According to a study in JAMA Internal Medicine, interpretation of mammograms with computer-aided detection (CAD) may be no better than without use of the technology. CAD is used for most screening mammograms in the United States at a cost of $400 million a year.
Researchers compared the accuracy of digital screening mammography that was interpreted with (n =495,818) versus without (n = 129,807) CAD from 2003 to 2009 among 323,973 women. The mammograms were interpreted by 271 radiologists across 66 facilities in the Breast Cancer Surveillance Consortium. Linking with tumor registries identified 3159 breast cancers in the women within 1 year of the screening.
Computer-aided detection did not improve screening performance by any assessed metric. Mammography sensitivity was 85.3% (95% CI, 83.6%-86.9%) with and 87.3% (95% CI, 84.5%-89.7%) without CAD. Specificity was 91.6% (95% CI, 91.0%-92.2%) with and 91.4% (95% CI, 90.6%-92.0%) without CAD. No difference was seen in the cancer detection rate (4.1 in 1000 women screened with and without CAD). Intraradiologist performance was not improved with CAD. Overall sensitivity was significantly decreased for mammograms interpreted with versus without CAD in the subset of radiologists who interpreted mammograms both with and without CAD (odds ratio, 0.53; 95% CI, 0.29-0.97).
The investigators concluded that CAD showed no improved diagnostic accuracy versus the traditional mammogram, suggesting that insurers are paying more for CAD, even though CAD provides no established benefit.