For-Profit Research More Favorable in Drug Trials

Friday, August 22nd 2003

• Tweet

Conclusions in randomized drug trials funded by for-profit organizations may be more positive than those funded by nonprofit organizations because of biased interpretations of trial results, a new study finds. Bias may be introduced by subtle factors, such as an emphasis on results from surrogate outcomes or subgroups, or by more explicit ones, such as a financial conflict of interest, according to the authors.

Regardless, the recurring evidence that suggests a link between industry-backed research and positive conclusions of drug trials argues for establishing an international register, available to the public, of all randomized clinical trials, according to the study published in the Aug. 20 issue of The Journal of the American Medical Association. The register would "enable the public to follow the development of drugs from the beginning of Phase 2 trials," said the authors, and would make available "data from unpublished randomized trials irrespective of funding source or results."

The authors, from the Center for Clinical Intervention Research at Denmark's Copenhagen University Hospital, conclude, "Editors, peer reviewers, and readers of trial reports should evaluate carefully the trial data to determine if the reported conclusions are supported by the data."

Officials at Pharmaceutical Research and Manufacturers of America (PhRMA), the industry group that represents drugmakers, had not seen the study and were not able to comment on its findings.

Based on the premise that industry-sponsored trials tend to draw conclusions in support of their drugs in development, the researchers sought to examine whether the link between funding and the conclusions derived from drug trials reflected the extent of the treatment effect (the difference in outcomes between groups) or the occurrence of adverse events. Adverse events were classified into four categories: no significant difference between study and control groups, significantly more frequent in the control group, significantly more frequent in the study group, or not reported.

Using a random sample of 167 reviews selected from the May 2001 Cochrane Library, the researchers identified 25 reviews of 370 drug trials that contained eligible meta-analyses, or models that contained the assessment of a binary outcome and noted the influence of a selection bias. The primary binary outcome, or research questions with only two possible outcomes, was considered the primary outcome for all of the trials included in each meta-analysis. Researchers then sought to measure whether the study drug was recommended as the treatment of choice.

In more than half (51%) of the trials funded by for-profit organizations, the study drug was recommended as the treatment of choice compared with 16% of the trials funded by nonprofit organizations, 30% of trials that did not report funding, and 35% of those funded by both nonprofit and for-profit organizations, the study found. In addition, adjusted analyses did not show an influence of treatment effect or adverse events on the association, according to the study. "The present study adds to previous evidence showing that this association does not reflect the quantitative trial results; neither the magnitude of the treatment effect nor the occurrence of adverse events could explain the association," the authors wrote.

The authors raise several possibilities for the association between funding source and conclusion, including violation of the so-called "uncertainty principle," publication bias, emphasis on subgroup or secondary analyses, or bias in drawing conclusions. "After having conducted exploratory randomized trials in Phase 2 drug development, several confirmatory randomized Phase 3 (proof of concept) trials usually are launched," the authors note. "Such trials may have a higher likelihood of favoring the experimental drug."

The design of the study by the Danish researchers, who were not available for comment, could not refute or confirm those hypotheses, the authors said. However, given that about half of all trials should favor the control, not the experimental intervention, the latest findings on the association between funding source and study conclusion should be reason for caution, they note.

"The combined evidence underlines the need for an international register of all initiated randomized clinical trials," the authors conclude.

Cathy Tokarski

JAMA. 2003;290(7):921-928

Reviewed by Gary D. Vogin, MD

 

If you need help or advice, you are welcome to email our naturopathic team with your health question.

Disclaimer: The health information presented here has been written for the New Zealand health consumer. It is of a general nature and is only intended to provide a summary of the subjects covered. The information is not intended to be comprehensive or to provide medical advice to you. While all care has been taken to ensure the accuracy of the information, no responsibility or liability is accepted, and no person should act in reliance on any statement contained in the information provided. All health ailments should be treated by a qualified health professional.

ref:n97

• Tweet