Project

Back to overview

Selective reporting of clinical trials? Multicentre study of protocols and published articles

English title Selective reporting of clinical trials? Multicentre study of protocols and published articles
Applicant Egger Matthias
Number 114142
Funding scheme Project funding (Div. I-III)
Research institution Institut für Sozial- und Präventivmedizin Universität Bern
Institution of higher education University of Berne - BE
Main discipline Methods of Epidemiology and Preventive Medicine
Start/End 01.06.2007 - 30.11.2008
Approved amount 167'475.00
Show all

Keywords (6)

Dissemination of research; selective reporting; publication bias; outcome reporting bias; research ethics; scientific integrity

Lay Summary (English)

Lead
Lay summary
Doctors, patients and health policy makers need to be able to appraise all the available evidence about medical interventions to make appropriate decisions. Systematic reviews of relevant clinical studies are considered the highest level of evidence. However, empirical studies suggest that not all studies are published. For instance, clinical trials with a negative effect of a new treatment or with inconclusive results are more likely to remain unpublished than confirmative studies (publication bias). Further, statistically significant outcomes within studies are being selectively reported (outcome reporting bias). Therefore, even systematic reviews of published studies may give an over-optimistic estimate of the effect of treatment. This can lead to health care recommendations that are at best inappropriate and at worst dangerous. The selection of clinical studies for publication has been demonstrated in Great-Britain, USA, Australia, and France. The selective reporting of study outcomes has been investigated in studies conducted in Denmark and Canada. This project will examine the publication of medical research conducted in Switzerland and Germany, and the biases affecting this process. The project is a collaboration with the German Cochrane Centre at the University of Freiburg / Germany. We aim to determine whether the full publication of medical research is influenced by characteristics of the studies such as trial design or commercial and non-commercial funding. For this purpose, we will perform a study of protocols of studies that compare the effectiveness or harm of at least one experimental drug with a control drug or placebo or no treatment by randomly allocating the study interventions to participants (randomised drug trials). We will include all protocols of such trials submitted to the research ethics committee of the Medical Faculty at the University of Berne from 1988 to 1998. Eletronic databases (e.g. Cochrane CENTRAL) will be searched for subesequent full publications that correspond to these protocols. We will estimate the publication rate of completed studies, examine key characteristics of study protocols and publications, and assess the discrepancy in the reporting of study outcomes. Factors associated with full publication, type of publication and time to publication will be analysed using appropriate statistical methods. The study has the potential to provide new empirical data on the dissemination of evidence from clinical studies conducted in research institutions in German-speaking countries.
Direct link to Lay Summary Last update: 21.02.2013

Responsible applicant and co-applicants

Employees

Publications

Publication
Cohort study of trials submitted to ethics committee identified discrepant reporting of outcomes in publications
Shelagh Redmond et al (2013), Cohort study of trials submitted to ethics committee identified discrepant reporting of outcomes in publications, in J Clin Epidemiol, 66, 1367-1375.
Publication and non-publication of clinical trials: longitudinal study of applications submitted to a research ethics committee
von Elm Erik, Roellin Alexandra, Bluemle Anette, Huwiler Karin, Witschi Mark, Egger Matthias (2008), Publication and non-publication of clinical trials: longitudinal study of applications submitted to a research ethics committee, in SWISS MEDICAL WEEKLY, 138(13-14), 197-203.

Associated projects

Number Title Start Funding scheme
133540 When clinical research fails: A study of controlled trials that were discontinued 01.02.2011 Project funding (Div. I-III)
149496 Learning from failure - Understanding the mechanisms of trial discontinuation 01.12.2013 Project funding (Div. I-III)

-