The accuracy of mammograms carried out on women with symptoms of breast disease varies widely and is linked to the radiologist who reads the specialised X-ray, research published last night has found.
Researchers from a number of centres in the US looked at how well 123 radiologists interpreted 36,000 diagnostic mammograms, which were carried out to evaluate breast lumps.
The results, published in the Journal of the National Cancer Institute, found a wide variation in both false negative and false positive rates among the radiologists studied.
A person with a false positive result does not have cancer even though the mammography report suggests they do.
A woman with a false negative outcome has a cancerous lump but the mammogram is reported as normal.
Prof Tom Keane, the new director of the National Cancer Control Programme, told The Irish Times last week that the best that can be said to a woman with a normal mammogram is that the doctor is 95 per cent certain that she does not have breast cancer.
Emphasising the importance of the public understanding that even the best quality cancer testing will never be 100 per cent accurate, he said: "False positive and false negatives are a feature of all diagnostic testing. No technology is 100 per cent accurate - it is a subjective science."
Dr Diana Miglioretti and her colleagues from the Group Health Centre for Health Studies in Seattle found the radiologists who read diagnostic mammograms most accurately tended to be those who were based at academic medical centres or who spent at least 20 per cent of their time on breast work.
According to the authors of the US paper, ideally the ability of mammograms to accurately detect cancer (known as sensitivity) would be consistently high, with few false positive results. The accuracy should not depend on which radiologist was reading the mammogram or on the characteristics of the patients.
"Because the rate of breast cancer is 10-fold higher among diagnostic mammograms than among screening mammograms, and the majority of women with breast cancer have a physical sign or symptoms at the time of diagnosis, this variability in interpretative performance is concerning" they conclude.
Meanwhile, a report into the alleged misdiagnosis of women who underwent mammography and breast ultrasound at the Midland General Hospital in Portlaoise between November 2003 and August 2007 has been completed.
It has been sent for legal vetting prior to publication.