The volume of mammograms read by individual breast imagers is likely to be important—if not directly, perhaps as a proxy for other characteristics such as advanced training, specialization in breast imaging, or working in an organization with a mammography quality improvement program.
One study reported that compared to general radiologists, breast imaging specialists detect more cancers, recommend more biopsies, and have lower recall rates. In general, the specialists, who interpreted more than 5,000 mammograms per year, found two to three more cancers than general radiologists for every 1,000 mammograms.11
However, only 12 percent of radiologists interpreting mammograms are specialists,23 and most women do not see specialists. The false positive rates of community radiologists (those who work outside academic research centers) are quite variable. One study reported rates of false positives that ranged from 3 to 16 percent.35
Accuracy depends on context. Rates of false positives in the United States have increased over the years, and they vary among countries and health care systems. Defensive medicine is widely presumed to be prevalent in the United States, especially in mammography. Leonard Berlin testified on behalf of the ACR in Congress that malpractice suits in the United States are decided in favor of plaintiffs so often that many radiologists do not attempt to contest even seemingly frivolous cases.14
Rates of false positives in the Unites States nearly doubled from 1985 to 1993, from roughly 5 to 10 percent.35 This increase parallels the steadily increasing rates of malpractice suits related to failures to detect breast cancer through mammography, which is often proposed as a driving force in rates of false positives. Radiologists in the United States may be practicing more defensive medicine because they fear malpractice suits, which their counterparts in the United Kingdom face to a much lesser degree.34 Although British radiologists also report that they worry about malpractice, the scope of the problem is considerably less than it is in the United States. Only about 25 percent of British practices admitted to being sued over breast cancer, with nearly all cases being dropped.89 Many believe that concern over malpractice is an important factor in the relatively high rate of false-positive results in the United States, and anecdotal evidence supports this view. However, there are no reliable data to measure the extent of this problem, and such sensitive data would be difficult to obtain.