Fate of clinical research studies after ethical approval--follow-up of study protocols until publication.

Détails

Ressource 1Télécharger: BIB_C5D9F1A2551D.P001.pdf (824.08 [Ko])
Etat: Serval
Version: de l'auteur
ID Serval
serval:BIB_C5D9F1A2551D
Type
Article: article d'un périodique ou d'un magazine.
Collection
Publications
Titre
Fate of clinical research studies after ethical approval--follow-up of study protocols until publication.
Périodique
PloS one
Auteur(s)
Blümle A., Meerpohl J.J., Schumacher M., von Elm E.
ISSN
1932-6203 (Electronic)
ISSN-L
1932-6203
Statut éditorial
Publié
Date de publication
2014
Peer-reviewed
Oui
Volume
9
Numéro
2
Pages
e87184
Langue
anglais
Notes
Publication types: Journal Article ; Research Support, Non-U.S. Gov't
Publication Status: epublish
Résumé
Many clinical studies are ultimately not fully published in peer-reviewed journals. Underreporting of clinical research is wasteful and can result in biased estimates of treatment effect or harm, leading to recommendations that are inappropriate or even dangerous.
We assembled a cohort of clinical studies approved 2000-2002 by the Research Ethics Committee of the University of Freiburg, Germany. Published full articles were searched in electronic databases and investigators contacted. Data on study characteristics were extracted from protocols and corresponding publications. We characterized the cohort, quantified its publication outcome and compared protocols and publications for selected aspects.
Of 917 approved studies, 807 were started and 110 were not, either locally or as a whole. Of the started studies, 576 (71%) were completed according to protocol, 128 (16%) discontinued and 42 (5%) are still ongoing; for 61 (8%) there was no information about their course. We identified 782 full publications corresponding to 419 of the 807 initiated studies; the publication proportion was 52% (95% CI: 0.48-0.55). Study design was not significantly associated with subsequent publication. Multicentre status, international collaboration, large sample size and commercial or non-commercial funding were positively associated with subsequent publication. Commercial funding was mentioned in 203 (48%) protocols and in 205 (49%) of the publications. In most published studies (339; 81%) this information corresponded between protocol and publication. Most studies were published in English (367; 88%); some in German (25; 6%) or both languages (27; 6%). The local investigators were listed as (co-)authors in the publications corresponding to 259 (62%) studies.
Half of the clinical research conducted at a large German university medical centre remains unpublished; future research is built on an incomplete database. Research resources are likely wasted as neither health care professionals nor patients nor policy makers can use the results when making decisions.

Mots-clé
Biomedical Research/ethics, Humans, Publishing/ethics
Pubmed
Web of science
Création de la notice
11/04/2014 18:45
Dernière modification de la notice
03/03/2018 21:16
Données d'usage