BACKGROUND: The performance of diagnostic mammography for women with signs or symptoms of breast cancer has not been well studied. We evaluated whether age, breast density, self-reported breast lump, and previous mammography influence the performance of diagnostic mammography. METHODS: From January 1996 through March 1998, prospective diagnostic mammography data from women aged 25-89 years with no previous breast cancer were linked to cancer outcomes data in six mammography registries participating in the Breast Cancer Surveillance Consortium. We used the final mammographic assessment at the end of the imaging work-up to determine abnormal mammographic examination rate, positive predictive value (PPV), sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve. We used age, breast density, prior mammogram, and self-reported breast lump jointly as predictors of performance. All statistical tests were two-sided. RESULTS: Of 41 427 diagnostic mammograms, 6279 (15.2%) were judged abnormal. The overall PPV was 21.8%, sensitivity was 85.8%, and specificity was 87.7%. Multivariate analysis showed that sensitivity and specificity generally declined as breast density increased (P =.007 and P<.001, respectively), that previous mammography decreased sensitivity (odds ratio [OR] = 0.52, 95% confidence interval [CI] = 0.36 to 0.74; P<.001) but increased specificity (OR = 1.43, 95% CI = 1.31 to 1.57; P<.001), and that a self-reported breast lump increased sensitivity (OR = 1.64, 95% CI = 1.13 to 2.38; P =.013) but decreased specificity (OR = 0.54, 95% CI = 0.49 to 0.59; P<.001). ROC analysis showed that higher breast density and previous mammography were negatively related to accuracy (P<.001 for both). CONCLUSIONS: Diagnostic mammography in women with signs or symptoms of breast cancer shows higher sensitivity and lower specificity than screening mammography does. Higher breast density and previous mammographic examination appear to impair performance.
BACKGROUND: The performance of diagnostic mammography for women with signs or symptoms of breast cancer has not been well studied. We evaluated whether age, breast density, self-reported breast lump, and previous mammography influence the performance of diagnostic mammography. METHODS: From January 1996 through March 1998, prospective diagnostic mammography data from women aged 25-89 years with no previous breast cancer were linked to cancer outcomes data in six mammography registries participating in the Breast Cancer Surveillance Consortium. We used the final mammographic assessment at the end of the imaging work-up to determine abnormal mammographic examination rate, positive predictive value (PPV), sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve. We used age, breast density, prior mammogram, and self-reported breast lump jointly as predictors of performance. All statistical tests were two-sided. RESULTS: Of 41 427 diagnostic mammograms, 6279 (15.2%) were judged abnormal. The overall PPV was 21.8%, sensitivity was 85.8%, and specificity was 87.7%. Multivariate analysis showed that sensitivity and specificity generally declined as breast density increased (P =.007 and P<.001, respectively), that previous mammography decreased sensitivity (odds ratio [OR] = 0.52, 95% confidence interval [CI] = 0.36 to 0.74; P<.001) but increased specificity (OR = 1.43, 95% CI = 1.31 to 1.57; P<.001), and that a self-reported breast lump increased sensitivity (OR = 1.64, 95% CI = 1.13 to 2.38; P =.013) but decreased specificity (OR = 0.54, 95% CI = 0.49 to 0.59; P<.001). ROC analysis showed that higher breast density and previous mammography were negatively related to accuracy (P<.001 for both). CONCLUSIONS: Diagnostic mammography in women with signs or symptoms of breast cancer shows higher sensitivity and lower specificity than screening mammography does. Higher breast density and previous mammographic examination appear to impair performance.
Authors: Sebastien Haneuse; Diana S M Buist; Diana L Miglioretti; Melissa L Anderson; Patricia A Carney; Tracy Onega; Berta M Geller; Karla Kerlikowske; Robert D Rosenberg; Bonnie C Yankaskas; Joann G Elmore; Stephen H Taplin; Robert A Smith; Edward A Sickles Journal: Radiology Date: 2011-11-21 Impact factor: 11.105
Authors: William E Barlow; Chen Chi; Patricia A Carney; Stephen H Taplin; Carl D'Orsi; Gary Cutter; R Edward Hendrick; Joann G Elmore Journal: J Natl Cancer Inst Date: 2004-12-15 Impact factor: 13.506
Authors: Monique D Dorrius; Erik F J de Vries; Riemer H J A Slart; Andor W J M Glaudemans Journal: Eur J Nucl Med Mol Imaging Date: 2015-03-12 Impact factor: 9.236
Authors: Diana L Miglioretti; Rebecca Smith-Bindman; Linn Abraham; R James Brenner; Patricia A Carney; Erin J Aiello Bowles; Diana S M Buist; Joann G Elmore Journal: J Natl Cancer Inst Date: 2007-12-11 Impact factor: 13.506