Table 2.
Screening outcomes of using biennial DBT in a population screening program
| Sensitivity of DBT | 65% | 70% | 75% | 80% | 85% | 90% | 95% | 100% |
|---|---|---|---|---|---|---|---|---|
| DBT for dense breasts compared to DM (scenario 1–reference) | ||||||||
| N screen-detected | − 12 (0) | − 3 (0) | + 6 (0) | + 15 (0) | + 24 (1) | + 31 (1) | + 40 (1) | + 47 (1) |
| N interval | + 12 (0) | + 4 (0) | − 4 (0) | − 12 (0) | − 19 (0) | − 25 (0) | − 32 (1) | − 39 (1) |
| LYG | − 39 (2) | − 14 (2) | + 12 (2) | + 40(4) | + 63 (4) | + 88 (5) | + 115(5) | + 138(5) |
| Discounteda LYG | − 14 (1) | − 5 (1) | + 5(1) | + 19 (2) | + 30 (2) | + 42 (2) | + 54 (2) | + 65 (2) |
| Discountedb LYG | − 23 (1) | − 8 (1) | + 8 (2) | + 27 (3) | + 43 (3) | + 60 (3) | + 78 (3) | + 94 (3) |
| DBT for all women compared to DM (scenario 2–reference) | ||||||||
| N screen-detected | − 74 (1) | − 47 (1) | − 22 (1) | + 2(1) | + 26(1) | + 48 (0) | + 71 (0) | + 91 (1) |
| N interval | + 66 (1) | + 43 (1) | + 21 (0) | + 1(1) | − 19 (0) | − 38 (1) | − 57 (0) | − 74 (1) |
| LYG | − 219 (3) | − 146 (4) | − 76 (3) | − 5 (4) | + 64 (4) | + 129 (3) | + 195 (4) | + 254 (5) |
| Discounteda LYG | − 76 (1) | − 51 (1) | − 25 (1) | + 3 (2) | + 31 (2) | + 62 (3) | + 92 (2) | + 120 (3) |
| Discountedb LYG | − 126(2) | − 84(2) | − 43(2) | + 1 (3) | + 44 (3) | + 88 (4) | + 133 (3) | + 172 (4) |
| DBT for all women compared to DBT for dense breasts (scenario 2–scenario 1) | ||||||||
| N screen-detected | − 62(1) | − 44(1) | − 28(1) | − 13(0) | + 3(0) | + 17(1) | + 31(1) | + 44(1) |
| N interval | + 54(1) | + 39(0) | + 25(1) | + 12(0) | 0(0) | − 12(0) | − 24(1) | − 35(1) |
| LYG | − 180(3) | − 132(4) | − 88(3) | − 45(2) | + 1(1) | + 41(2) | + 80(3) | + 117(3) |
| Discounteda LYG | − 62(1) | −46(1) | − 30(1) | − 16(1) | + 1(0) | + 19(1) | + 38(1) | + 54(1) |
| Discountedb LYG | − 104(2) | − 76(2) | − 51(2) | − 26(1) | + 1(1) | + 28(1) | + 54(2) | + 79(2) |
Data shown as differences in number of DBT screen-detected tumours (N screen-detected) and number of interval tumours (N interval) for scenarios 1 and 2 with respect to the reference scenario of biennial DM screening for women 50–75 years of age, and for scenario 2 with respect to scenario 1. All data expressed as mean (SEs) per 10,000 women screened. Discountinga: 3% for both cost and LYG. Discountingb: 4% for cost and 1.5% for LYG
N = number; LYG = life years gained; DM = digital mammography; DBT = digital breast tomosynthesis.
Values in italics indicate the scenario ourperforms the comparison in the screening outcome