Selective reporting of biomedical research - a longitudinal study of protocols submitted to a research ethics committee and published articles
Final Report Abstract
In summary, our two projects “Selective reporting of biomedical research – a longitudinal study of protocols submitted to a research ethics committee and published articles” and “Empirical study of controlled clinical trials that were discontinued” enabled us to better understand the mechanisms of reporting and dissemination of results of clinical research conducted at a German academic research centre and elsewhere. By accessing study protocols archived by a research ethics committee we utilised a novel and unbiased source of information on clinical research projects. As compared to samples using studies that were registered prospectively and received competitive funding (Jones BMJ 2013 and Gordon NEJM 2013) our study results gain in external validity as they better reflect the landscape of clinical research in Germany, also including smaller studies and those without extramural funding. While archived study protocols have proven very useful as a resource for methodological research, the extraction of data from this heterogeneous body of documents turned out to be difficult to standardize and consequently time-consuming. Both our projects have contributed valuable evidence to the growing knowledge of how scientific data is selectively disseminated between the approval of study protocols, study conduct and completion and subsequent publication in the scientific literature. The results are consistent with the findings of recently published international research. Up-to-date evidence is urgently needed in order to support current initiatives in science policy e.g. to achieve public access to study data from human research (AllTrials Initiative) and to overcome the failure to publish negative findings (OPEN project funded within the 7th EU Framework Programme and led by the German Cochrane Centre, Freiburg / Germany). Using our comprehensive dataset we were able to address several topical questions, which led and will lead to highly visible publications in international journals such as the BMJ. For instance, we showed that eligibility criteria are not reported as pre-specified in study protocols, and that, in general, a substantial share of clinical studies of all types remain unpublished. With the second project the established cohort of trials became one of the foundation blocks of the DISCO study, an empirical research project on trial discontinuation conducted in three countries. So far, this project has produced robust estimates of the frequency of this problem, which were presented and discussed at several key conferences. It will enable the involved researchers to pursue their methodological research with the aim to better understand why trials are discontinued and to develop effective measures to prevent this particular type of waste. The projects’ outcomes have been instrumental in securing new funding that will help to continue this line of research in international collaboration.
Publications
- Clinical research projects at a German medical faculty: follow-up from ethical approval to publication and citation by others. 2008. Journal of Medical Ethics, Vol. 34. 2008, Issue 9, e20.
Blümle A, Antes G, Schumacher M, Just H, von Elm E.
(See online at https://dx.doi.org/10.1136/jme.2008.024521) - Reporting of eligibility criteria of randomised trials: cohort study comparing trial protocols with subsequent articles. . BMJ, Vol. 342.2011: d1828.
Blümle A, Meerpohl JJ, Rücker G, Antes G, Schumacher M, von Elm E, et al.
(See online at https://dx.doi.org/10.1136/bmj.d1828) - Learning from failure - rationale and design for a study about discontinuation of randomized trials (DISCO study)BMC Medical Research Methodology , Vol.12. 2012. p. 131.
Kasenda B, von Elm E, You J, et al.
(See online at https://doi.org/10.1186/1471-2288-12-131) - Fate of clinical research projects after ethical approval – follow-up and comparison of protocols with subsequent publications. PLOS ONE, Vol.9.2014, Issue 2: e87184.
Blümle A, Meerpohl J, Schumacher M, von Elm E.
(See online at https://doi.org/10.1371/journal.pone.0087184)
