Table 1. Examples of risk prediction models for asymptomatic individuals that have been validated in different populations.
|
Summary of performance in validation studies |
||
|---|---|---|
| Risk model | Discrimination (AUROC, 95% CI) | Calibration (O/E, 95% CI) |
|
Breast (Meads et al, 2012) | ||
| Colditz | 0.63 (0.63–0.64) | 1.01 (0.94–1.09) |
| Gail 2 | 0.63 (0.59–0.67) | 0.95 (0.88–1.01) |
| Rosner and Colditz | 0.57 (0.55–0.59) | 0.96 (0.92–1.02) |
| Tyrer and Cusick | 0.762 (0.70–0.82)a | 1.09 (0.85–1.41)a |
|
Lung (Li et al, 2015) | ||
| Bach | 0.81 (0.76–0.86) | 0.88 (0.72–1.08) |
| Sptiz | 0.78 (0.73–0.83) | 3.75 (3.06–4.60) |
| LLP | 0.79 (0.73–0.83) | 1.12 (0.92–1.37) |
| PLCOM2012b | 0.81 (0.76–0.86) | 1.03 (0.87–1.23) |
|
Colorectal (Win et al, 2012) | ||
| Harvard Cancer Risk Index | Men: 0.71 (0.68–0.74) Women: 0.67 (0.64–0.70) | — |
| Imperiale | 0.74 (SD=0.06) | — |
| Freedman | Men: 0.61 (0.60–0.62) Women: 0.61 (0.59–0.62) | Men: 0.99 (0.95–1.04) Women: 1.05 (0.98–1.11) |
| Ma | 0.64 (0.61–0.67) | 1.09 (0.98–1.23) |
|
Prostate (Louie et al, 2014) | ||
| Prostataclass | 0.79 (0.75–0.84) | — |
| Finne | 0.74 (0.70–0.77) | — |
| Karakiewcz | 0.74 (0.69–0.80) | — |
| Prostate Cancer Prevention Trial | 0.66 (0.63–0.68) | — |
| Chun | 0.76 (0.72–0.79) | — |
| ERSPC RC3c | 0.79 (0.77–0.81) | — |
|
Melanoma (Usher-Smith et al, 2014) | ||
| Fortes | 0.79 (0.70–0.86) | — |
| Williams | 0.70 (0.64–0.77) | |
Abbreviations: AUROC=area under receiver operator characteristic curve; CI=confidence interval; LLP, liverpool lung project; O/E=ratio of observed to expected events.
a
In a high-risk sample.
b
The 2012 version of the prostate, lung, colorectal, and ovarian cancer screening trial model.
c
European Randomized Study of Screening for Prostate Cancer Risk Calculator 3.