Pharmacological Meta-Analyses Rarely Report Disclosures
Meta-analyses of pharmacological interventions rarely disclose funding or conflicts of interest
TUESDAY, March 8 (HealthDay News) -- Meta-analyses of pharmacological treatments rarely include information addressing primary study funding and conflicts of interest (COIs) of the authors for the included randomized control trials (RCTs), according to a study published in the March 9 issue of the Journal of the American Medical Association.
Michelle Roseman, from McGill University in Montréal, and colleagues determined whether meta-analyses of pharmacological treatments in high-impact journals report COIs that are disclosed in the RCTs they include. Twenty-nine meta-analyses of patented pharmacological treatments published between January and October 2009 in five areas of medicine were reviewed to identify study funding disclosure, author financial relationships, and author employment for meta-analyses and their RCTs.
The researchers found that only two meta-analyses reported RCT funding sources, and none reported RCT author-industry relationships or pharmaceutical industry employment. The meta-analyses reviewed covered 509 RCTs, with 318 RCTs reporting funding sources, of which, 69 percent disclosed funding from industry. Of 132 RCTs that reported author financial disclosures, 69 percent had at least one author with a financial relationship to the pharmaceutical industry. In seven of the 29 meta-analyses, all of the included RCTs had at least one form of disclosed COI, but only one reported RCT funding sources, and none reported RCT author-industry ties or employment.
"This study found that meta-analyses of pharmacological interventions published in high-impact medical journals rarely reported the funding sources or author-industry financial ties of included RCTs, even when these sources of COIs were disclosed in RCT reports," the authors write.
One author disclosed a financial relationship with legal firms involved in cases dealing with the pharmaceutical industry.