Publication bias is the tendency to favour the publication of results when these are positive, particularly favouring one treatment rather than another or a placebo. This is a long-standing problem in scientific literature [1]. The first Open Science ‘health’ barometer (2022) indicates that only 57% of French clinical trials result in the publication of results in the clinical trial registries Clinicaltrials.gov and/or Clinical Trials Information System – CTIS. This percentage is only 31% when the trials have public sponsors. These indicators were confirmed by the results published in the new 2023 Barometer of Open Science [2].

In health research, this bias leads to decisions being taken on the basis of partial information and also contributes to ‘research waste” – the waste of funds allocated to research. This is despite a European regulation voted in 2014 [3] which came into force in 2022 and makes the summary publication of results compulsory.

The National Plan for Open Science [4] and the French Digital Health Acceleration Strategy also feature commitments to extending this obligation to make the results of health research transparent for non-drug clinical trials and observational studies as well as drug clinical trials. However efforts are required to enforce existing regulations before extending the regulations to new areas.

[1] Fanelli D. Negative results are disappearing from most disciplines and countries. Scientometrics. 2012;90(3):891-904. doi:10.1007/s11192-011-0494-7

[2] https://barometredelascienceouverte.esr.gouv.fr/

[3] Posting of clinical trial summary results in European Clinical Trials Database (EudraCT) to become mandatory for sponsors as 21 July 2014. European Medicines Agency. https://www.ema.europa.eu/en/news/posting-clinical-trial-summary-results-european-clinical-trials-database-eudract-become-mandatory

[4] https://www.enseignementsup-recherche.gouv.fr/fr/le-plan-national-pour-la-science-ouverte-2021-2024-vers-une-generalisation-de-la-science-ouverte-en-48525