Conflict of Interest Reporting in Dentistry Randomized Controlled Trials: A Systematic Review

• Conflict of interest (COI) reporting in dentistry was examined using 66 studies
• 75% of the 66 studies mentioned the presence or absence of potential COI
• There was no correlation amid direction of results and whether COI was declared
• There was no association between funding source and whether COI was discussed
• In what journal the article was published correlated with whether COI was mentioned

ObjectivesConflict of interest (COI) reporting in the medical field has been a concern for a number of years. As the impact of industry on medical research increases, the need for transparency in these relationships increases. In this present study we aim to assess current practices and associations for COI reporting in the field of dentistry.Study designWe conducted a systematic review of conflict of interest reporting for randomized controlled trials (RCTs) in 6 journals of Dentistry published between Jan 2011 and Mar 2012. We extracted study characteristics in duplicate and used descriptive statistics and Chi-square tests to assess associations.ResultsOf 1755 studies across seven journals, we identified 66 eligible RCTs. Many included studies were conducted in Europe (39%), with an average sample size of 355. A total of 76% of studies mentioned the presence or absence of potential COI. No correlation between the direction of results and whether a COI was declared was found (p = 0.328), nor was there an association between funding source and whether COI was discussed (p = 0.120). The journal in which the article was published did however have a correlation as to whether COI was discussed (p ≤ 0.0001).ConclusionsRCTs published in the field of dental research inconsistently report the presence or absence of a conflict of interest. Dental journals should standardize conflict of interest reporting to aid dental researchers in understanding when a conflict of interest exists, and to provide transparency to readers and patients alike.