Radiologists with recent training best at detecting breast cancer on mammograms

WASHINGTON (AP) _ Practice may not make perfect when it comes to spotting cancer on mammograms. <br><br>A new study casts doubt on the belief that doctors who read the most mammograms each year are the

Tuesday, February 18th 2003, 12:00 am

By: News On 6

WASHINGTON (AP) _ Practice may not make perfect when it comes to spotting cancer on mammograms.

A new study casts doubt on the belief that doctors who read the most mammograms each year are the most proficient. Instead, the research found the most recently trained radiologists did best in a test of cancer-detection accuracy.

Other factors that may give a more reliable mammogram: using a center that requires two radiologists to read each X-ray, and that performs more sophisticated breast-imaging procedures as well as routine mammograms, concludes the study published in the Journal of the National Cancer Institute.

It's counterintuitive that a doctor fresh out of school would do a better job than a seasoned veteran.

Indeed, the study is surprising and conflicts with other research, cautioned Dr. Joann Elmore of the University of Washington, in an accompanying editorial.

Mammograms are considered the best tool available for spotting breast cancer early, when it's most treatable. But they're not perfect. One reason is that they require someone to decide that an often blurry spot on the X-ray is suspicious enough to biopsy _ without subjecting too many women to unnecessary false alarms.

Federal regulations require that U.S. radiologists read 480 mammograms a year for certification _ although many read many more _ while British doctors are required to read 5,000 a year, Elmore wrote.

A year ago, California researchers compared British radiologists with U.S. counterparts who read varying numbers of mammograms. That study concluded doctors who performed the most mammograms found more cancer with fewer false alarms.

And other studies have found that young radiologists have higher rates of unnecessary biopsies than older colleagues.

In the new study, Craig Beam of the University of South Florida asked 110 radiologists to examine the mammograms of 148 women, 43 percent of whom had breast cancer.

How many mammograms the radiologists had read the previous year had no impact, Beam reported. But he found a small but significant drop in cancer detection for each year beyond a doctor's residency training, plus better accuracy among radiologists who practiced in the more sophisticated centers.

Beam acknowledges the study is small and doesn't prove mammogram volume plays no role. After all, some highly accurate veterans may not have participated. But he said reading the X-rays is a skill that must be maintained, and he questioned whether radiologists have enough continuing education.

Elmore noted the study has some flaws. Because it was just a test, participants may have been less cautious in calling slightly suspicious spots cancer _ yielding a cancer-detection rate of 91 percent, higher than the nation's average of 75 percent.

Probably a combination of factors _ expertise from reading lots of mammograms plus continuing education to keep skills sharp _ ultimately will prove important for accuracy, Elmore said. She and Beam agreed more research is needed to figure that out.

Meanwhile, how should women pick a mammogram center?

Some advocacy groups advise asking if your center does double-reads or uses new computer technology to double-check readings. Elmore advises going to the same facility each year, or at least making sure prior X-rays are available for a new radiologist to compare _ it helps in spotting subtle changes.