Selective reporting of biomedical research - a longitudinal study of protocols submitted to a research ethics committee and published articles

• Applicant: Dr. Erik von Elm
• Subject Area: Epidemiology and Medical Biometry/Statistics
• Term: from 2008 to 2013
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 65205989

Patients and doctors should be able to appraise all the available evidence about medical interventions to make appropriate decisions. Empirical studies suggest that not all research results are published. For instance, clinical trials with a negative effect of a new treatment are more likely to remain unpublished than confirmative studies (positive outcome bias). Further, outcomes of studies may be selectively reported (outcome reporting bias) and cited in subsequent studies (citation bias). If only part of the existing evidence can be used later, health care recommendations can be biased and lead to inappropriate decisions.This project conducted at two universities in Switzerland and Germany will examine the chain of dissemination of medical research results from the planning stage of a study protocol until full publication and subsequent citation by others. We aim to determine whether the publication and citation of research is associated with study characteristics, e.g. trial design or funding. We will access the protocols of clinical studies of any design submitted to the research ethics committee of both universities (Freiburg: years 1998 to 2000). Electronic literature databases will be searched for corresponding full publications; searches will be complemented by questionnaires mailed to the investigators. We will investigate the publication probability of completed studies, key characteristics of study protocols and publications, and the discrepancy in the outcome reporting between protocols and publications. To estimate different types of reporting bias, factors associated with study completion, full publication and later citation will be analysed statistically. The study will provide new insights into the chain of dissemination of evidence from clinical studies conducted in German-speaking research institutions.

• DFG Programme: Research Grants