BY: ELAINE CAREY
MEDICAL REPORTER
Mammograms in Canada and the
United States routinely produce more false positive readings
than those in other developed countries, according to a new study.
The number of mammograms deemed abnormal was 2 to 4 percentage
points higher in North America, without any corresponding increase
in the number of breast cancers detected, says the study, published
today in the Journal of the National Cancer Institute.
That leads to unnecessary follow-up procedures, higher medical
costs and anxiety for patients, says the study by the University
of Washington's Harborview Medical Center in Seattle. It estimates
that those costs in the United States add up to an additional
$750 million (U.S.) a year.
The data looked at 32 studies of community-based mammogram programs
in 18 countries, including 12,000 mammograms in British Columbia,
which had one of the highest false positive rates in the world.
Only 4.6 per cent of the Canadian women or one in every 22 who
were told they had an abnormal mammogram actually had cancer,
a rate of false positive tests slightly behind only those in
Hong Kong, Portugal and New Mexico, the study found. Of the Canadian
women who had a biopsy because of the mammogram reading, only
27.7 per cent had cancer.
By comparison, more than half of North Carolina women who had
biopsies based on mammograms had cancer, and more than two-thirds
of those in the Netherlands, the country with the best mammogram
record.
"If you look at these numbers, I'd rather be in North Carolina,"
said Dr. Cornelia Baines, a professor in the department of public
health sciences at the University of Toronto.
"You can see the relief on a woman's face when you tell
her that eight of 10 abnormal mammograms are false readings,"
she said. "But from a population point of view, eight out
of 10 of these follow-ups are unnecessary."
The new study, by a respected research group, is "very,
very sound in documenting persuasively this is a problem that
is more important in North America than in Europe and other countries,"
said Baines, a principal researcher on the Canadian National
Breast Screening Study in the 1980s.
"If you look at the countries and compare, you have to ask:
Why are countries like Iceland and Australia so much better than
Canada?" she said. "It goes back to the contrast between
entrepreneurial medicine and population programs that are rigorously
controlled."
While a number of other possible reasons are cited for the high
false positive rate, "I feel strongly this is the unfortunate
impact of medical malpractice," said Dr. Joann Elmore, the
study's lead investigator.
Failure to detect cancer is the number-one cause of malpractice
suits in the U.S., and breast cancer tops the list, she said
in an interview. Doctors are afraid not to order more tests for
even the slightest possibility of a problem with the mammogram.
"Radiologists have it tough, and primary-care doctors are
also sued for failing to follow up," she said.
Baines said malpractice is also an issue in Canada, although
not to the same extent as in the U.S. "But it's still a
factor any doctor would have in the back of their mind,"
she said.
Another factor is the quality of mammograms, which in the U.S.
have been found to be "atrocious," she said. "There
are bad mammograms done in Toronto, too, but there are also good
ones."
Financial incentives built into the health care system and how
different doctors interpret the tests could be other reasons,
the study suggests. Even in Canada, the fee-for-service system
encourages more testing, and the Ontario and B.C. breast screening
programs have been "basically persuaded that the screening
has to occur in private doctors' offices," Baines said.
Elmore said she hypothesized that the study would find higher
rates of abnormal mammograms in North America but also that they
would detect more breast cancers, which they didn't.
"It would be okay if the higher recall rate gave us more
bang for our buck, but we didn't find that," she said. "The
majority of recalls were false positives."
The differences persisted in the North American studies even
after controlling for the age of the women screened, the number
of readers for each mammogram and the number of pictures taken
in each mammogram. This isn't the first study to come up with
surprising news about mammography. A study by Toronto doctors
about a year ago showed no reduction in breast cancer death rates
for women who get the screening test in their 40s. And an international
uproar has surged in the past year about whether routine mammography
saves lives.