Study 329
BMJ 2015; 351 doi: https://doi.org/10.1136/bmj.h4973 (Published 17 September 2015) Cite this as: BMJ 2015;351:h4973
We need your help with hidden dataFill in this form with details of any examples you have seen
All rapid responses
The responsibility for the study that is unpublished or misreported, must be shared between the principle researcher, the funding body, the institution where it was conducted, the organisation that provided national data where this was used, and with the research ethics committee that approved the study. (I propose a closed season for editor and referee harrying for now, their shortcomings have been sufficiently aired of late, and anyway the majority are not obdurate.)
An understanding of the natural history of the individuals and bodies involved is required if measures are to be put in place to prevent the scandals to which your leader refers.
Researchers are but mortal and may change their jobs and abandon their study, or if they change jobs and take the data with them intent on completion, this is thwarted by the overwhelming demands of the new job and new interests.
The funding body's trustees owe a responsibility to their founders and to the Charity Commissioners to ensure that the completed work is appropriately published, but chasing a grantee presents problems particularly when he/she has moved abroad. Where the award was to an institution, it is uncommon to hear of charitable funding bodies clawing back funds when the terms of the contract were not met. Where funding comes from a commercial organisation, early on, principled institutions declined contracts that made publication conditional on their approval.
As for national registers that have generously provided data to researchers, while it is implicit that their services will lead to publication, there are no sanctions against defaulting researchers or their institutions.
Since the early 1960s, the research ethics committee who made publication an essential condition of a study being approved should have constituted a long stop. Subsequently they have been fine and coarsely tuned, but invisible and abandoned trials have still been identified.
It all boils down to political will; until an organisation is designated to be responsible for policing the publication of research and provided with authority backed by adequate facilities to exercise it, RIAT will continue to have work to do.
Competing interests: No competing interests
Legally, it is relatively easy for governments to require that data be made public
The matter of making public the data on each subject participating in a clinical study has been debated in the US since at least 1975, when I participated in those discussions and arranged to allow researchers access to health data. In general, whenever governments provide funding for research, they create a “contract” and can impose, as a condition of receiving funding, that the data be made public (with appropriate safety constrains). As a condition of paying for drugs, the government can require manufacturers and sellers to make public all data relevant to use, such as safety, efficacy, side effects, doses, molecular structure. Concealing or misrepresenting material information is often illegal in the US, and entities involved in misrepresentations should reimburse taxpayers. Users and prescribers of drugs are entitled to analyze the data to verify drug properties.
While working in the office of the US Secretary for Health, Education and Welfare (now renamed Health and Human Services, HHS), I proposed many times that the US government require that all entities with public funding, and those requesting FDA approval, make data public. The government has authority to require it. I worked for The White House and the Office of Management and Budget (OMB). Among many duties, OMB reviews all surveys and forms, reviews statistical programs, government grants and contracts and government rules or regulations. In general, with regard to entities receiving government funding (such as grants or contracts, the typical source of funding used in medical research), the government can and does impose a variety of requirements as a condition to get the funding. These requirements include reporting employment patterns, financial data, etc. The government requires that recipients of federal funding explain what they did with the money; provide reports on outcomes, provide products, etc. If the government pays for a Jeep, it has the right to get the Jeep; if it pays for data, it should get the data. The government can require that the data collected, with individual identifiers removed or distorted (e.g., no subject names, only an ID code, no date of birth) be provided to the government and be available to qualified individuals (e.g., researchers with specific criteria).
The FDA could require the submission of individual data on a format suitable for further analysis (as a condition to evaluate it). There are many ways in which data reported to the FDA could be made available to qualified researchers (with appropriate mechanism to protect individual’s privacy). These actions could be taken by President of the US (POTUS) via an executive order. It can be done in one week.
I worked for the National Institutes on Drug Abuse (NIDA) to supervise the design and implementation of one of the largest health data systems in the US. At that time the federal government funded about 1,500 clinics that provided treatment to about over 30,000 individuals every month. We implemented a system to separate data collection from data analysis. The data was reviewed for error detection and correction and eventually gathered into a data base without individual’s names. Before my time, government programmers spent months writing special purpose software to create tables. Because drug abuse is a critical issue, I had two “bosses”. I reported to the director of NIDA who reported to a US Secretary. However, there was a separate office for drug abuse at The White House. I got frequent requests for data analysis directly from The White House or the US Secretary (who got them from Congress). We spent huge amounts of time and money responding to data requests (e.g., tables and graphs of users of specific drugs by age, sex, geographic location, etc.). We could not keep up with the demand. I separated data collection from data analysis and used off the shelf statistical software to analyze the data. The reports did not look as nice as those made with special software. So we used scissors and a typewriter. We wrote nice titles and pasted the data on blank paper with nice titles.
To improve the quality of our analysis, I arranged with a university to become the official “deposit” for drug abuse data. The data was probably more “confidential” than data from clinical trials. The individuals treated for drug abuse could be subject to substantial penalties. They often chose drug abuse treatment instead of jail.
Even with the primitive computer systems of the 1970s, it was feasible to make health data available to scientists for research purposes. We made available the data from more than 1,500 clinics. Data from clinical trials, particularly data reported in journals or submitted to the FDA, should be available to qualified researchers. POTUS could require it any day he is not visited by the Pope and has a pen available. May be if the US government shuts down, he could use the time to organize the government to be more efficient and effective. It is not a difficult task; government inefficiency and ineffectiveness is widespread and prevalent. The problem is critical in health care. The huge costs of drugs are interfering with optimal use of health resources and better research. Costs are huge, often from drugs used off-label or without adequate review by scientists other than those who participated in the clinical trial. There are no valid scientific or logistic or ethical reasons why data on clinical trials should be kept secret from qualified researchers.
In the US, POTUS can easily “ensure that the results of all clinical trials are made publicly available and that individual patient data are available for legitimate independent third party scrutiny.” (BMJ Editorial, http://www.bmj.com/content/351/bmj.h4973?etoc=). Ask POTUS to do it. He does not need an Act of Congress or a regulation or help from the Pope. If he does not know how, he can ask me (the NSA has my number). We can get it done in a few days. There are probably similar ways in the UK and Europe. Ex: don’t make payments until data are disclosed.
References
https://simple.wikipedia.org/wiki/United_States_Secretary_of_Health,_Edu...
https://www.whitehouse.gov/sites/default/files/omb/assets/inforeg/pra-fa.... The Paperwork Reduction Act (PRA) requires Federal Agencies to submit proposed collections of information for review and approval by OMB.
http://ocio.os.doc.gov/ITPolicyandPrograms/Information_Collection/dev01_...
http://babel.hathitrust.org/cgi/pt?id=pur1.32754081175618;view=1up;seq=1
http://link.springer.com/chapter/10.1007/978-94-011-7715-3_2
Competing interests: I conduct many activities that influence my opinions (e.g., my research, write books or articles or patents, have websites, design foods, lecture about health and disease, health policies). I describe the types of fats to eat, exercise, stress management, diets, eating less, periodic fasting. I am developing a new theory of disease, including lipid metabolism and the cause of cardiovascular disease. I intend to profit from my intellectual property (IP) (e.g, write articles, web sites, books, patents, lectures, ads, biomarker tests, etc.). I wrote a patent to measure fatty acids. I wrote one book on essential and trans fatty acids. I gave talks to corporations, conferences and trade shows, for which I was paid. I may receive compensation from food or other companies. This comment is an extract of books and articles either written or in progress (authorized by the author under the non-exclusive common license CC BY-NC 4.0 requested by BMJ).