Medical Tests Skewed, Study Finds
NEW YORK -- The scramble for profit warps the way scientists and universities conduct medical experiments, undermining the integrity of research, said Yale University investigators who studied the impact of commercial funding on science.
One-quarter of the biomedical researchers at universities had commercial ties serious enough to raise questions of financial conflict, the analysts found. In many cases, it was enough to skew their research.
Moreover, the universities expected to police the integrity and ethics of faculty scientists have their own commercial research interests and financial conflicts. At least two-thirds of the universities also were involved in commercial ventures, holding equity shares in start-up companies whose research they were also expected to monitor. Twenty-seven universities had equity in 10 or more start-up companies, the researchers said.
The result can be slanted science.
Industry-sponsored research is 3.6 times more likely to produce results favorable to the company that helped pay for it, the Yale researchers determined in a university-funded study published today in the Journal of the American Medical Assn.
“Medical research is a matter of life and death,” said Yale University medical analyst Justin Bekelman, who led the study team. “The guidance patients receive from their doctors relies on valid scientific research.”
In all, the Yale researchers assembled and analyzed data from 37 previously published peer-reviewed studies -- covering hundreds of research projects, thousands of scientists, and more than 1,000 peer-reviewed scientific papers -- on the extent, impact and management of financial conflicts.
They named no scientists or schools involved, nor did they reveal the names of any firms. Conducting the most extensive study of commercialism and science so far, they did identify a troubling trend across the realm of biomedical research.
Their review covered the decades from 1980 to 2000, a period during which the share of commercial funding grew to 62% of all U.S. spending on biomedical research. Even as public spending on biomedical research doubled in the last five years, the financial ties between academic scientists, universities and industry became more common and more likely than ever to influence research findings.
“Financial conflicts were even more prevalent among institutions than among individuals,” said Yale research analyst Cary Gross. “How workable is oversight and the policing of financial conflicts by institutions if they themselves have conflicts?”
By necessity, the public looks for reliable medical guidance almost every day.
Public health warnings, medical claims and policy pronouncements come draped in the cloak of scientific authority, usually backed by a bible of published research findings. But along the frontier of science and business, the line between research and marketing can be hard to find.
Every day seemingly brings a blizzard of new assertions: Roller coasters are safer than bunk beds. Wine and walnuts improve your health. This vaccine is harmless. That drug works better than a placebo or competitor.
Who among the chorus of experts should the public believe?
“It is getting harder and harder to find someone who can conduct the research or evaluate the research who does not have a financial interest in a particular product or particular company or a particular outcome,” said Stanford University biomedical ethicist Mildred Cho.
This financial tangle is the consequence of a basic change in the way biomedical research is conducted in the United States, several experts said.
Commercial funding, once anathema to ivory-tower academics, has become the lifeblood of biomedical research, fostered by federal laws and regulatory changes since 1980 that give scientists and schools more freedom to profit from work once considered in the public domain.
At most universities and medical centers, researchers are obliged to tell people whom they want to enroll in an experiment that the researchers have a financial stake in the drug or treatment on trial. Many medical journals, which publish new research, also have rules requiring scientists to reveal any financial stake they may have in the work.
Such disclosure policies are meant to alert the public to the possibility of bias in published research, but they are honored more in the breach than in practice, medical research analysts at Stanford and Tufts universities said.
When researchers checked 47 of the most influential biomedical journals, fewer than half had disclosure policies. When Richard Smith, editor of the British Medical Journal, checked 3,642 articles published in the five leading medical journals, only 52 declared the authors’ financial stake in the works.
In response to growing qualms about clinical trials, the American Assn. of Medical Colleges, which represents the nation’s 125 accredited medical schools and nearly 400 major teaching hospitals, recently issued stricter guidelines to control financial relationships for medical research, urging that no one be allowed to experiment with human beings if they have a financial stake in the outcome.
So did the American Society for Gene Therapy. Its new, voluntary guidelines insist that anyone involved in choosing patients or managing a clinical trial “must not have equity, stock options or comparable arrangements in companies sponsoring the trial.”
Even so, it can be unclear who has most at stake in a clinical trial; the patient, the doctor or a corporate research partner.
“The results of [a clinical] investigation can swing huge amounts of money one way or another,” said David Korn, senior vice president for biomedical and health sciences research at the American Assn. of Medical Colleges. “The way the world is now, we need a very high degree of sensitivity and a very high degree of scrutiny.”
Voluntary guidelines and good intentions are hardly enough, said Tufts University policy analyst Sheldon Krimsky, author of a forthcoming book called “Science in the Private Interest.”
“The conventional view was that the integrity of the scientists put them beyond the corruption of funding,” Krimsky said. The new study, however, shows “that private funding of research can bias the results.”
Whether it comes in direct grants and consultant fees or through gifts of research materials and travel expenses, commercial funding can affect a researcher’s behavior, said health care analyst David Blumenthal at the Harvard University Medical School.
It can sway the kinds of scientific questions researchers ask, the way they frame their experiments, and the role they play in influencing the medical decisions that people make about their health.
Several medical research analysts said that the pharmaceutical industry is especially aggressive in its efforts to influence researchers.
By itself, the sheer size of the pharmaceutical research enterprise is persuasive enough. A single company -- drug giant Pfizer Inc. -- spent $5.3 billion on research and development last year, more than the National Science Foundation.
The drug companies also proselytize aggressively. In 2000, the drug industry sponsored more than 314,000 events for doctors, ranging from luncheons to weekends, some in exotic locales, at a cost of about $2 billion.
“Physicians who take part in research studies often become spokespersons for the companies or join advisory boards,” said Jerome Kassirer, a former editor of the New England Journal of Medicine who teaches at Yale and the Tufts University Medical School. “Then they move up to become [medical] guideline writers. They get involved with professional organizations. They write blurbs and books, all of them supported by industry.”
