Study finds physicians need to more accurately disclose industry relationships

Self-disclosure rates were highest at one meeting when payments were directly related to research.

There was a 79% rate of self-disclosure at the 2008 American Academy of Orthopaedic Surgeons annual meeting of payments that hip and knee replacement specialists received from industry when those payments were directly related to the topic of the physician’s presentation or role at the meeting, an observational study found.

The study, published in the New England Journal of Medicine, also found the rate was 50% when payments were indirectly related to the research topic.

Mininder S. Kocher, MD, MPH, one of the study authors, told Orthopedics Today the methods he and colleagues used to assess accuracy of these conflict of interest disclosures were “fairly robust.”

The American Academy of Orthopaedic Surgeons (AAOS) was unaware the study had been conducted until just prio to publicatin, according to academy officials.

Settlement data

Kocher and co-investigators accessed public information from five major orthopedic companies that manufacture total hip and knee prostheses about payments they made to physicians in 2007, data which was available following the companies’ settlement with the U.S. Department of Justice (DOJ).

“The thrust of disclosure from the point of view of presenting research articles at meetings or publication in journals has been voluntary self-disclosure by physicians. It is hard to look at the accuracy or validity of that, so I think the DOJ settlement allowed us to do that,” Kocher said.

To determine nondisclosure rates, he and colleagues compared the company-declared payments to the physician-supplied payment information in the final meeting program, not taking into account what was said or displayed during actual presentations.

They surveyed the 91 physicians who received payments, but did not declare that relationship in the program, asking why that occurred. Of the 36 respondents, 39% said the payment did not directly relate to their presentation; and 14% said they misunderstood the disclosure requirement.

“One of the main solutions to conflict of interest is disclosure,” but it must be accurate so that fair, yet individual assessments can be made concerning whether the conflict of interest may have influenced the physician or research outcome, Kocher explained.

The study had some limitations, he noted.

“We were a little limited by the data,” which assumed DOJ payment declarations and physicians’ statements were accurate. Company payments listed in the DOJ-mandated disclosures did not indicate which products the payments related to, so investigators drew some conclusions of their own, Kocher said. To determine relatedness of a payment to the subject being presented “also was a bit of a judgment call,” he noted.

AAOS responds

Orthopedics Today discussed the study with AAOS President Joseph D. Zuckerman, MD, who took issue with the timeline involved, saying disclosure data for the 2008 AAOS meeting was due in August 2007, but DOJ settlement data appeared later in October 2007 and the meeting was held in March 2008.

“I would like to think if [the study] was done in 2009, it would be different. There are a number of things the academy has done without knowledge of this study as part of our ongoing process of evaluating our disclosure requirements,” including launching an electronic method of filing conflict of interest data that allows for frequent updates and training AAOS members on proper disclosure, Zuckerman said.

“We, like most other organizations, have looked to do more detailed disclosure than less.”

Kocher and Zuckerman agreed greater transparency and national registers of physician and industry disclosure data, as the 2009 physician payment Sunshine Act proposes, are reasonable approaches.

Kocher said the findings can be generalized across all medical specialties. “Relationships between physicians and industry are common. They have the potential to be positive and really benefit the patient, but there are these negative potentials we need to be aware of.”

For more information:

• Mininder S. Kocher, MD, MPH, can be reached at Children’s Hospital Boston, 319 Longwood Ave., Boston, MA 02115; 617-355-8423; e-mail: mininder.kocher@childrens.harvard.edu. He receives grant support from ConMed Linvatec and is a consultant to ConMed Linvatec, Covidien, EBI and Smith and Nephew.
• Joseph D. Zuckerman, MD, can be reached at NYU Hospital for Joint Diseases, 301 E. 17th St., 14th Floor, New York, NY 10003; 212-598-6674; e-mail: joseph.zuckerman@nyumc.org. He receives royalties from Exactech for a shoulder replacement system design.

Reference:

• Okike K, Kocher MS, Wei EX, et al. Accuracy of conflict-of-interest disclosures reported by physicians. N Engl J Med. 2009;361:1466-1474.