Declaration of transparency for each research article
BMJ 2013; 347 doi: https://doi.org/10.1136/bmj.f4796 (Published 07 August 2013) Cite this as: BMJ 2013;347:f4796
All rapid responses
I congratulate Altman and Moher on an excellent suggestion. The "transparency declaration" would indeed be incredibly easy for journals to implement. While it does not of course provide any guarantee that it will make authors report trials more fully, it would shift many deficiencies of reporting from the category of "inadvertent error" to "frank dishonesty". I suspect that many authors who might be willing to be careless in their reporting would be far less comfortable with being actively dishonest.
However, I do have to take issue with one small part of Altman and Moher's article. They state "There are concerns that commercially sponsored research may be more likely to remain unpublished", which they back up with 2 citations.
The first of those citations, their reference No 2, looks at publication bias, but is a small study on a single drug, and does not compare commercially sponsored research with any kind of control group, so is unable to support the assertion that commercially sponsored research is more likely to be unpublished.
The second of those citations, their reference No 31, compared industry and non-industry publications on many outcomes, but likelihood of remaining unpublished was not one of them, so it is equally incapable of supporting their statement. (As an aside, there are also some serious flaws in that paper, as I have written about previously.)
So what does the research tell us? A systematic review published in 2010 found mixed evidence on whether industry or non-industry studies were more likely to be published, but with the majority of evidence showing that industry studies were more likely to be published. A large and well conducted study published since that systematic review also found that, although industry studies were slower to be published, they were more likely to be published eventually. (It also found that industry studies were less likely to have discordant primary outcomes between registration and publication.)
Incomplete or misleading publications are indeed a serious problem, and I welcome Altman and Moher's suggestion for one step towards tackling it. However, it is important to recognise that poor publication is a general problem of clinical research, and not one specifically limited to commercial sponsors.
Competing interests: My company provides clinical trial reporting services to pharmaceutical companies and academic researchers, but more often to the former.
Sir, the editorial by Altman and Moher should shake the conscience of each and every researcher who has pledged the Hippocratic oath. No words can emphasize the need for transparency in reporting of results. As put in by the authors, not reporting the results of research, or withholding negative results should, in fact, be labelled as 'bad clinical practice' (the antithesis of GCP).
The funding agencies ought to ensure that every project funded by them is published in time and reported back to the agency. A glimpse of the dismal state of affairs in India regarding reporting of publications to the national agency Indian Council of Medical Research is provided below. Indeed, it is time for responsible and transparent research, not research for the sake of research. After all research entails drain on the country's economic resources. drjagjitsingh1@gmail.com
Competing interests: No competing interests
Re: Declaration of transparency for each research article
I read with great interest the article by Altman and Moher (1), who suggest that authors should sign a publication transparency declaration as a measure to help ensure that the scientific community and the public get accurate and complete record of research.
Indeed, the Singapore Statement on Research Integrity, the first international guideline set to foster greater integrity in research worldwide (http://www.singaporestatement.org/), emphasizes that researchers should take responsibility for the trustworthiness of their research and that the primary responsibility has to rest with the individual, nonetheless it must be emphasized that also research istitutions are expected to adopt proper measures to ensure the accountability of research activity performed in their context.
Scientific misconduct includes not only serious breaching of integrity, like fabrication, falsification, and plagiarism, but also questionable research practices such as failure to keep records, mismanagement of data, proper acknowledgements, ghost/guest authorship, etc. Prevention of scientific misconduct is of paramount importance and should start from proper training into research methodology and ethics. Research integrity issues should be part of the basic training of PhD students and postdoc researchers, who need to have clear in mind that avoiding questionable research practice (by e.g. knowing and adhering to regulations and policies related to research, employ appropriate methods during research and in the critical analysis, report and interpretation of results, as well as keep accurate research records) will result in higher quality of research results and increased potential for scientific and personal achievements in the medium-long term. Such a training should be the responsibility of senior researchers, however in the context of appropriate institutional policies and strategies.
Unfortunately, when considering responsibilities of research institutions, emphasis is still put preferentially on procedures for responding to allegations of misconduct. However investigations about individuals under suspicion, in particular in the case of senior researchers, may bring with them significant conflicts of interest which may ultimately compromise the investigation (2). Yet, even from a pragmatic and selfish point of view, research institutions are likely to suffer significant damage from individual researchers' misconduct in terms of reputational damage, possibly leading to impaired attractiveness towards potential funders as well scientific research partners.
According to the Singapore Statement on Research Integrity (#13. Research Environments), "Research institutions should create and sustain environments that encourage integrity through education, clear policies, and reasonable standards for advancement, while fostering work environments that support research integrity." Institutional policies and standards should also consider scientific colleagues who are "bystanders", and who may be involved either because they may suspect scientific misconduct and would like to report possible concerns or just because they may indirectly suffer from cases of misconduct e.g. belonging to the same department.
In summary, research institutions have a key role in ensuring research integrity, therefore they need not only to be ready to respond to allegations against researchers, but also (and mainly) to prevent misconduct, ensuring that research they perform has a high degree of integrity and accountability. This can be achieved through the development of internal policies and procedures and in particular through the supply of continuous tranining into research integrity issues, targeting primarily young researchers. Individual departments need to be empowered as they (should) represent homogeneous communities of researchers. A recent study aimed at investigating the regulatory framework regarding research integrity in Europe (3). According to their results, specific laws to deal with research misconduct exist only in Denmark and Norway, many countries have multiple guidelines with little internal consensus and with different lists of principles and definitions, and no guidelines could be retrieved for 12 countries. It seems thus that much work remains to be done. In any case, although main responsibility must remain on individual researchers, research institutions need to take fully their responsibilities, as their role is crucial in the process towards ensuring research integrity.
References
1. Altman DG, Moher D. Declaration of transparency for each research article. BMJ 2013;347:f4796
2. Responsibilities of research institutions, in Scientific misconduct, Wikipedia, the free encyclopedia URL: http://en.wikipedia.org/wiki/Scientific_misconduct#Responsibilities_of_r... (last accessed: August 10th, 2013)
3. Godecharle S, Nemery B, Dierickx K. Guidance on research integrity: no union in Europe. The Lancet 2013;381:1097-8
Marco Cosentino, MD PhD
Center for Research in Medical Pharmacology
University of Insubria, Via Ottorino Rossi n. 9
21100 Varese VA - Italy
E-mail: marco.cosentino@uninsubria.it
Mario Picozzi, MD
Dipartimento di Biotecnologie e Scienze della Vita (DBSV)
University of Insubria, Via Ottorino Rossi n. 9
21100 Varese VA - Italy
E-mail: mario.picozzi@uninsubria.it
Competing interests: No competing interests