|
Epi Wit & Wisdom Articles
Review of Epidemiologic Data on
Breast Cancer Screening Evokes Strong Reactions
“As a panel member and an
epidemiologist, I was asked to evaluate the benefits and risks
associated with mammographic screening among women 40 - 49 years of
age... The mortality benefit for women 50 and above is not
questioned—the benefit to very young women (i.e. those in their 20’s
and 30’s) is not beneficial. But the benefit to women 40 - 49 is
uncertain. Do we make a definite recommendation to all women 40 - 49
—all healthy women 40 - 49—when we are not certain? I think not...the
panel recommended that each woman should consider her own risk-benefit
profile and with the best possible information, make her own
decision...I felt and still feel, we arrived at the best conclusion
based on the available data.”
So spoke Susan Chu, Seattle
based epidemiologist at Group Health Cooperative of Puget Sound, when
called to testify before Congress in early February following the
presentation of a statement on breast cancer screening by an NIH
consensus panel.
The uproar caused by the panel’s
recommendation was amply described in the press last month. According
to a New York Times account, Leon Gordis, a Johns Hopkins University
epidemiologist who chaired the panel, said that nothing had prepared
him for the venomous reaction his panel got when it said in a report
that it had no reason to recommend routine mammograms for women under
50.
The Panel’s Responsibility
Exactly what did the panel get
asked to do? According to the National Cancer Institute’s Richard
Klausner, the panel was asked to address the often confusing and
sometimes contentious debate that surrounds the question of the age at
which a woman should begin getting regular mammography screening.
According to Klausner, the best data is from eight randomized clinical
trials involving about 180,000 women, including five Swedish studies.
Few trials have enough deaths from breast cancer to achieve
statistical significance, but using meta-analysis there is about a 15%
reduction in mortality. He estimates that over 30,000 women in their
forties will be diagnosed with breast cancer this year and a 15%
reduction in mortality would mean over 1,600 lives saved.
Despite press accounts that
Klausner said he was “shocked” at the time by the conclusions of the
NIH panel, he indicated in his testimony to Congress that he agrees
with the conclusion of the panel, but “it is my opinion that the draft
report of the panel overly minimizes the benefits and overly
emphasizes the risks for this population. A balanced statement of the
pros and cons of screening is essential for a woman to make an
informed decision whether to initiate regular mammography in her
forties.”
Among the risks associated with
mammography are the occurrence of false positives and the procedures
involved in the follow-up of these false positives. Also, the test
fails to detect up to one fourth of breast cancer cases in this age
group. There is no direct evidence that the levels of radiation used
in mammography cause breast cancer.
Why did the benefits not seem
strong or clear enough? According to Chu, not all the randomized
trials showed the reduction in mortality. Two studies showed a
significant benefit and two studies showed no significant benefit. “We
should not create certainty where there is none to reassure the public
or to ‘please the audience’. Our task was to objectively weigh the
health benefits and risks of an universal screening program for women
40 - 49. I feel we met that task,” she said.
Why the Strong Reaction?
Several potential reasons for
the strong reactions to the panel’s recommendation have been mentioned
in the press accounts about this report. One reason involves the
marketing of mammography—by instrument makers, medical centers, and
even the American Cancer Society—as a way to ease the fears women have
about breast cancer. Because it has been touted as a way to find
early-stage tumors and therefore strongly influence a women’s chances
of a positive outcome, “few want to hear that its much touted
effectiveness may be in doubt for women under age 50” (NYT, January
28, 1997).
Reaction to the panel’s
recommendations have been so strong that even Congress has jumped into
the act. Olympia Snowe (R-Maine) introduced a resolution urging the
NCI advisory panel to consider reissuing guidelines recommending that
women in their forties seek routine mammograms. “Good science dictates
that mammography for women in their forties saves lives,” she
insisted.
Not the Senate’s Business
Others are questioning the
appropriateness of having politicians involved in the evaluation of
scientific evidence. According to a Washington Post article (February
10, 1997) by Jessica Matthews, senior fellow at the Council on Foreign
Relations, “the Senate is once again trespassing where it has no
business to be. On the basis of some mysteriously acquired
epidemiological insight, 98 senators...recently concluded that women
in their forties benefit from routine mammograms...The National Cancer
Institute is under no illusion as to the source of its funding. The
chilling effect of a debacle like this one is not lost on those in
Bethesda and around the country whose research is funded by the
various National Institutes of Health. Sadly, for some, (including
often the most conscientious researchers) the lesson is to stay away
from any highly controversial issue. The larger result is to make the
NIH and the government’s other health-related agencies, whose job is
to evaluate data dispassionately, into less valuable protectors of
Americans’ health.”
[Editor’s note: The current
controversy is only the latest flare up in a long-standing
disagreement about the value of breast cancer screening for women in
their forties. For a detailed account of the history behind this
debate, see Gary Taubes’s recent article in Science, (February 21).
Believe it or not, epidemiologists are not painted in negative terms.]
Published March 1997 v
|
