Table 2.
Predictive accuracies of ARRS.
| Category | ARRS≥2 | ARRS≥8 | ||||
|---|---|---|---|---|---|---|
| Data (95% CI) | I2(%) or Z* | p | Data (95% CI) | I2(%) or Z* | p | |
| SE % | 0.98 (0.94–0.99) | 10.1 | 0.35 | 0.58 (0.51–0.65) | 51.15 | 0.020 |
| SP % | 0.30 (0.22–0.39) | 89.64 | <0.001 | 0.86 (0.81–0.89) | 79.9 | <0.001 |
| PLR | 1.42 (1.26–1.61) | 86.3 | <0.001 | 3.81 (2.88–5.05) | 63.6 | 0.001 |
| NLR | 0.13 (0.08–0.20) | 0 | 0.952 | 0.51 (0.43–0.61) | 51.7 | 0.019 |
| DOR | 15.08 (8.87–25.63) | 0 | 0.897 | 7.59 (5.82–9.90) | 43.7 | 0.052 |
| AUROC | 0.82 (0.78–0.85) | – | – | 0.77 (0.73–0.80) | – | – |
| Beta# | −0.28 (−1.19–0.64) | −0.6 | 0.552 | 0.41 (−0.54–1.37) | 0.85 | 0.394 |
| Lambda# | 2.23 (0.76–3.71) | – | – | 1.84 (1.22–2.45) | – | – |
ARRS, ANCA renal risk score; SE, sensitivity; SP, specificity; PLR, positive likelihood ratio; NLR, negative likelihood ratio; DOR, diagnostic odds ratio; AUROC, area under the receiver operating characteristic curve; CI, confidence interval.
#
From hierarchical summary receiver operating curves (HSROC) model.
*
Z value only for Beta row, I2 value for rest.