| Sign In to gain access to subscriptions and/or personal tools. |
Improving the Clarity and Transparency of Reporting Health Research: a Shared Obligation and Responsibility
1 Unit of Periodontology, International Centre for Evidence-Based Oral Health, UCL Eastman Dental Institute, 256 Gray’s Inn Road, London WC1X 8LD, UK; Correspondence: * corresponding author, i.needleman{at}eastman.ucl.ac.uk
Key Words: clinical research • ethics • CONSORT • moral obligations INTRODUCTION Progress in healthcare depends on the application of innovation from many types of biomedical research. While appropriate design, conduct, and analysis are key characteristics of quality, a critical aspect of translating findings to healthcare is clear and transparent reporting. In this paper, we hope to stimulate debate on research reporting and in particular to ask the question, whose responsibility is this? WHY IS CLEAR AND TRANSPARENT REPORTING IMPORTANT? For research to influence healthcare, its publication needs to be critically evaluated. If reporting is inadequate—namely, information is missing, incomplete, or ambiguous—assumptions have to be made, and, as a result, important findings could be missed and not acted upon. Alternatively, false outcomes might be identified and used in practice. THE TOTALITY OF EVIDENCE For many reasons, not least the play of chance, the findings from an isolated study might, or might not, represent an important discovery. More convincing are similar findings from several independent studies (Clarke and Chalmers, 1998), particularly if synthesized in high-quality systematic reviews (Egger and Davey Smith, 1997). Studies with missing, incomplete, or ambiguous information do not contribute meaningfully, despite containing potentially important information (Moles et al., 2005). PROTECTION FROM BIAS The reporting quality of clinical research, especially for bias protection, is not yet ideal, with little difference between medicine and dentistry (Sjogren et al., 2002; Kunz et al., 2007; Pildal et al., 2007; Needleman et al., 2008). This is worrying, since inadequate protection from bias has been associated with an exaggeration of treatment effect (Kunz et al., 2007; Pildal et al., 2007; Wood et al., 2008). Clearly, though, adequate reporting of other dimensions relating to participant characteristics, their health status, and the intervention(s) being evaluated is also essential. ETHICS The Declaration of Helsinki (1964) and Good Clinical Practice (1996) provide clear guidance for the design and conduct of trials. Surprisingly, there has been little guidance about the adequacy of publication. The Declaration of Helsinki does stress accurate and full publication of results, while the ICH E8 (1997) states that clinical trials ‘should be reported appropriately’. Full and transparent publication should be considered within informed consent. If a participant consents on the understanding that participation could contribute to better future healthcare for others, then such developments are likely to be subverted by inadequate publication. Poor reporting should therefore be considered a breach of the ethical framework of biomedical research conduct. Current guidance (International Committee of Medical Journal Editors [ICMJE], 2007) stresses the importance of the use of reporting guidelines, although not as an ethical responsibility. In view of the importance of adequate reporting, we believe that reporting should be considered an ethical and moral responsibility of researchers, editors, publishers, and funders, and should be managed accordingly (Moher, 2007). WHAT REPORTING STANDARDS EXIST? The most recognized guideline for reporting is the CONSORT (Consolidated Standards of Reporting of Trials) Statement, and a new initiative, the EQUATOR Network (www.equator-network.org), provides a central and updated resource with links to the best available guidance for reporting experimental, observational, and other research types. SHARED RESPONSIBILITIES
Researchers
Editors and Publishers
Others Funders spend billions of dollars to fund what they believe to be research of the highest possible quality, with minimal bias but little attention to how that research is reported. Funders should become more active partners in ensuring that the research they fund is reported to the highest possible quality (O’Toole, 1997). Global organizations, such as the International Association for Dental Research (IADR), could lead educational activity to promote the use and importance of reporting guidelines, perhaps in partnership with funders, editors, and publishers. The regional network might also help to promote such reporting guidance locally. Oral health researchers should become involved in the development of reporting guidelines of particular importance to oral health researchers—for instance, the reporting of split-mouth studies (Lesaffre et al., 2007). CONCLUSIONS Improved reporting quality would be a major advance in biomedical research, offering the potential to accelerate advances in healthcare. Compared with the actual development and conduct of research studies, improving reporting quality is relatively straightforward and has modest requirements for resources. In addition, change could be brought about rapidly. A concerted and shared approach by stakeholders would have a profound impact on the quality of publications, maximizing benefits to consumers and other stakeholders reading research reports. This should be a shared obligation and responsibility. ACKNOWLEDGMENTS We are grateful to the Editors of the following journals for information on the adoption of CONSORT: Journal of Dental Research, Journal of Endodontics, Clinical Oral Implants Research, Journal of Periodontal Research, Journal of Clinical Periodontology, Caries Research, Oral Oncology, Community Dentistry and Oral Epidemiology, European Journal of Oral Sciences, International Journal of Oral & Maxillofacial Surgery, Journal of Periodontology, Journal of Dentistry, and the International Endodontics Journal. REFERENCES
Journal of Dental Research, Vol. 87, No. 10,
894-895 (2008)
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
