True, False, Whatever
Physicians are putting a stop to the publication of misleading drug data
Catherine DeAngelis is not happy. The editor of the Journal of the American Medical Association knows that she is responsible for publishing deliberately misleading research that could have untoward consequences for thousands of patients. A year ago, she ran the results of a six-month study of the popular arthritis drug Celebrex; that probe showed the drug caused fewer gastrointestinal problems than comparable medications. But when the Food and Drug Administration reviewed the same trial five months later, she learned the study had actually lasted a year, and that the full collection of data in fact pointed to a quite different result. Indeed, Celebrex was not easier on the stomach than the other drugs. "The company had 12 months of data and didn't tell me," she says. "They know how upset I am."
The editors from the other major medical journals share DeAngelis's anger and frustration. In a joint editorial published this week, the heads of 12 international journals, including the New England Journal of Medicine, the Lancet, and the Canadian Medical Association Journal, warn that pharmaceutical companies have gained too much control over both medical research and the publication of experimental results. Clinical investigators are often excluded from study design and data interpretation, the editors contend, and in some cases unfavorable results are intentionally buried to avoid public dissemination. In an effort to win back some control over the scientific process, the editors are declaring a radical change in policy: "We will not review or publish articles based on studies that are conducted under conditions that allow the sponsor to have sole control of the data or to withhold publication."
The move is meant to give scientists more "clout" when negotiating contracts with drug companies, DeAngelis says. Harold Sox, editor of the Annals of Internal Medicine, agrees. "In the worst cases, the drug firms design the trial, explain to physicians how to carry it out, analyze the study, do not let the researchers see all of the data, and then control the publication," he says. "Physicians are not happy with this arrangement."
Well, perhaps not completely happy, but they are not shunning the drug companies either. In fact, the relationship between researchers and industry is far from adversarial, and it's often mutually beneficial, with the academic centers receiving much-needed funding and companies gaining access to prestigious university scientists. Only in rare instances, some observers argue, do researchers and their corporate sponsors strongly disagree about what to do with study results.
Cash cow. Still, few dispute the fact that academic medical centers have in the past few years become increasingly dependent on pharmaceutical company grants. Managed care and cuts in Medicare reimbursements have left many university hospitals cash strapped, says Marcia Angell, senior lecturer at Harvard Medical School. Pharmaceutical companies financed 70 percent of the clinical drug trials performed in the United States last year, investing a total of nearly $60 billion. And there are powerful incentives for medical researchers to go after that action. Not only do most rely on such "soft money" to keep their labs running, but the grants often are generous enough to support other research completely unrelated to the drug in question. Moreover, says Sox, "typically universities will get a negotiated amount of money over and above the cost of carrying out the trial." This "indirect cost recovery" pays for expenses such as electricity, library resources, and other essential features that make it possible to run a research university.
But such coziness has a price, and it's that conflict that has the editors concerned. Companies often want researchers to keep findings secret in order to protect proprietary information. Steven Rosenberg, chief of surgery at the National Cancer Institute, says one company asked him to keep study results confidential for 10 years; when he refused, the collaboration fell through. "Science depends on the open and free flow of information," Rosenberg says. "As more and more research is performed by private companies, the normal ethos of business intrudes on the normal ethos of science."
Some researchers are so cowed by the power of drug companies that they're reluctant even to voice disagreement. Says Steven Cummings, director of the clinical research program at the University of California-San Francisco: "Industry sponsors decide who to work with. Sometimes researchers are excluded from trials and opportunities to help with research if their views conflict with the company." By contrast, those whose views are favorable to a company's product are invited to speak at expense-paid conferences and to sign on to papers that appear in prestigious publications. In short, a good relationship with industry can advance an academic career.
Numbers game. How study data are analyzed and reported is at the heart of many of the conflicts between researchers and company sponsors. Curt Furberg, a professor of public health sciences at Wake Forest University School of Medicine in Winston-Salem, N.C., ran a trial comparing a new calcium channel blocker with a diuretic. The purpose was to see if the drug lowered blood pressure and reduced hardening of the arteries, but Furberg says the findings showed the diuretic worked better than the calcium channel blocker. The company, Sandoz (now Novartis), disagreed with Furberg's analysis of the data, and the drafting of the paper was contentious. "We went through the draft and presented our version of the results to the company," he says. "But the company kept changing it back to an older version that we didn't agree with." After 10 drafts, Furberg and four others gave up and took their names off the paper. "It was five years of work, and we got nothing out of it," he says. "But you have to have principles."
Stephen Lagakos, a professor of biostatistics at Harvard University School of Public Health, along with James Kahn, associate professor of medicine at UCSF, were denied access to complete data from a study of an HIV vaccine after they refused to include the company's analysis of the results in their research paper. They published the study in the Journal of the American Medical Association, using the partial data they had, and the company, Immune Response, responded by suing UCSF for damages of up to $10 million. UCSF countersued to get the rest of the data; the university won, but it took considerable time and money for Lagakos and Kahn to get access to the material.
Researchers and companies share the goal of publication in prestigious journals. It is a powerful medium by which scientists communicate, and it allows companies to promote new uses of their products with the imprimatur of rigorous peer review. Harvard's Angell, who is also the former editor of the New England Journal of Medicine, says she hopes the new policies of the medical journals will help to create more of the arm's-length relationship that academic researchers once had with companies. "All of the conflicts of interest cannot be eliminated," she says. But "the public should not have to wonder whether medical research can be believed."
THE CRACKDOWN
The journals signing on to the new policy:
Annals of Internal Medicine
Journal of the American Medical Association
New England Journal of Medicine
Canadian Medical Association Journal
Journal of the Norwegian Medical Association
Lancet
MEDLINE/Index Medicus
New Zealand Medical Journal
Norwegian Medical Association
Dutch Journal of Medicine
Medical Journal of Australia
Western Journal of Medicine
This story appears in the September 17, 2001 print edition of U.S. News & World Report.