwed all resulting trial publications. They similarly reviewed all 2004 protocols (but not publications, as it is too early for their results to be published).

43 of 44 trials from 1994-5 had "multinational pharmaceutical firms as sponsors." For half of the trials, the protocol assured that the sponsor would own the data and/0r would need to approve any manuscript submitted for publication. 41% of trials had other constraints by the sponsors on publication, such as requirements that the sponsor review the manuscript, or that the sponsor's views must be represented in the manuscript. 61% of the 2004 protocols required that the sponsor would own the data and/or would need to approve any manuscript submitted for publication, and 32% had other constraints.

Furthermore, the investigators noted that none of the constraints on publication found in trial protocols in 1994-5 were mentioned in the articles published about the trials. But none of the trial publications "stated that investigators had access to all of the data generated from the trial or had final responsibility for the decision to submit for publication without requiring approval from the sponsor."

We had posted a while back about the study by Mello and colleagues that showed that many US medical schools and academic medical centers would be willing to contractually cede control over aspects of clinical research, and publications resulting from such work to the work's commercial sponsor. That study was of what academic contracting officers said they were willing to do. This new study was of how clinical research was actually designed. It shows that in Denmark most clinical trial protocols as early as 1994 had already ceded substantial control over publication of trial results to commercial sponsors. The studies' nominal "principal investigators" did not have full control over the data they collected, nor over what they could say about it in print. And none of the investigators were apparently able to even publicly acknowledge their lack of such control.

This quantitative study adds to vivid, but case-based data (such as that of the unfortunate case of Aubrey Blumsohn's inability to get the data from his own research from the work's commercial sponsor, as per this post) suggesting that the ostensible sponsors or funders of research increasingly control how the work is done, and how the results are disseminated.

Thus it gets harder and harder to trust that published clinical research studies sponsored by commercial firms are not manipulated by studies' sponsors to produce results favorable to their financial interests. There is more and more reason to be concerned that the altruism of the people who volunteer to participate in human clinical research may be exploited. Patients and physicians need to be increasingly skeptical of the results produced by commercially sponsored research, and applying them to clinical decisions.

If we cannot persuade commercial research sponsors to let researchers do their work unaffected by commercial interests, then the only alternatives may be the heavy hand of government regulation, or even an outright ban on commercial firms having any role in clinical research.

No comments: