THE British Medical Journal carried an editorial last month
dismissing the notion that antidepressants such as Prozac are addictive鈥攁
claim that some pressure groups would dispute. It later emerged that the three
European doctors who wrote the article had been flown out to Phoenix, Arizona,
by the drugs company Eli Lilly鈥攖he manufacturer of Prozac鈥攖o attend
its closed symposium in December 1996.
This perceived conflict of interest has been strongly criticised by the
British group Social Audit, which among other things campaigns for the release
of more information about medicines. Its director, Charles Medawar, thinks the
authors鈥� links with Eli Lilly should have been disclosed.
鈥淚 didn鈥檛 tell the BMJ because I didn鈥檛 think it was important,鈥�
says Allan Young, one of the authors of the editorial (vol 316, p 1105) and a
lecturer in psychiatry at Newcastle General Hospital. 鈥淢aybe I鈥檝e been naive.
But I鈥檝e been to events organised by other drugs companies, too.鈥� The
BMJ鈥檚 editor, Richard Smith, says that if the editorial is shown to involve
a conflict of interest, the journal will publish an announcement to this
effect.
Advertisement
There is no suggestion that the BMJ authors based their opinions on
anything other than their clinical expertise. But when it comes to medical
research, critics say drugs companies often have too much influence. 鈥淒rugs
company money certainly funds useful research, but inevitably it鈥檚 going to
influence what鈥檚 happening,鈥� says Medawar. Several recent reports in leading
American medical journals back this up.
In a paper that appeared in The New England Journal of Medicine(
NEJM) in January (vol 338, p 101), a team led by Henry Stelfox at the
University of Toronto reviewed all the English-language articles published
between March 1995 and September 1996 on the safety of calcium-channel blockers,
used by millions for the treatment of high blood pressure and angina. These
drugs are highly controversial鈥攖here is concern that they may cause cancer
or heart failure鈥攁nd the researchers wanted to find out whether money from
pharmaceuticals companies influenced the outcome of studies into their
safety.
Undeclared interest
The results were alarming. Authors who supported the use of calcium-channel
blockers were nearly three times as likely to have financial links with the
company that made the drugs as critical or neutral authors
(This Week, 17 January, p 21).
Yet the vast majority of researchers with such links had not
declared a conflict of interest.
In some cases a drugs company鈥檚 influence on research can be overriding. In
1989, Boots Pharmaceuticals commissioned Betty Dong, a clinical pharmacist at
the University of California in San Francisco, to compare its treatment for
hypothyroidism 鈥攁 condition that can cause tiredness and hair
loss鈥攚ith cheaper alternatives made by other companies. The Boots drug,
Synthroid, is taken by millions of people every day and was the medicine of
choice for American doctors. A complete switch to cheaper brands would have
saved the US health department $356 million a year.
Dong and her colleagues compiled evidence that Synthroid was no better than
the cheaper drugs. But Boots Pharmaceuticals, which had since been bought by
Knoll Pharmaceuticals, withdrew the paper just as the research was about to be
published in The Journal of the American Medical Association(
JAMA)鈥攁s the contract allowed it to do鈥攃laiming it was
鈥渇undamentally flawed鈥�. The paper finally appeared in 1997 (JAMA, vol
277, p 1238). Synthroid users sued Knoll, which settled out of court for
$98 million.
Critics of the way industry funds research say that companies generally give
money only when they are confident that a study鈥檚 results will help them
commercially. No company would pay for research that it expected would show that
its products were inferior. But as Boots found, the results of research are not
always predictable.
In another instance Burroughs Wellcome, now Glaxo Wellcome, confident that
its drug AZT when used alone would delay the onset of AIDS in HIV-positive
people who did not yet show any symptoms, funded a huge Anglo-French trial. To
the surprise of many, the results showed that AZT did not delay AIDS鈥攁nd
Wellcome began making critical noises about the trial.
A necessary burden
鈥淎 drugs company will always try to put a spin on results,鈥� says Ian Weller,
an AIDS specialist at University College London, who led the trial. 鈥淭he more
independent you are, the more you can stop that. But you also have to bear in
mind that without the companies, we wouldn鈥檛 have these treatments.鈥�
Industry funding is necessary, agrees Janet Darbyshire, head of the Medical
Research Council鈥檚 HIV clinical trials centre at University College London,
because drugs studies are very expensive. The fact that they often involve many
researchers means that there 鈥渁re likely to be a lot of independent-minded
people taking part鈥�, which acts as a safeguard.
Some researchers are prepared to go further and promote the products they
research. In February, 快猫短视频 received a letter on University of
Ottawa headed paper signed by Bill Cameron, associate professor of medicine in
the Division of Infectious Diseases. It read: 鈥淚 am delighted to inform you that
The Lancet, dated Saturday 21 February 1998, includes the results of a
large-scale international study which demonstrates that ritonavir (Norvir) . . .
reduces the risk of AIDS complications and prolongs survival in patients with
advanced HIV disease.鈥�
The study in The Lancet (vol 351, p 543), which Cameron led, was
sponsored by the drug鈥檚 manufacturer Abbott Laboratories. Senior HIV researchers
to whom 快猫短视频 spoke were critical of Cameron鈥檚 letter, asking
whether a scientist who receives funding from a drugs company to test its
product should endorse promotional letters to journalists. Richard Nicholson,
editor of the Bulletin of Medical Ethics, said: 鈥淲hy wasn鈥檛 it enough
to have the trial published in The Lancet?鈥�
Allan Detsky, one of the authors of the NEJM paper on
calcium-channel blockers, thinks that the medical profession needs to develop a
more effective policy on conflicts of interest. He calls for the complete
disclosure of relationships with drugs companies by clinicians and researchers
who write articles about such products.
Smith agrees that such links should be brought into the open. Industry
funding strongly affects 鈥渉ow researchers think, how they act and what kind of
research they do鈥�, he says. A company鈥檚 financial involvement often goes beyond
research, he adds, and may include consultancy fees for the scientists involved.
鈥淵ou can鈥檛 get rid of these conflicts of interest. So what you have to do is
disclose them. And at the moment, we in the medical press aren鈥檛 doing a very
good job of it.鈥�
The problem was underlined by another report in last month鈥檚 JAMA
(vol 279, p 995). A team of Harvard researchers looked at how often life
sciences researchers in the top 50 US universities were offered research-related
gifts from industry. They found that 43 per cent of the 2128 respondents had
been offered presents from companies (see figure). Almost a third of recipients
reported that the donors wanted pre-publication review of any articles stemming
from use of their gifts.
The JAMA paper offers this advice to researchers involved in
industry-funded studies: 鈥淎t times it may be prudent鈥o look a gift horse
in the mouth.鈥�