BMJ  2008;336:170 (26 January), doi:10.1136/bmj.39469.465139.80

Editorials

Mandatory disclosure of trial results for drugs and devices

New US law will require public posting of all the main results and data on harms

Last week the drug industry was shown, twice, to have suppressed trial results that were commercially unfavourable—an all too familiar tale.^{1 2} But from September 2008, a new US law will mandate the public disclosure of such results, and it will no longer be left to investigative journalists and other researchers to dig out the findings of trials.

The FDA Amendments Act, passed quietly last year, rules that any ongoing clinical trial involving a drug, biological product, or device regulated by the US Food and Drug Administration (FDA) must be registered at clinicaltrials.gov and that, from 27 September 2008, triallists must start to post in that same registry the results of those trials.³ This applies to all clinical trials of drugs and devices except phase I drug trials (preliminary safety studies for new products) and small feasibility studies of a device. Furthermore, it covers all trials—whether or not they are conducted and sponsored by industry and wherever they are conducted—if the products concerned need approval by the FDA.

This law provides much needed enforcement. "Responsible parties" (sponsors or principal investigators of trials) who fail to comply will be fined $10 000 (£5100; 6900) for each infringement, with no upper limit, and they will be publicly named and shamed in non-compliance notices posted on clinicaltrials.gov. Moreover, triallists will not be able to drag their feet—the FDA requires disclosure of the results within one year of the date that "the last subject was examined or received intervention for purposes of final collection of data for the primary outcome." Although the act allows for delayed disclosure of results in exceptional circumstances, for instance in the interests of national security, it does not mention delay while awaiting publication in a peer reviewed journal. Will editors count such disclosure as prior publication and refuse to consider such trials for publication in their journals?

Urged on by the International Committee of Medical Journal Editors (ICMJE), many journals including the BMJ will no longer consider unregistered trials for publication. The ICMJE ruled last summer that its member journals would consider trials whose results had been disclosed at a registry, as long as this did not exceed one table of results or an abstract of up to 500 words.⁴ The architects of the Ottawa statement on trial registration (http://ottawagroup.ohri.ca/) and the World Health Organization’s international clinical trials registration platform (www.who.int/ictrp/en/) have called for fuller disclosure, but they have assumed that this would occur after publication in a peer reviewed journal. Yet the FDA has now mandated release of much more information, and this poses quite a challenge for editors and researchers. From September 2007, responsible parties for eligible trials must post on clinicaltrials.gov two tables giving all the key results for the main outcomes, and from next year they must post two tables of any harms reported during the trial. Further public consultations in 2009 may lead to further expanded disclosure under the act.

The BMJ will consider disclosed trials and urges other journals to do the same for several reasons. Firstly, disclosure will be a legal requirement, so there is nothing editors can do about it if they still want to publish important trials of drugs and devices. Moreover, journals will continue to add value by publishing useful and readable trial reports that clinicians, the media, and patients can interpret and use. And, most importantly, the results disclosed for the FDA will not have been externally peer reviewed and will be preliminary. Peer review not only provides a stamp of quality assurance, it often leads to reanalysis of results.

Two challenges remain. Will a journal still want to publish a paper if its results have been posted on clinicaltrials.gov; if they have been debated widely in the mass media along with quotes from respected clinical experts, and—as is often the case—if they have led to a widely publicised rise or fall in the share price of the company that sponsored the trial? And, in the long run, will the role of peer reviewers and medical journals as the gatekeepers of new medical evidence become redundant?

At the BMJ we support this great leap forward for public disclosure. And, if you have an important and robust trial of a drug or device that you would like published before disclosing the results on clinicaltrials.gov, contact us early and request fast track review. We do not plan to turn down trials that have already been disclosed, but—through rapid open access publication—we can get the full peer reviewed work into the public domain quickly on bmj.com and, where the results are important, we are keen to do so.

¹ BMJ, London WC1H 9JR

Shortcuts, doi: ; News, doi: 10.1136/bmj.39465.409051.80 10.1136/bmj.39468.610775.DB

---

Competing interest: TG is the BMJ’s senior research editor, whose remit includes encouraging authors to submit research to the BMJ.

Peer review and provenance: Not commissioned; not externally peer reviewed.

References

1. Schering-Plough Press Release. Merck/Schering-Plough Pharmaceuticals provides results of the ENHANCE trial. 2008. www.schering-plough.com/schering_plough/news/release.jsp?releaseID=1095943.
2. Turner EH, Matthews AM, Linardatos E, Tell RA, Rosenthal R. Selective publication of antidepressant trials and its influence on apparent efficacy. N Engl J Med 2008;358:252-60.[Abstract/Free Full Text]
3. US Food and Drug Administration. Law strengthens FDA. 2007. www.fda.gov/oc/initiatives/advance/fdaaa.html.
4. Laine C, Horton R, DeAngelis CD, Drazen JM, Frizelle FA, Godlee F, et al. Clinical trial registration. BMJ 2007;334:1177-8.[Free Full Text]

This article has been cited by other articles:

• Lenzer, J., Brownlee, S. (2008). An untold story?. BMJ 336: 532-534 [Full text]