New Study Challenges Mammogram Value

N E W   Y O R K, Sept. 19, 2000 -- A new Canadian study of 50 to 59-year-old women who received annual mammograms challenges the conventional wisdom that the diagnostic test lowers the incidence of death from breast cancer.

The 13-year study of almost 40,000 women, led by Dr. Anthony Miller, professor emeritus of the University of Toronto’s department of public health sciences, found that those who undertook annual mammograms and physical exams did not experience fewer deaths than the women who just had annual exams.

U.S. doctors have questioned the findings, saying the quality of the mammograms used in the study were poor — the X-rays did not detect sufficient breast cancers and therefore biased the results.

The study involved 15 screening centers in six Canadian provinces and will be published in the forthcoming Journal of the National Cancer Institute. The women had mammograms between January 1980 and March 1985 and their health was followed for an average of 13 years and up to 16 years.

No Difference in Death Rates

During the screening and subsequent follow-up period, doctors detected 622 breast cancers in the mammography and physical examination group, compared with 610 in the physical examination group. Between 1980 and 1993, 88 women died from breast cancer in the mammogram and exam group, while 80 women died in the exam-only group.

The mammography and physical exam group detected more lymph-node negative, or less advanced cancers, and more small breast cancers, the researchers said.

“The results show that mammography was not associated with a reduction in breast cancer deaths even though mammography detected smaller cancers than physical examination did,” the researchers reported.

They suggest that for women older than 50, thorough annual physical breast exams plus the teaching of breast self-examination may be a useful alternative to mammography. But the doctors emphasize that physical examinations for routine screening involve far more time and skilled attention to relatively minor signs than more cursory ones often done in seconds in doctor’s offices.

Critics: Poor Mammograms and Bias

But American doctors say the study is flawed. Dr. Daniel Kopans, director of breast imaging for the Massachusetts General Hospital, in Boston, says the patients were not randomly assigned to the different groups and the mammography performed was below par.

“The quality of the mammography was terrible,” Kopans said. “Even the reference physicist quoted in the [Journal] that the mammograms were below average for that period of time.”

Dr. Robert Smith, director of cancer screening for the American Cancer Society, said all other studies of mammograms show an increase in survival. All studies show that mammograms detect earlier stage tumors that can be treated earlier and reduce deaths.

He agreed the Canadian mammograms were of low quality and the patient selection might have been biased.

Between the poor mammogram quality and the fact that there was a greater likelihood someone with more advanced cancer was put in the mammogram group, there was a higher death rate in the mammogram group than there should have been, Smith explained.

But Smith did acknowledge that a good clinical exam may be of value, too.The American Cancer Society recommends that all women over the age of 40 get an annual mammogram and a clinical physical examination. The death rate from cancer has been decreasing since 1989, with the average percentage each year decreasing by 2 percent.

The peer-reviewed Journal of the American Cancer Institute has been sold to Oxford Press and is independent of the National Cancer Institute and its research.

Mary Harris of ABCNEWS contributed to this story.