Radiologists who examine larger numbers of mammograms have fewer false-positives, according to a new study, leading researchers to call for changes to certification requirements.

The study, touted as one of the largest of its kind, was undertaken in response to a call from the Institute of Medicine to see if interpretive volume affected performance in screening mammography.

In the study, the researchers analyzed almost 780,000 screening mammograms from 120 radiologists from registries in California, North Carolina, New Hampshire, Vermont, Washington and New Mexico, over five years.

The findings weren't entirely intuitive: reading more mammograms a year didn't help radiologists find more cancers. Rather, it helped them become better at ruling out cancer.

"[R]adiologists with higher 'interpretive volumes' could identify the same number of cancers, while making fewer women come in for extra tests that showed they did not have cancer," said study leader Diana S.M. Buist, a senior investigator at Group Health Research Institute in Seattle.

Around 22 women were called back for more tests for every one cancer found, the researchers said.

False-positives from mammograms -- where cancer appears to be, but is not, present -- lead to biopsies and other follow-up tests, which can cause anxiety and discomfort to women and which cost the health care system $1.6 billion every year, the researchers said.

The findings have convinced the researchers that mammographers should be required to read more studies than they currently do. Currently, the U.S. Food and Drug Administration requires radiologists who interpret screening mammograms to read 960 every two years. But Buist said based on models her team developed, raising volume requirements to 1,000 per year would lower the number of women with false-positive workups by more than 71,000. And raising it to a 1,500 annual minimum would lower it by more than 117,000, without affecting radiologists' ability to detect cancers.

However, the researchers acknowledged that raising volume requirements could drive low-volume mammographers out of the business. According to their research, more than one-third of radiologists in the study interpreted fewer than 1,500 mammograms each year.

Plus, the pool of radiologists who read mammograms could be shrinking as the doctors age and retire, the researchers said. In their study, the median age of a radiologist was 54.

"Unlike the mammography debate about whether women in their 40s should be screened, which is based on the weight of harms of false positives, the tradeoff around volume policy will concern workforce issues and reporting requirements that would necessitate changes to how the FDA collects information on how many mammograms radiologists interpret," Buist said.

The results are published in the April issue of Radiology.

More Industry Headlines