Meta-analyses Rarely Report Conflicts of Interest in Underlying Trials

Few published meta-analyses of trials involving pharmacological therapies disclose the funding sources or author conflicts of interest for the primary studies, researchers report in the March 9, 2011, issue of the Journal of the American Medical Association.

Although several guidelines exist regarding reporting of conflicts of interest for study authors publishing in medical journals, no guidelines exist for such reporting in meta-analyses, the authors say.

For the study, researchers led by Brett D. Thombs, PhD, of McGill University (Montreal, Canada), examined a sample of meta-analyses of pharmacological treatments published from January through October 2009 in 3 categories of high-impact journals: general medicine journals, journals representing the top 5 medical specialties based on 2008 global pharmaceutical sales, and the Cochrane Database of Systematic Reviews. All of the meta-analyses investigated at least 1 drug patented in the United States and used in oncology, respiratory medicine, endocrinology, or gastroenterology.

Conflicts of Interest Not Carried Forward

Separately, of the 509 randomized controlled trials included in the 29 meta-analyses selected, 62.5% reported funding sources. Nearly 69% were funded in part or fully by the pharmaceutical industry.

Author financial disclosures were reported in 25.9% of the individual trials, and among these, 68.9% reported 1 or more authors with financial ties to the pharmaceutical industry.

However, only 2 of the 29 meta-analyses reported the funding source of included trials—1 in a table footnote and the other in a table following the references. One of the meta-analyses reported no author ties to industry, while the other reported that 1 of 3 authors had a link to the pharmaceutical industry.

In 7 of the 29 meta-analyses, all of the included trials had at least 1 form of disclosed conflict of interest, yet only 1 of these reported funding sources and none reported author-industry ties or industry employment for trial investigators.

The vast majority of the meta-analyses assessed quality or risk of bias in the included trials, but few did so in a way that revealed the potential influence of industry relationships. Only 2 meta-analyses recorded and reported trial funding sources, while 5 recorded but did not report funding sources, and 20 did not record funding sources at all. Twenty-seven of 29 meta-analysis authors did provide information on data extraction protocols.

Authors Not Assessing Biases

According to Dr. Thombs and colleagues, their results highlight a major gap in the reporting of conflicts of interest and suggest that, unless medical journals institute a formal reporting policy, conflicts will continue to go unreported when data are synthesized in meta-analyses.

In addition, they say the authors of meta-analyses are expected to “transparently assess and interpret potential sources of bias from included studies that could influence outcomes.” When the authors fail to do so, readers are likely to trust the conclusions of a given meta-analysis when they potentially should not.

“Meta-analysts should evaluate the potential for bias due to pharmaceutical industry study funding and author-industry financial ties as part of their standard risk of bias assessment,” the investigators write. “As with other potential sources of bias, results from this domain should be transparently documented and used in sensitivity analyses or qualitatively.”

 

---

Source:
Roseman M, Milette K, Bero LA, et al. Reporting of conflicts of interest in meta-analyses of trials of pharmacological treatments. JAMA. 2011;305:1008-1017.

 

 

Related Stories:

• Physician-Industry Relationships Decreasing but Still Prevalent
• Patients, Research Participants Prefer Physician Disclosure of Financial Ties