"I'm amazed at what he did," said Furberg, a professor at Wake Forest University. "And to do it in secrecy I find very objectionable. This is a critical change in the NIH policy."
In 1999, Varmus wrote a letter to the institute directors that cautioned them to "avoid even the appearance of a conflict of interest." But in an attachment to the letter, he told them that employees "may briefly discuss or mention current work" to outsiders, in effect giving agency scientists permission to reveal their unpublished, confidential research.
Varmus, now president and chief executive of the Memorial Sloan-Kettering Cancer Center in New York, declined to be interviewed for this article. His spokeswoman, R. Anne Thomas, said that Varmus, who in 1989 shared a Nobel Prize for research into the genetic basis of cancer, believed that NIH employees should take personal responsibility for avoiding conflicts of interest, regardless of what agency rules allow.
Kirschstein, after taking over as Varmus' interim successor at the NIH three years ago, said in a May 2000 speech to medical researchers that conflicts of interest posed "a major concern."
"While the federal government was once the dominant force for supporting clinical research, today we share the arena with biotechnology companies, pharmaceutical firms and many others -- all interested in the possibility of financial gain from their research.
"Profit raises issues of public trust," she said. "When scientific inquiry generates findings that can make a profit for the researcher and the institution, their images become clouded."
Yet officials have lifted controls on consulting even as industry's stake in NIH research has deepened. When Zerhouni, the current NIH director, appeared before the House Subcommittee on Environment, Technology and Standards last year, he cited 274 ongoing research and development agreements between the federal agency and industry.
At the same time, NIH leaders have moved to what they describe as "managing" conflicts of interest. Employees are allowed to consult if they receive prior clearance from an administrator at their institute or, in the case of most institute directors, from NIH headquarters.
An Honor System
Potential conflicts are typically addressed by allowing employees to sign "recusals." Under these agreements, NIH employees pledge not to participate in decisions affecting an outside client. Agency officials, Kirschstein said, rely on an honor system to enforce recusals and other conflict-of-interest rules.
