ONE of the most repugnant episodes in the history of medical research is the
Tuskegee Syphilis Study. From 1932 to 1972, American researchers followed 400
poor African Americans with syphilis to see how their disease progressed. Even
after penicillin became available in the 1950s, the researchers withheld the
cure, so that medical science could benefit from observing how syphilis would
progress untreated.
Such a horrible injustice could never happen today. Or could it? Some
ethicists say that researchers are still withholding proven therapies from sick
people. They do so not to follow the natural course of an illness, but rather to
compare patients without treatment with patients given a new drug. In other
words, some ethicists believe that placebo-controlled trials of new drugs are
unethical if doctors already have a therapy that could treat the disease, even
when patients give their consent to the study.
That idea is enshrined in the Declaration of Helsinki, a document adopted by
the World Medical Association in 1964 in response to Nazi experiments during the
Second World War. The declaration notes quite simply that 鈥渆very
patient鈥攊ncluding those in a control group, if any鈥� should be
assured of the best proven diagnostic and therapeutic method鈥�.
Advertisement
But some medical researchers are challenging their critics. They say we must
keep placebos鈥攅ven if there are available treatments. 鈥淢ost people ignore
the Declaration of Helsinki, even while they claim to honour it,鈥� says Robert
Levine of the Yale University School of Medicine in New Haven, Connecticut, who
chaired a committee that looked at revising the 36-year-old document. Levine
says that there are many circumstances where using placebos is scientifically
important and ethically justified. The World Medical Association may adopt a new
draft of the declaration next month when it meets in Edinburgh鈥攐ne that
will allow placebo trials for treatable conditions in certain circumstances when
the illness is not life-threatening.
Michael Baum of University College London medical school also supports a
pragmatic approach: 鈥淚t is a sign of maturity to reject the claims of the
absolutists and accept that ethical dilemmas are dilemmas that cannot be solved
by the rule book but have to be debated on their own merits,鈥漢e said last
month.
On the opposite side, Peter Lurie of pressure group Public Citizen in
Washington DC says that too many studies involve placebos. The most notorious
and controversial studies are those in Africa where pregnant women were given a
short course of AZT to see if it could reduce transmission of HIV to their
infants as effectively as a longer course
(New 杏吧原创, 1 July, p 42).
Ethicists argue that the researchers could have compared the short course
with the longer course, not placebo.
In another study, researchers compared a new drug called nifedipine with a
placebo in more than 1600 Chinese people with high blood pressure for an average
of 2.5 years. This happened even though doctors have known for years that
existing drugs can reduce mortality. Similar concerns have surfaced over the
ethics of studies on schizophrenia and anti-emetic treatments for cancer
patients.
But Levine notes that critics of trials such as the African AZT studies are
inconsistent, as they oppose placebo use while failing to call for medical and
social facilities that are up to the best Western standards are provided to the
participants. 鈥淚t is necessary to acknowledge with regret that there are great
imbalances in the distribution of wealth among the nations of the world.鈥�
The scope of medical research has changed considerably in the decades
following the Second World War. When the Declaration of Helsinki was first
adopted, randomised controlled trials of any kind were rare, says Levine. Now,
pharmaceuticals companies can鈥檛 get a drug on the market without testing it in a
controlled trial. Levine argues that the Declaration of Helsinki would stymie
research if taken literally. You couldn鈥檛 have placebo-controlled trials of
analgesics for headaches, or even drugs designed to treat baldness. His proposed
change would allow the use of placebos in trials as long as it did not cause
disability or death.
Karin Michels of Harvard Medical School and Kenneth Rothman of Boston
University argue that doctors have a duty not to withhold something that might
relieve their patients鈥� suffering, no matter how minor or transient it is. 鈥淲hy
should the patient have to suffer a headache? They鈥檙e coming to the doctor to
get rid of it, not contribute to science,鈥� says Michels. She rejects arguments
that informed consent from a willing patient relieves physicians of their
responsibility. Two weeks ago she and Rothman wrote an article in the
British Medical Journal arguing that the Declaration of Helsinki should be
rewritten to make it clear that all placebo controls are unethical if a current
treatment exists (vol 321, p 442).
But in the same issue, Baum, an eminent breast surgeon, argued against such
鈥渁bsolutism鈥�. As an example he cites the study of cyclical mastalgia (breast
pain), which affects around a quarter of women in their reproductive years.
Breast pain is unpleasant but not life-threatening and varies in its severity.
鈥淎 placebo in this setting is not the same as no treatment because it can
produce relief [the placebo effect] through mechanisms we can only speculate
on,鈥� he says. Specific treatments such as prolactin inhibitors exist 鈥渂ut at
considerable cost in side effects and to the drug budget鈥�. 鈥淚 would therefore
have no problem with a placebo-controlled trial in this setting.鈥�
However, placebos aren鈥檛 the only way to control a trial. Researchers can
give a new drug to patients and then compare the outcome with historical data on
the disease. Or they can add the new drug to an existing therapy and compare the
combination with the old drug. Lurie argues that comparing two drugs side by
side actually gives physicians the most relevant data, because they want to know
how a new drug compares with an existing one, not how it compares with nothing.
鈥淎ctive controls are essential. But they鈥檙e scientifically less desirable,鈥�
counters Levine.
Critics argue that if the FDA can use statistics to tease out meaningful data
from active control trials of antibiotics, why can鈥檛 they do so for things like
drug addiction treatments? 鈥淭he methodologies exist to do these trials as
comparison trials,鈥� says Michels. 鈥淚t鈥檚 easier to do a placebo-controlled trial
than an active-controlled trial.鈥�
Because the FDA carries such weight when it comes to drugs approval,
ethicists also worry about the International Conference on Harmonization, a
project shared by Europe, the US and Japan that aims to standardise the data
pharmaceuticals companies collect to get clearance for new drugs. A recent draft
document on choosing control groups seems to suggest that the emphasis on
placebos in the US may expand into other countries. 鈥淚t used to be that Japan
didn鈥檛 allow placebos at all, and Europe was very reluctant,鈥� says Michels.
Use of placebos may also increase if another proposed change to the
Declaration of Helsinki goes through. Levine鈥檚 committee argued that the
declaration should be changed so that patients were guaranteed not the best
proven therapy but the one 鈥渢hat would otherwise be available to him or her鈥�.
Levine says that this will help researchers working on affordable treatments for
people in developing countries. He says that these trials will not exploit
disadvantaged populations because they will be bound by The International
Ethical Guidelines for Biomedical Research Involving Human Subjects created in
1993 by the Council for International Organizations of Medical Sciences and the
World Health Organization.
But Lurie argues that vulnerable people in richer countries aren鈥檛 immune
from this kind of research. He cites a study of heroin addicts in Anchorage,
Alaska. Half got needles in an exchange programme, and the other half were
denied them. The researchers were then to see how many in each group got
hepatitis B鈥攄espite the existence of a highly effective hepatitis B
vaccine.
Lurie refused to serve on their panel of outside monitors, though hepatitis B
vaccine was subsequently offered to participants. 鈥淭hose people are ripe for
exploitation,鈥� he says. The researchers designing the trial ultimately wanted to
help addicts by proving that needle exchange programmes work.
This is one argument that is unlikely to go away.