Table 5.
Test parameters for MPR cut-offs in detecting unfavourable viral outcomes
| MPR category | Number of individuals in category | Number of individuals with unfavourable outcomes | Sensitivity % (95% CI) | Specificity % (95% CI) | PPV % (95% CI) | NPV % (95% CI) | ROCa area (95% CI) |
|---|---|---|---|---|---|---|---|
| < 0.80 | 97 | 21 | 65.6 (52.3–77.3) | 12.2 (9.8–15.1) | 6.8 (4.9–9.2) | 78.4 (68.8–86.1) | 0.39 (0.33–0.45) |
| < 0.90 | 137 | 24 | 60.7 (47.3–72.9) | 18.2 (15.2–21.5) | 6.8 (4.9–9.2) | 82.5 (75.1–88.4) | 0.39 (0.33–0.46) |
| < 0.95 | 197 | 31 | 49.2 (36.1–36.1) | 26.7 (23.3–30.4) | 6.2 (4.2–8.7) | 84.3 (78.4–89.1) | 0.38 (0.31–0.45) |
| < 1.00 | 346 | 44 | 27.9 (17.1–40.8) | 48.6 (44.6–52.6) | 5.1 (3.0–8.0) | 87.3 (83.3–90.6) | 0.38 (0.32–0.44) |
| < 1.05 | 510 | 52 | 14.8 (7.0–26.2) | 73.8 (70.1–77.2) | 5.2 (2.4–9.7) | 89.8 (86.8–92.3) | 0.44 (0.39–0.49) |
aThe average of sensitivity and specificity for the binary MPR cut-point being tested
PPV Positive predictive value, NPV Negative predictive value, ROC receiver operating characteristic