Association between funding, risk of bias, and outcome of randomised controlled trials in oral and maxillofacial surgery

M.A.E.M. Oomens, S. Lazzari, M.W. Heymans, T. Forouzanfar

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

The influence of funding on the main outcome of a random control trial (RCT) is important, as it could potentially lead to bias towards industry, and results that are too optimistic. We investigated the association between funding, the published outcome, and the risk of bias in trials in oral and maxillofacial surgery (OMFS) published from January 2000 to May 2013 listed in PubMed. The methods used were scored using the risk of bias items given in a Delphi List. Sources of funding were recorded and categorised five ways: not funded, funded by industry, not funded by industry, supported by industry, and source of funds not clear. A total of 390 RCT met the inclusion criteria, and there was a correlation between funding and favourable main outcomes, although this was not significant. There was no correlation between the risk of bias and favourable results of the main outcome of a trial, or between the risk of bias and the reported source of funding in post-hoc analysis. We were unable to show a significant correlation between funding and a higher likelihood of a favourable result for the primary outcome in RCT in OMFS. We also failed to show a significant correlation between the risk of bias of a trial and its main outcome. In contrast, the source of funding proved to affect the risk of bias of a trial significantly, although not in post-hoc analysis. Funded trials were better organised, and so had a lower risk of bias.
Original languageEnglish
Pages (from-to)46-50
JournalBritish Journal of Oral & Maxillofacial Surgery
Volume54
Issue number1
DOIs
Publication statusPublished - 2016

Fingerprint

Dive into the research topics of 'Association between funding, risk of bias, and outcome of randomised controlled trials in oral and maxillofacial surgery'. Together they form a unique fingerprint.

Cite this