Self-Reported Financial Conflicts Have No Effect on Cardiovascular Trial Outcomes
A new analysis of major cardiovascular trials published over an 8-year period in 3 top-tier journals suggests that conflicts of interest may not matter as much as has been previously thought. In particular, no association was found between researchers’ self-declared financial relationships and the likelihood of a trial reporting favorable results, according to findings released online February 6, 2013, ahead of print in the Journal of the American College of Cardiology.
Michael E. Farkouh, MD, MSc, of Mount Sinai School of Medicine (New York, NY), and colleagues identified 550 papers on cardiovascular disease published between January 1, 2000, and April 15, 2008, in the New England Journal of Medicine, Lancet, and Journal of the American Medical Association.
Researchers reported financial conflicts with industry such as holding stock, being an employee, serving on a speaker’s bureau, and consulting in 281 papers (51.1%). The remaining 269 papers either reported no conflict (7.1%) or mentioned research funding only (41.8%).
Level of Conflicts Differs But Not Outcome
Papers listing self-reported conflicts were no more likely to have favorable results than those without conflicts (60.5% vs. 59.5%; P = 0.81). The lack of effect remained after multivariable adjustment for study type, center location, trial registration, funding source, and intervention type. Moreover, the distribution of favorable vs. unfavorable results was consistent across various types of conflicts.
Trials having a surrogate endpoint, meanwhile, showed a borderline tendency toward being positive compared with those having a clinical endpoint (68.7% vs. 58.1%; P = 0.051). Studies that included independent statistical analyses were less likely to produce favorable outcomes than those without such independent assessments (56.7% vs. 69%; P = 0.008).
Physician-reported conflicts increased alongside the amount of industry funding, ranging from 21.5% in studies receiving no such sponsorship to 50.0% in those receiving some funding and 75.0% in those entirely backed by industry (P < 0.0001 for both). Device trials were just as likely to include conflicts as drug studies (68.4% vs. 60.8%; P = 0.57) and trials testing both drugs and devices (43.8%; P = 0.14) but were more likely to include conflicts than those not involving drugs or devices (26.9%; P < 0.0001). Conflicts also were more likely when trials tested equivalence/noninferiority vs. superiority (61.4% vs. 47.4%; P = 0.006) and clinical vs. surrogate endpoints (53.9% vs. 38.4%; P = 0.005).
Importantly, investigators who listed their trials on the Web site ClinicalTrials.gov, a process that became mandatory in late 2007, also were more likely to report conflicts than those who did not register (72.3% vs. 27.7%; P = 0.006).
Researchers reported financial conflicts with industry such as holding stock, being an employee, serving on a speaker’s bureau, and consulting in 281 papers (51.1%). The remaining 269 papers either contained no conflict (7.1%) or mentioned research funding only (41.8%).
Findings Reassuring and Unexpected
“Our data demonstrate that authors’ self-declared [financial conflicts of interest] are ubiquitous in major cardiovascular clinical trials and do not seem to have an impact on their outcomes,” Dr. Farkouh and colleagues conclude.
Study coauthor Ashish Aneja, MD, of Case Western Reserve University (Cleveland, OH), told TCTMD in a telephone interview, “These findings are as unexpected to us as they would be to any person who is practicing out there. They’re reassuring to some extent but we can’t get complacent about the issue.”
However, due to the nature of self-reporting, the observed amount of conflicts “may represent the tip of the iceberg,” the paper cautions. “The potential impact of publication bias also is difficult to assess and overcome and could have impacted the results of our study. . . . Additionally, we limited our study to the top 3 general medical journals for practical reasons. The potential impact of casting an even wider net and including articles from other cardiovascular journals is unknown.”
Higher impact journals were chosen, Dr. Aneja said, “specifically because [they] have the potential to change practice.”
In the same telephone interview, Dr. Farkouh commented, “We think it’s a limited snapshot but one that informs this argument, which has always been that conflicts do influence outcomes. [The current study] swings the argument back—not necessarily to say there is no influence—but rather to a more neutral position. We need to continue to monitor this and be aware of it.”
Hard to Account for All Conflicts
Robert M. Califf, MD, of Duke University Medical Center (Durham, NC), however, expresses more reservations in an accompanying editorial.
“Such journals are hardly representative of the general medical literature, because they are highly selective and typically publish findings from large, adequately powered trials that tend to be clinically interesting regardless of whether the results are positive or negative,” he comments, pointing out that the study contained no evaluation of whether disclosures were accurate.
Dr. Califf also highlights nonfinancial conflicts of interest. “Investigators who hold strong beliefs or who are engaged in competing work often can affect the interpretation of a study’s results or the editorial perspective taken by a journal,” he writes. “These types of conflicts are well described, but they are not evaluated as often because they are less overt and are harder to quantify than [financial relationships].”
Dr. Aneja agreed that there are less tangible influences, often difficult to assess. “If you work at a certain institution, you are sort of beholden to [it]. That itself, I think, is a conflict of interest. The institution may want you to practice medicine in a certain way [and] use a certain company’s devices or medications,” he said.
Though many of the issues raised by the editorial are valid, Dr. Aneja countered that it comes down to readily available information. Physicians may, for example, not be able to recall conflicts from many years ago, prior to the era of mandatory reporting, he noted.
Moving Toward Consistency
While the process of self-reporting financial conflicts “alone probably does not eliminate potential or perceived problems stemming from [such relationships],” Dr. Farkouh et al note in the paper, it “represents an important first step in eradicating their impact.”
Current reporting strategies are in flux, they observe, adding that, “An effective but ongoing surveillance and standardization of procedures at the individual and institutional level, as suggested by the Institute of Medicine, is required to address this phenomenon.”
Dr. Aneja noted that the International Committee of Medical Journal Editors, “which is the body that controls the way journals publish their data, has issued a set of guidelines that are in the process of being implemented. . . . All the prominent journals have certainly embraced those guidelines and they’ve become fairly stringent about the way these conflicts are reported.”
Overall, 34.6% of trials were funded by nonprofit organizations, 48.3% by industry, and 17.1% by a combination of the two. Surrogate endpoints were assessed by 18% of trials and clinical endpoints by 82%. The majority of trials looked at drugs (58.9%), with the remainder investigating devices (9.8%), both drugs and devices (2.9%), or other types of intervention such as behavioral, dietary, procedural, or surgical (28.4%). Most were multicenter (89.1%), involved a superiority hypothesis (72.3%), and took place in the United States and/or Canada (65.8%).
1. Aneja A, Esquitin R, Shah K, et al. Authors’ self-declared financial conflicts of interest do not impact the results of major cardiovascular trials. J Am Coll Cardiol. 2013;Epub ahead of print.
2. Califf RM. Conflicting information about conflict of interest. J Am Coll Cardiol. 2013;Epub ahead of print.
- Drs. Farkouh and Aneja report no relevant conflicts of interest.
- Dr. Califf reports receiving research grants that are paid directly to Duke University and partially support his salary as well as consulting fees from numerous drug, device, and media companies that are donated to nonprofit organizations. He also holds equity in Nitrox, N30 Pharma, and Portola.