From: Methods for Evaluation of medical prediction Models, Tests And Biomarkers (MEMTAB) 2018 Symposium
 | Semi-structured Reference Standard N studies = 29 N participants = 6,725 N major depression = 924 | Fully structured Reference Standard N studies = 14 N participants = 7,680 N major depression = 839 | Difference across reference standards (Semi- Fully) | |||
---|---|---|---|---|---|---|
Cutoff | Sensitivity | Specificity | Sensitivity | Specificity | Sensitivity | Specificity |
5 | 0.98 | 0.55 | 0.93 | 0.54 | 0.05 | 0.01 |
6 | 0.98 | 0.63 | 0.91 | 0.61 | 0.07 | 0.02 |
7 | 0.98 | 0.69 | 0.86 | 0.69 | 0.12 | 0.00 |
8 | 0.95 | 0.75 | 0.82 | 0.75 | 0.13 | 0.00 |
9 | 0.91 | 0.80 | 0.74 | 0.79 | 0.17 | 0.01 |
10 | 0.88 | 0.85 | 0.70 | 0.84 | 0.18 | 0.01 |
11 | 0.84 | 0.89 | 0.62 | 0.87 | 0.22 | 0.02 |
12 | 0.79 | 0.91 | 0.57 | 0.89 | 0.22 | 0.02 |
13 | 0.70 | 0.93 | 0.49 | 0.92 | 0.21 | 0.01 |
14 | 0.64 | 0.95 | 0.44 | 0.94 | 0.20 | 0.01 |
15 | 0.56 | 0.96 | 0.35 | 0.96 | 0.21 | 0.00 |