Table 3.
Benefits and harms of the USPSTF criteria and selected LCDRAT model screening strategies (screening age 55–80 y) compared with no screening
| Strategy description | Corresponding risk threshold, % | Percentage ever screened, % (CMR†) | Number of CT screens per 100 000 (CMR) | Lung cancer deaths prevented per 100 000 (CMR) | Lung cancer mortality reduction, % (CMR) | Life-years gained per 100 000 (CMR) | Life-years gained per lung cancer death prevented (CMR) | Number of overdiagnosed lung cancers per 100 000 (CMR) | Percentage of screen-detected cases % (CMR) | Average number of screens per lung cancer death avoided (CMR) | Average number of screens per life year gained (CMR) | Average number of screens per person screened (CMR) | Average age at first screening (CMR), y |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| USPSTF criteria | USPSTF criteria | 19.9 (18.7–21.2) | 326 608 (301 659–337 726) | 613 (337–865) | 10.8 (6.7–16.9) | 8590 (4665–11 922) | 14.0 (13.0–15.3) | 115 (49–156) | 7.3 (5.3–10.4) | 533 (385–1003) | 38 (28–72) | 16 (16–18) | 55.6 (55.0–56.0) |
| Similar proportion of individuals selected as the USPSTF criteria in the PLCO control arm | 0.96 | 33.4 (31.6–35.1) | 501 689 (468 648–517 665) | 844 (493–1 151) | 14.9 (9.8–22.5) | 11 018 (6298–14 955) | 13.0 (12.4–14.3) | 168 (77–220) | 7.7 (5.6–10.6) | 594 (450–1041) | 46 (34–81) | 15 (14–16) | 62.9 (62.3–64.4) |
| Similar sensitivity as the USPSTF criteria in the PLCO control arm | 1.34 | 26.4 (24.9–27.8) | 394 238 (368 077–406 682) | 766 (418–1 060) | 13.5 (8.3–20.7) | 9734 (5145–13 380) | 12.7 (12.1–13.9) | 157 (68–210) | 7.8 (5.8–10.4) | 515 (384–968) | 41 (30–79) | 15 (14–16) | 62.5 (62.1–63.9) |
| Similar CT screens required as the USPSTF criteria | 1.70 | 23.6 (22.3–24.9) | 325 056 (303 273–336 570) | 696 (366–977) | 12.3 (7.3–19.1) | 8631 (4365–11 838) | 12.4 (11.8–13.5) | 150 (60–201) | 7.8 (6.0–10.0) | 467 (343–917) | 38 (27–77) | 14 (13–15) | 63.7 (63.2–65.0) |
| Similar lung cancer deaths averted as the USPSTF criteria | 2.10 | 20.9 (19.8–22.1) | 259 582 (242 052–270 053) | 619 (310–885) | 10.9 (6.1–17.3) | 7438 (3574–10 184) | 12.0 (11.4–13.1) | 140 (53–190) | 7.9 (6.2–9.8) | 419 (302–868) | 35 (25–75) | 12 (12–14) | 65.2 (64.6–66.5) |
| Similar life-years gained as the USPSTF criteria | 1.70 | 23.6 (22.3–24.9) | 325 056 (303 273–336 570) | 696 (366–977) | 12.3 (7.3–19.1) | 8631 (4365–11 838) | 12.4 (11.8–13.5) | 150 (60–201) | 7.8 (6.0–10.0) | 467 (343–917) | 38 (27–77) | 14 (13–15) | 63.7 (63.2–65.0) |
Results are per 100 000 individuals alive at age 45 years. Lung cancer incidence in the no-screening strategy group was 7116 (6518–8450) per 100 000 persons. Lung cancer mortality was 5670 (5010–7114) per 100 000 persons.
All results were summarized as the mean across the four CISNET models. The numbers in parentheses denote the lower and upper range of the results across the four CISNET models.
CISNET = Cancer Intervention and Surveillance Modeling Network; CMR = CISNET model range; CT = computed tomography; LCDRAT = Lung Cancer Death Risk Assessment Tool; PLCO = Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial; USPSTF = United States Preventive Services Task Force.