Chicago: A team of U.S. researchers on Tuesday revealed that mammograms’ accuracy varies widely. According to the researchers conducting a study on the subject, the women with lumps in their breasts rely on their radiologists to accurately read their mammograms, but the accuracy of those readings varies widely.

Diana Miglioretti, the lead researcher at the Group Health Center for Health Studies in Seattle, told that the variation in the quality of screening mammograms have also been revealed by earlier studies, but the recent study found inconsistencies even when a lump was present, leaving some women open to false positive results or even missed diagnoses.

In the study published in the Journal of the National Cancer Institute, Miglioretti and her team of researchers studied 123 radiologists who looked at 36,000 diagnostic mammograms from 1996 through 2003 at 72 U.S. facilities, including six from Group Health, a nonprofit health maintenance organization in Washington. The study also included data from rural hospitals, where radiologists may not handle a high volume of diagnostic mammograms.

Funded by grants from the National Cancer Institute, Agency for Healthcare Research and Quality and Breast Cancer Surveillance Consortium, the study found that sensitivity -- the ability to accurately detect cancer -- ranged from 27 percent to 100 percent. False positives ranged from 0 to 16 percent.

The results surprised Miglioretti.

Miglioretti said, “On average, 21 percent of breast cancers were missed and 4.3 percent of women underwent a biopsy even though they didn't have breast cancer.”

This variability was evident even after accounting for differences in physical characteristics.

She said, “I was expecting diagnostic mammograms to be less variable than screening mammograms.”

According to Miglioretti, women might want to consider scheduling diagnostic exams at centers with breast imaging experts. She said, “It might be worth driving the extra hour to find one. It is not always possible.”

Diagnostic mammograms are those performed on women whose initial, screening mammograms led to breast concerns. According to an estimate, there will be 178,000 new cases of breast cancer, with 40,000 deaths in the United States, in 2008.