Author Affiliations: Departments of Psychiatry (Mss Roseman and Milette and Dr Thombs), Epidemiology, Biostatistics, and Occupational Health (Dr Thombs), and Medicine (Dr Thombs), McGill University, Montréal, Québec, Canada; Lady Davis Institute for Medical Research, Jewish General Hospital, Montréal, Québec, Canada (Mss Roseman and Milette and Dr Thombs); Department of Clinical Pharmacy, School of Pharmacy and Institute for Health Policy Studies, School of Medicine, University of California, San Francisco (Dr Bero); Department of Psychiatry, University of Pennsylvania School of Medicine, Philadelphia, and Health Psychology Section, Department of Health Sciences, University Medical Center Groningen, University of Groningen, the Netherlands (Dr Coyne); School of Health Policy and Management, York University; Emergency Department, University Health Network; and Department of Family and Community Medicine, University of Toronto, Toronto, Ontario, Canada (Dr Lexchin); and Department of Psychiatry, Oregon Health and Science University, and Portland Veterans Affairs Medical Center, Portland (Dr Turner).
Context Disclosure of conflicts of interest (COIs) from pharmaceutical industry study funding and author-industry financial relationships is sometimes recommended for randomized controlled trials (RCTs) published in biomedical journals. Authors of meta-analyses, however, are not required to report COIs disclosed in original reports of included RCTs.
Objective To investigate whether meta-analyses of pharmacological treatments published in high-impact biomedical journals report COIs disclosed in included RCTs.
Data Sources and Study Selection We selected the 3 most recent meta-analyses of patented pharmacological treatments published January 2009 through October 2009 in each general medicine journal with an impact factor of at least 10; in high-impact journals in each of the 5 specialty medicine areas with the greatest 2008 global therapeutic sales (oncology, cardiology, respiratory medicine, endocrinology, and gastroenterology); and in the Cochrane Database of Systematic Reviews.
Data Extraction Two investigators independently extracted data on disclosed study funding, author-industry financial ties, and author employment from each meta-analysis, from RCTs included in each meta-analysis, and on whether meta-analyses reported disclosed COIs of included RCTs.
Results Of 29 meta-analyses reviewed, which included 509 RCTs, only 2 meta-analyses (7%) reported RCT funding sources; and 0 reported RCT author-industry ties or employment by the pharmaceutical industry. Of 318 meta-analyzed RCTs that reported funding sources, 219 (69%) were industry funded; and 91 of 132 (69%) that reported author financial disclosures had 1 or more authors with pharmaceutical industry financial ties. In 7 of the 29 meta-analyses reviewed, 100% of included RCTs had at least 1 form of disclosed COI (pharmaceutical industry funding, author-industry financial ties, or employment), yet only 1 of these 7 meta-analyses reported RCT funding sources, and 0 reported RCT author-industry ties or employment.
Conclusion Among a group of meta-analyses of pharmacological treatments published in high-impact biomedical journals, information concerning primary study funding and author COIs for the included RCTs were only rarely reported.
Roseman M, Milette K, Bero LA, Coyne JC, Lexchin J, Turner EH, Thombs BD. Reporting of Conflicts of Interest in Meta-analyses of Trials of Pharmacological Treatments. JAMA. 2011;305(10):1008-1017. doi:10.1001/jama.2011.257