Although mammography saves lives, it is not perfect. Depending on the study, the sensitivity of screening mammography ranges from 83 to 95 percent, which means that as many as 17 percent of cancer cases may go undetected by mammography. As many as three-quarters of the breast lesions that are biopsied in the United States as a result of a suspicious mammogram are benign.9,11 More effective approaches to the early detection and diagnosis of breast cancer would go a long way toward improving the care of women concerned about their risk of breast cancer—both by reducing the number of false alarms and unnecessary biopsies and by decreasing the number of cases that go undetected.
There are several potential ways to improve detection of breast cancer: more widespread use of mammography, better quality mammography, or development of new technologies. Of these three, greater use of mammography, as it now exists, even though it remains an imperfect screening technology, would likely save the most lives in the short run.
Although a number of new technologies are poised to expand the suite of current options, most advances are likely to be incremental improvements in existing technologies. The most significant technology changes to be adopted in clinical practice since 2001, when the Institute of Medicine and the National Research Council’s Mammography and Beyond report was published, have been improvements in existing technology. Four new digital mammography systemsa and three new systems for computer-aided detectionb were approved by the Food and Drug Administration and are all on the market now. Thus far, the accuracy of digital mammography has been shown to be equivalent to, but not superior to, traditional film-screen mammography, although clinical studies are still under way.14
There have also been changes relevant to breast cancer detection that are only indirectly related to technology developments. First, the sense of crisis concerning the shortage of breast imagers has deepened, and mammography facilities continue to close. Second, the Health Insurance Portability and Accountability Act Privacy Rule took effect in April 2003 and is a source of great concern to those conducting certain types of clinical studies. Third, the extent to which ductal carcinoma in situ (DCIS) cases are overtreated remains unclear, but the most recent data suggest that until it is