Article type
Year
Abstract
Background:
Cochrane systematic reviews are marketed as independent, reliable summaries of research evidence. Potential Cochrane authors are encouraged to work in areas that they are interested in, and may also be investigators on trials included in the review. However, it is possible that they will treat their own trials differently, with regard to the rigor of the application of the inclusion criteria, data extraction, and interpretation, resulting in a biased review.
Objectives:
To quantify the number of recent Cochrane systematic reviews that include authors that are also authors of included trials; and to describe any steps taken to mitigate the risk of conflict of interest influencing the results.
Methods:
Audit recent Cochrane systematic reviews from October 2013 with the intention of evaluating a total of 210 reviews.
Results:
The initial evaluation of 79 studies showed that 12 reviews did not include any studies; and four (6%) of the remaining 67 reviews had review authors who were also authors on one of the included trials. In two of the reviews, the conflict of interest was noted, with one stating that studies authored by a review author were appraised independently by the other review authors.
Conclusions:
Early results suggest conflicts of interest are being inadequately addressed in Cochrane systematic reviews, with a number of authors of Cochrane Reviews also being an author of a paper included in the review. Three of the four reviews where a conflict of interest was noted did not mitigate the risk of conflict of interest influencing the results. Full results and associated recommendations to assist teams manage this form of conflict of interest will be available by August 2014.
Cochrane systematic reviews are marketed as independent, reliable summaries of research evidence. Potential Cochrane authors are encouraged to work in areas that they are interested in, and may also be investigators on trials included in the review. However, it is possible that they will treat their own trials differently, with regard to the rigor of the application of the inclusion criteria, data extraction, and interpretation, resulting in a biased review.
Objectives:
To quantify the number of recent Cochrane systematic reviews that include authors that are also authors of included trials; and to describe any steps taken to mitigate the risk of conflict of interest influencing the results.
Methods:
Audit recent Cochrane systematic reviews from October 2013 with the intention of evaluating a total of 210 reviews.
Results:
The initial evaluation of 79 studies showed that 12 reviews did not include any studies; and four (6%) of the remaining 67 reviews had review authors who were also authors on one of the included trials. In two of the reviews, the conflict of interest was noted, with one stating that studies authored by a review author were appraised independently by the other review authors.
Conclusions:
Early results suggest conflicts of interest are being inadequately addressed in Cochrane systematic reviews, with a number of authors of Cochrane Reviews also being an author of a paper included in the review. Three of the four reviews where a conflict of interest was noted did not mitigate the risk of conflict of interest influencing the results. Full results and associated recommendations to assist teams manage this form of conflict of interest will be available by August 2014.