Skip to main content
. 2009 Mar 3;100(6):901–907. doi: 10.1038/sj.bjc.6604954

Table 2. Breast cancers and tumour characteristics at different reading strategies.

Reading strategy Single radiologist reading Radiologist double-reading Single radiologist reading and radiographer double-reading with referral of all positive readings Radiologist double-reading followed by radiologist review of positive cases at radiographer double-reading Double-reading by radiologists and radiographers with referral of all positive readings
Referral rate, % (95% CI) 1.24 (1.17–1.31) 1.36 (1.30–1.43) 1.96 (1.87–2.04) 1.48 (1.41–1.55) 2.04 (1.95–2.12)
Mammographic abnormality, no. (%) 1315 1448 2076 1570 2161
Density 877 (66.7) 967 (66.8) 1386 (66.8) 1041 (66.3) 1448 (67.0)
Microcalcifications 257 (19.5) 289 (20.0) 447 (21.5) 331 (21.1) 464 (21.5)
Density with microcalcifications 106 (8.1) 114 (7.9) 140 (6.7) 116 (7.4) 145 (6.7)
Architectural distortion 44 (3.3) 47 (3.2) 62 (3.0) 51 (3.2) 64 (3.0)
Breast parenchyma asymmetry 31 (2.4) 31 (2.1) 41 (2.0) 31 (2.0) 40 (1.9)
Breast cancers, no. 492 528 579 564 592
CDR, per 1,000 women (95% CI) 4.64 (4.23–5.05) 4.98 (4.55–5.40) 5.46 (5.01–5.90) 5.32 (4.88–5.75) 5.58 (5.13–6.03)
Sensitivity, % (95% CI) 63.9 (60.5–67.3) 68.6 (65.3–71.9) 75.2 (72.1–78.2) 73.2 (70.1–76.4) 76.9 (73.9–79.9)
Specificity, % (95% CI) 99.2 (99.2–99.3) 99.1 (99.1–99.2) 98.6 (98.5–98.7) 99.0 (99.0–99.1) 98.5 (98.4–98.6)
PPV of referral, % (95% CI) 37.4 (34.8–40.0) 36.5 (34.0–38.9) 27.9 (26.0–29.8) 35.9 (33.6–38.3) 27.4 (25.5–29.3)
           
Type of breast cancer, no. (%)
 DCIS 80 (16.3) 90 (17.0) NA 98 (17.4) NA
 Invasive 412 (83.7) 438 (83.0) NA 466 (82.6) NA
 T1a–c 321 (77.9) 342 (78.1) NA 366 (78.5) NA
 T2 88 (21.4) 93 (21.2) NA 96 (20.6) NA
 Unknown 3 (0.7) 3 (0.7) NA 4 (0.9) NA

CI=confidence interval; CDR=cancer detection rate; DCIS=ductal carcinoma in situ; PPV=positive predictive value.

NA: several cancers were diagnosed as interval cancers or detected at subsequent screening. Consequently, exact tumour stages of these cancers at the time of the index screening examination are not available.