A new analysis of older studies sparked a debate about whether the use of mammograms to screen women for breast cancer saves lives. After reviewing eight published studies on the effectiveness of screening, a team of scientists found that both showed no benefit while six others were potentially biased and therefore of little use. But other researchers strongly dispute the conclusions
Much of the debate about mammograms sought to determine if they are useful for women in their 40s. most scientists accepted the benefits for women over 50 and proven ( Science , 21 February 1997, p. 1056 and April 4, 1997, p. 27). But new issues of meta-analysis that conclusion.
internist Peter Gøtzsche from the Nordic Cochrane Centre in Copenhagen and his colleague Ole Olsen watched eight screening effectiveness studies conducted since 1963. In each study, women were randomly assigned to either a group that mammograms regular checks, or to a control group that did not get the chest x-ray procedure. In six studies, the Danish duo claims, the two groups differed significantly in age or socioeconomic status, can skew the results in either direction, or the authors describe their randomization procedure thoroughly enough to convince Gøtzsche Olsen and that the subjects had an equal chance of being in the two groups. Of the two studies, the authors consider well randomized and therefore impartial - one from Canada and one in Malmö, Sweden - nor found that mammograms increased life expectancy. Therefore, "breast cancer screening by mammography is unjustified," the team writes in Lancet tomorrow .
"Nonsense," says Daniel Kopans radiologist at Massachusetts General Hospital in Boston . the bias and Gøtzsche Olsen report are trivial, he says, for example, they dismiss a study in Gothenburg, in Sweden, because the control group was older, on average, than the mammography group - but a months. And the Canadian trial was so much prejudice that one of the licensees studies, he said. Significantly more women in the group of mammography than in the control group began in the Canadian experience with cancer advanced breast, Kopans said, a difference that he and others suspect perhaps arisen because nurses assigned to women in the other group also conducted an initial manual breast examination and may have affected women with pieces within the group of mammography in the hope they would be diagnosed more quickly, said Kopans. A survey of a subset of Canadian data found no evidence of bias, but Kopans and others are suspects.
Gøtzsche stands by its results. Scrutinizing the effectiveness of mammography is particularly important, he said, because this process disadvantages, too: Mammograms often produce false positives, which cause anxiety and may lead women to undergo unnecessary biopsies or surgery . Each mammography study was dissected and supported for years, and Gøtzsche also provides that people will come gunning for his study. "We look forward to the scientific debate that will come now," he said.
0 Komentar