Controlling Conflicts of Interest

Following a broad review begun in early 2003, Harvard Medical School (HMS) this May formally reaffirmed and updated its policies governing faculty members' potential conflicts of interest in conducting their research. The revisions maintain the basic architecture of prohibited and permitted activities, and of universal disclosure of researchers' pertinent financial interests, first drafted 15 years ago, but carefully address new aspects of faculty involvement in medical clinical trials and business management, among other matters. The changes reflect current relationships between the academy and industry in the push to move promising research into practical use: "from bench to bedside."

HMS dean Joseph B. Martin, who commissioned the policy review, cites "two very important principles" that governed the work: "to be absolutely certain that human subjects be protected" at any stage of research, and to assure that "there is not even a perception of bias" in basic or clinical research.

Professors Barbara McNeil and Christopher T. Walsh have worked on HMS conflict-of-interest policies since the late 1980s. Photograph by Stu Rosner

Why should such problems even arise? First, faculty members are permitted to use 20 percent of their time on outside activities—from lecturing to entrepreneurship—for which they may be compensated by honoraria, consultant fees, stock options, or equity. Second, since 1980, federal policy has encouraged commercialization of faculty inventions to speed relief to patients, allowing scientists and their institutions to benefit financially if federally sponsored research yields marketable products. Third, beyond the patent and royalty arrangements and ownership stakes offered in return for academic intellectual property, industrial support of research is common and may grow more so as new kinds of science encourage sharing of costly laboratories and tools; companies such as Merck and Novartis have recently established major facilities near HMS and its affiliated hospitals.

As a result, Martin says, "over the last 15 years there has been much broader acceptance of the idea that our intellectual property" has wider application, and a recognition that such application might yield the "occasional real financial boon back to the institution." That in turn has spurred wider acceptance of commercial ties to research, particularly in the affiliated hospitals—where much clinical research is based and where most of the several thousand professionals holding HMS faculty appointments work.

The school's initial response to such changes in academic medicine emerged from a committee chaired in the late 1980s by Barbara McNeil, a radiologist who is Watts professor of health care policy and head of the department of health care policy. She remembers its work as "fairly contentious," as members struggled to define a "financial interest," "family" who might benefit, the precise meaning of "royalties," and the distinctions between "clinical" and "basic" research.

The guidelines (including subsequent minor amendments, available as "Faculty Policies on Integrity in Science" at www.hms.harvard.edu/integrity) prohibit two situations. A faculty member may not participate in clinical research on a technology owned by or obligated to a business in which he or she or a family member has any financial interest (consulting, ownership). Clinical or basic research, not involving human subjects, is also forbidden if the faculty member's institution receives research funding for the work from a business in which he or she holds stock. Exceptions permitted include "de minimis" holdings of $20,000 of publicly held equity—so long as there is no tie between that stake and specific research to be conducted—and $10,000 per year of consulting fees or honoraria. Other activities are governed by disclosure and prior approval requirements, or are routinely permitted (for example, receiving royalties for published work).

For most of the decade, says associate dean for faculty affairs Margaret L. Dale, who joined HMS in 1991, the conflict of interest procedures provoked little comment.

But in what he characterizes as the "boom time" in the years after his appointment as dean in 1997, Martin recalls, some faculty members "felt we were missing out in biotech and our investment in our intellectual capital" and encouraged HMS to review its policies. That review began as the ugly side of conflict problems became evident—notably the death of a patient in a gene-therapy experiment at the University of Pennsylvania. In May 2000, Martin announced that HMS would not modify its policies, and called for national debate on how best to maintain the integrity of research. That work, effected through discussion with peer institutions and through the Association of American Medical Colleges (AAMC), led to model guidelines in 2002.

When Martin initiated a new review of HMS practices in 2003, he appointed McNeil to chair a clinical-research committee. Kuhn professor of biological chemistry and molecular pharmacology Christopher T. Walsh, a member of McNeil's initial group in the late 1980s, this time was tapped to chair a separate committee examining basic research.

Reflecting a fundamental change in science, McNeil's group clarified the definition of "clinical research" so it continues to include "human subjects in vivo or the use of human samples" but now excludes "commercially obtained de-identified" cell lines and tissues.

The committees also broadened the definition of who is considered a "participant" in research to include study designers—who may play a critical role in determining what samples are included in a research panel—and anyone who may be an author on the publication of research findings. The clarifications reflect the growth in and complexity of contemporary clinical studies and trials. McNeil notes the high stakes: trial design plays an important role in securing Food and Drug Administration approval of therapies and in subsequent decisions by healthcare payers to reimburse new treatments—a critical factor in financial success if an invention is marketed.

In other refinements, the new language specifies that the de minimis levels of stock ownership and fees be adjusted upward to $30,000 and $20,000, respectively. At the same time, to be sure the rules are not circumvented, the committees insisted that any allowed royalties shared with a faculty member under an institutional licensing agreement must be made after an invention is marketed; no earlier-stage payments—equity, say, or payments for research in progress—are permitted outside the de minimis rules. A faculty member's freedom to acquire an equity interest in a company will now begin not when sponsored research funding ends, but only after publications on the relevant research are completed. Finally, to clarify the limit on outside obligations, faculty members may not hold such positions as chief executive officer, scientific director, or medical director of for-profit biomedical enterprises.

Walsh says the new policies represent "incremental" change in Harvard's rules, which he terms a "strict constructionist" guide to suitable behavior. "We didn't see any reason to change the basic fabric of prohibition" of research where a faculty member held equity in a company, he says, while permitting consulting within the HMS context of securing disclosure of possible conflicts. (Such disclosure now requires that prospective students, trainees, and new faculty members be informed of potential conflicts before they join a laboratory.)

Walsh says his committee members did not see Harvard's policies as unduly restraining the "translation" of research to applied therapies. As part of its work, the committee "got a very careful description of MIT's policies," relevant because MIT is widely considered to conduct itself with the highest integrity and to pursue innovative research while being "very embedded in the world." That Harvard's policies are "very parallel to and congruent with MIT's" was reassuring, he says.

The HMS faculty approved the new language in May "without dissent," Martin notes. The perception that Harvard's policies are unrealistic, he says, has been dispelled; the challenge he perceives is communicating fully with faculty members and the wider biomedical industry that the standards protect human subjects, maintain integrity, and accommodate appropriate translation from discovery to medical practice.

Looking ahead, Martin sees three related priorities. First, he worries that some academic research centers "haven't come to address the issues relevant to their institutions" despite the AAMC guidelines. Future conflicts of interest or harm to patients elsewhere could erode public support for biomedical research generally. Second, HMS's policies govern the behavior of individual researchers; policies on institutional conflicts of interest—where the University or an affiliated hospital is offered private equity in return for early access to research, for example—have yet to be formulated. That issue is on Harvard's policymaking agenda now.

Third, as those issues await resolution, opportunities are at hand "to find ways to work together" with the pharmaceutical enterprises that have set up shop in Cambridge and Boston. Merck's research center, for instance, is adjacent to HMS's huge New Research Building, commissioned last year. The company's focus on cancer, Alzheimer's disease, and obesity all complement HMS strengths, and "positive" meetings have been held on research contracting and investments in core facilities to screen chemical compounds, conduct high-cost imaging studies, and pursue other common needs. Says Martin, "Boston is now the center for all this."