Dr. Ellen B. Mendelson, a 52-year-old radiologist, had her first mammogram when she was 35. She had another one when she was 38. Since age 40, she has had routine screening mammograms every year, and she passionately urges women in their 40s to do the same.
Dr. Mendelson, who directs the Breast Diagnostic Imaging Center of the Western Pennsylvania Hospital in Pittsburgh, says she knows four radiologists who had breast cancer that was found after routine mammograms when they were in their late 30s or 40s. Mammography saved them, she says. "It seems really immoral to me to remove this potentially lifesaving benefit from women who are right at the height of their careers and family lives," she says.
Dr. Barbara Rimer, the 45-year-old director of cancer prevention and research at the Duke University School of Medicine, used to agree with Dr. Mendelson. She had a screening mammogram two years ago and has written papers on the need for women in their 40s to have such mammograms. But now, after immersing herself in the world's data on mammography, she has decided to wait until she is 50 to have her next mammogram.
"There was something transforming about reading all those articles together," Dr. Rimer says. "The science almost called out to you with the incredible weight of the evidence." And, she concluded, the science convinced her that routine mammograms before age 50 do not save lives.
The question of when to start having regular mammograms has never been more perplexing or more disturbing. After years of recommending that women start having them every year or two at age 40, the National Cancer Institute in Bethesda has now changed its stance, saying instead that their benefits for women under age 50 are not documented and that women should look at the facts and make up their own minds. But, as the opposite views of Dr. Mendelson and Dr. Rimer illustrate, this emotion-laden issue can elicit very different responses.
Dr. Edward Sondik, the deputy director of the Division of Cancer Prevention and Control at the National Cancer Institute, says he has seen "women who understand the facts and who could not be more convinced and passionate that screening in the 40s is the right thing to do." But he has also seen "others who feel exactly the opposite, and they also understand the facts."
"My view is that if a woman is informed at age 40 and then makes a decision about mammograms, whatever she decides is fine," he says. But he adds, "If she makes a decision on a rote basis, without understanding that there may be harms that go along with benefits that have not been proven, then she's making a mistake."
Anticipation of regret
Dr. Ann Flood, director of policy studies at the Center for Evaluative Clinical Sciences at Dartmouth Medical School in Hanover, N.H., agrees. She says the most important thing is that women in their 40s be informed of the drawbacks of mammography as well as the possible benefits.
Dr. Flood says that often the most compelling factor is whether a woman feels she may regret her choice. That means, she says, that even though a woman in her 40s might think that not having a mammogram is the more rational decision, she also might feel she could not live with herself if she made that choice and then developed cancer before she was 50. The question, Dr. Flood says, is, "would you feel guilty and would you say, 'My life could have been saved?' A lot of people feel compelled by the anticipation of regret."
The basic facts are not in dispute. Eight studies over the last 30 years have consistently found that routine mammograms for women over the age of 50 reduce the death rate from breast cancer by 25 to 30 percent. But none of these studies has found a similar effect in younger women. And even when all the world's studies are lumped together and the 173,000 women in their 40s who participated are considered as one group, no benefit has emerged.
This may reflect the fact that most women survive breast cancer, making it difficult to show that mammography saves lives at any age.
The cancer institute says that more than 79 percent of women with breast cancer are still alive five years later. Since younger women are less likely to get breast cancer than older ones, the breast cancer death rate becomes so small that it is difficult to show that screening them with mammography saves their lives.
Another difficulty is that mammography is a less efficient tool to examine younger women. Researchers find that mammograms miss as many as 40 percent of cancers in women in their 40s but just 14 percent of cancers in older women.
But while they have failed to show that mammography saves lives of women in their 40s, researchers have found some drawbacks associated with it. Dr. Sondik says that in about half of all women between 40 and 50 who have annual mammograms a suspicious lesion that turns out not to be cancer will be detected. About 5 percent will be found to have an invasive cancer, he says.
Dr. Rimer says that months after a breast cancer false alarm, women show signs much like those of post-traumatic stress syndrome. "It is not a benign experience," she says.
Another drawback, says Dr. H. Gilbert Welsh, a senior research associate at the Department of Veterans Affairs in White River Junction, Vt., is that in 40 percent of the breasts removed on autopsy from women in their 40s who died from causes other than cancer, microscopic lesions in the milk ducts are found; it is unknown which of these common lesions develop into cancer. The lesions sometimes show up on mammograms and few if any doctors or women would ignore them. The standard treatment, however paradoxical it may seem, is a mastectomy.
The dispute over mammography in younger women, however, is not over the facts; it is over how to interpret them. The studies do not prove that mammograms fail to help younger women. Instead, they show that if there is a benefit, the maximum it could be is about a 20 percent reduction in the death rate from breast cancer.
However, the same studies, taking into account margins of error, also show that the numbers could be read to show that screening actually increases the death rate, again by about 20 percent. The most likely effect, however, is that mammograms neither helped nor harmed women in their 40s.
The death rate
In 1990, the death rate from breast cancer for women in their 40s was about 30 per 100,000, meaning that if there were a 20 percent reduction it would save 6 lives per 100,000. If there were a greater effect, the studies would have shown it, says Dr. Peter Greenwald, director of the Division of Cancer Prevention and Control at the cancer institute.
Dr. Mendelson and others, like Dr. Gerald Dodd, a radiologist at the M.D. Anderson Cancer Center in Houston, are convinced that despite the statistical data, routine mammograms can save the lives of women in their 40s. If the studies did not show this, they say, it is because they are flawed.
For example, says Dr. Dodd, a past president of the American Cancer Society, researchers may simply have to wait longer to see a benefit of mammography in younger women. Most studies of younger women so far followed them for no more than 12 years, but it may take 18 years or longer to see a reduction in the death rate from breast cancer, Dr. Dodd says.
Another drawback to the studies, critics say, is that they used outdated technology. "There is a significant difference in the quality of mammograms and in the quality of interpretation now as compared to even 10 years ago, and most of these studies took place 7 to 10 years ago," Dr. Dodd says.