FRIDAY, Oct. 19, 2001 (HealthDayNews) -- When two Danish researchers declared last year that routine mammograms didn't save lives, they were roundly criticized, and their work was discredited even by many colleagues.
But in either a display of gluttony for punishment or scientific martyrdom, the two scientists say they've reviewed the data a second time and reached the same conclusion: Screening by mammography is at best a sketchy use of resources. It may lead to more aggressive treatments that could put women at unnecessary risk of harm, they reiterate. The new review appears in the latest issue of the prestigious British medical journal The Lancet.
And just as they did last year, screening advocates dismissed those conclusions and rallied to the defense of mammograms.
Robert Smith, director of cancer screening at the American Cancer Society, says women should ignore the Danish research. "We know that early detection saves lives." As for the suggestion that X-ray mammography doesn't lower deaths from all causes of disease, "None of the experts in screening would agree with them. In fact, no one agrees with this concept," Smith says.
In the first go-round, Ole Olsen and his colleague Peter Gøtzsche of the Nordic Cochrane Center in Copenhagen tried to find out why, although routine mammography has been available in Sweden since 1985, breast cancer death rates in that country weren't dropping.
The Danish scientists analyzed seven earlier trials that screening advocates considered proof of the tests' utility and found crippling flaws in several. They asserted that even the two cleanest studies found that screening by mammography didn't reduce a woman's overall risk of dying. And since they found no solid evidence of the exam's ability to reduce mortality, they said it was "unjustified."
During the storm of controversy that followed, Smith says the American Cancer Society conducted a study last May that found that screening women in Sweden has led to a 63 percent drop in deaths from breast cancer there.
Meanwhile, Olsen and Gøtzsche revisited the seven studies. This time, their findings were reviewed by the Cochrane Breast Cancer Group, a well-regarded collaboration of scientists who assess scientific research.
"The [new] analysis is much more rigorous. It is now based on more than 200 reports of the trials and a meticulous and transparent assessment of the trial quality comprising eight pages of text. Never before has so much information about the trials been summarized systematically in one place," Olsen says in an e-mail interview.
Olsen, a statistician, says he expects the follow-up report will be better received than the first, which he admits was not as well "underpinned" as it might have been. Still, he says the outcry last year was partly the reaction of optimists to skepticism. "Additionally, humans in general tend not to be happy if weaknesses in their work are pointed out. And it is always difficult to change views if you have had a strongly held position," he says.
Richard Horton, editor of The Lancet, supports that view in an editorial accompanying the Danish researchers' new article, and goes on to endorse the paper's conclusions. "At present, there is no reliable evidence from large randomized trials to support screening mammography programs," he writes.
In addition to the journal article, The Lancet is
publishing the researchers' review in full on its Web site, but not a
version prepared by the Cochrane group, which had agreed
to release its work.
As Horton details in his editorial, the Cochrane group made a
number of changes to the new review, alterations the two
Danish scientists say undermined their latest research.
Smith says The Lancet has taken the "unorthodox" step of endorsing a journal article that relies on disputed research methods, then published only one version of the research on its Web site. He says that does "a disservice" to scientists who read the journal and the additional material, and whose decisions may impact patient care.
Smith says since it's clear that mammography saves lives, the latest stir offers more insight into the Byzantine nature of scientific publication than it does into the viability of the screening test. "This entire debate provides an enormous amount of material for probably more than one dissertation on the sociology of science," he says.
What To Do
This study notwithstanding, the American Cancer Society says every woman over age 40 should have a routine mammogram each year, in addition to monthly breast self-examinations for those over 20.
For more on screening by mammography, check the American Cancer Society or the Northwestern Memorial Hospital Web sites.
