Table 4 Discriminatory performance measures for each of the risk models for 5 year risk of developing colorectal cancer in women
From: External validation of risk prediction models for incident colorectal cancer using UK Biobank
| Â | Colditz | Freedman | Guesmi | Johnson | QCancer10 | Tao | Wei Y-S | Wells | Driver a | Ma (simple) a | Ma (cox) a |
|---|---|---|---|---|---|---|---|---|---|---|---|
Total n | n=164 034 | n=130 188 | n=202 620 | n=203 390 | n=193 365 | n=189 097 | n=194 601 | n=191 475 | n=201 474 | n=174 297 | n=174 297 |
Incident CRC | n = 592 | n = 562 | n = 752 | n = 754 | n = 714 | n = 696 | n = 716 | n = 713 | n = 745 | n = 628 | n = 628 |
Top 10% | |||||||||||
Sensitivity | 12.3 | 16.5 | 11.2 | 19.8 | 19.0 | 18.0 | 11.0 | 18.9 | 17.4 | 19.6 | 19.3 |
Specificity | 90.0 | 90.0 | 90.0 | 90.0 | 90.0 | 90.0 | 90.0 | 90.0 | 90.0 | 90.0 | 90.0 |
LR+ | 1.23 | 1.66 | 1.12 | 1.00 | 1.91 | 1.80 | 1.10 | 1.90 | 1.75 | 1.97 | 1.93 |
LR− | 0.97 | 0.93 | 0.99 | 0.98 | 0.90 | 0.91 | 0.99 | 0.90 | 0.92 | 0.89 | 0.90 |
PPV (%) | 0.4 | 0.7 | 0.4 | 0.4 | 0.7 | 0.7 | 0.4 | 0.7 | 0.6 | 0.7 | 0.7 |
NPV (%) | 99.6 | 99.6 | 99.6 | 99.6 | 99.7 | 99.7 | 99.6 | 99.7 | 99.7 | 99.7 | 99.7 |
Top 20% | |||||||||||
Sensitivity | 21.5 | 30.1 | 28.3 | 22.9 | 36.1 | 33.9 | 19.3 | 33.1 | 30.9 | 31.2 | 32.6 |
Specificity | 80.0 | 80.0 | 80.0 | 80.0 | 80.1 | 80.1 | 80.0 | 80.0 | 80.0 | 80.0 | 80.0 |
LR+ | 1.07 | 1.51 | 1.42 | 1.15 | 1.81 | 1.70 | 1.01 | 1.66 | 1.55 | 1.56 | 1.64 |
LR− | 0.98 | 0.87 | 0.90 | 0.96 | 0.80 | 0.83 | 0.95 | 0.84 | 0.86 | 0.86 | 0.84 |
PPV (%) | 0.4 | 0.6 | 0.5 | 0.4 | 0.7 | 0.6 | 0.4 | 0.6 | 0.6 | 0.6 | 0.6 |
NPV (%) | 99.6 | 99.6 | 99.7 | 99.6 | 99.7 | 99.7 | 99.6 | 99.7 | 99.7 | 99.7 | 99.7 |
Top 80% | |||||||||||
Sensitivity | 80.2 | 86.1 | 92.7 | 76.5 | 93.6 | 91.4 | 78.2 | 93.1 | 91.9 | 93.0 | 92.0 |
Specificity | 20.0 | 20.0 | 20.0 | 20.0 | 20.1 | 20.0 | 20.0 | 20.0 | 20.0 | 20.0 | 20.0 |
LR+ | 1.00 | 1.08 | 1.16 | 0.96 | 1.17 | 1.14 | 0.98 | 1.16 | 1.15 | 1.16 | 1.15 |
LR− | 0.99 | 0.69 | 0.36 | 1.17 | 0.32 | 0.43 | 1.09 | 0.34 | 0.40 | 0.35 | 0.40 |
PPV (%) | 0.4 | 0.5 | 0.4 | 0.4 | 0.4 | 0.4 | 0.4 | 0.4 | 0.4 | 0.4 | 0.4 |
NPV (%) | 99.6 | 99.7 | 99.9 | 99.6 | 99.9 | 99.8 | 99.6 | 99.9 | 99.9 | 99.9 | 99.9 |
Top 90% | |||||||||||
Sensitivity | 90.4 | 94.0 | 97.2 | 88.2 | 97.1 | 95.1 | 88.8 | 96.9 | 95.7 | 96.7 | 95.7 |
Specificity | 10.0 | 10.0 | 10.0 | 10.0 | 10.0 | 10.0 | 10.0 | 10.0 | 10.0 | 10.0 | 10.0 |
LR+ | 1.00 | 1.04 | 1.08 | 0.98 | 1.08 | 1.06 | 0.99 | 1.08 | 1.06 | 1.07 | 1.06 |
LR− | 0.96 | 0.60 | 0.27 | 1.18 | 0.29 | 0.49 | 1.12 | 0.31 | 0.43 | 0.33 | 0.43 |
PPV (%) | 0.4 | 0.5 | 0.4 | 0.4 | 0.4 | 0.4 | 0.4 | 0.4 | 0.4 | 0.4 | 0.4 |
NPV (%) | 99.6 | 99.7 | 99.9 | 99.6 | 99.9 | 99.8 | 99.6 | 99.9 | 99.8 | 99.9 | 99.8 |
