Mammograms Every Other Year OK for Women Over 50?
Researchers found similar risk for advanced disease as with yearly screening, but some experts disagree
WebMD News Archive
The task force guidelines are at odds with those of many other organizations, including the American Cancer Society, which recommends annual screening beginning at age 40.
The task force guidelines consider only a woman's age, according to Kerlikowske. Her team decided to study the benefits and harms of screening based not just on age but also on breast density and hormone therapy use.
Women in the study were 40 to 74. Most diagnosed with breast cancer during the years studied, 1994 to 2008, were 50 or older. They typically had dense or very dense breasts.
The study has flaws, said Dr. Daniel Kopans, a professor of radiology at Harvard Medical School and senior radiologist in the breast imaging division of Massachusetts General Hospital. He is also a member of the American College of Radiology's Breast Imaging Commission.
The groups weren't identical, he said, and that would have been the best way to study the issue. Information is lacking, too, he said, on why some women got screened annually and others did not. "Those screened every year may have had different risk factors," he said.
"I would tell women it makes sense to get screened every year," Kopans said.
In a statement, the American College of Radiology (ACR) pointed to an analysis published in 2011 in the American Journal of Roentgenology finding that under the biennial model, about 6,500 more women annually in the United States would die of breast cancer.
Looking at early versus late-stage cancer is not the best way to judge the best interval for mammograms, according to the ACR statement. Rather, it said, researchers should look at such factors as tumor size and other markers of detecting cancers early.
Dr. Robert Smith, senior director of cancer screening for the American Cancer Society, said that the study is "not an accurate look at one year versus two years." For the study, annual was defined as intervals of nine to 18 months, for instance, and two years as more than 18 to 30 months.
Kerlikowske, in response, said this interval variability reflects real life.