Table 3.
Predictive ability of risk stratification scores
| Sensitivity, % (95% CI) | Specificity, % (95% CI) | Positive predictive value (%) | Negative predictive value (%) | Area under the receiver operating curve (95% CI) | |
|---|---|---|---|---|---|
| All cases | |||||
| Existing risk stratification scores | |||||
| qSOFA ≥ 2 | 50 (25–75) | 87 (82–91) | 19 | 97 | 0.79 (0.69–0.89 |
| SIRS ≥ 2 | 88 (62–98) | 44 (38–51) | 9 | 98 | 0.74 (0.62–0.86) |
| MEWS ≥ 5 | 63 (35–85) | 80 (75–85) | 16 | 97 | 0.80 (0.71–0.90) |
| UVA ≥ 2 | 88 (62–98) | 82 (77–87) | 23 | 99 | 0.89 (0.78–1.0) |
| UVA high ≥ 5 | 50 (25–75) | 98 (96–99) | 62 | 97 | |
| Cases with an infection | |||||
| Existing risk stratification scores | |||||
| qSOFA ≥ 2 | 55 (23–83) | 82 (76–88) | 16 | 97 | 0.77 (0.63–0.91) |
| SIRS ≥ 2 | 82 (48–98) | 32 (26–40) | 7 | 97 | 0.70 (0.52–0.88) |
| MEWS ≥ 5 | 64 (31–89) | 72 (65–79) | 13 | 97 | 0.72 (0.58–0.87) |
| UVA ≥ 2 | 91 (59–100) | 78 (72–84) | 21 | 99 | 0.90 (0.78–1.0) |
| UVA high ≥ 5 | 55 (23–83) | 97 (94–99) | 55 | 97 | |
| Adapted risk stratification scores | |||||
| qSOFA (1) ≥ 1 | 82 (48–98) | 80 (73–86) | 20 | 99 | 0.82 (0.68–0.96) |
| qSOFA (2a) ≥ 2 | 73 (39–94) | 79 (72–85) | 18 | 98 | 0.81 (0.69–0.94) |
| qSOFA (2b) ≥ 1 | 82 (48–98) | 73 (66–79) | 50 | 97 | 0.83 (0.68–0.98) |
| qSOFA (3) ≥2 | 55 (24–83) | 82 (76–88) | 16 | 97 | 0.78 (0.64–0.93) |
| UVA (1) ≥ 2 | 82 (48–98) | 81 (75–87) | 21 | 99 | 0.86 (0.70–1.0) |
| UVA (1) ≥ 5 | 55 (23–83) | 99 (96–100) | 75 | 97 | |
| External validation of existing and adapted qSOFA in the Malawi cohort* | |||||
| qSOFA ≥ 2 | 72 (62–80) | 68 (63–73) | 40 | 50 | 0.73 (0.68–0.78) |
| qSOFA (1) ≥ 1 | 58 (48–67) | 53 (48–59) | 27 | 81 | 0.58 (0.51–0.64) |
| qSOFA (3) ≥ 2 | 79 (70–87) | 63 (58–68) | 39 | 91 | 0.77 (0.72–0.82) |
CI = confidence interval; qSOFA = quick Sequential Organ Failure Assessment; SIRS = systemic inflammatory response syndrome; MEWS = modified early warning score; UVA = universal vital assessment. qSOFA (1): qSOFA with respiratory rate and systolic BP thresholds according to UVA criteria but no other changes, using a cutoff ≥1 to identify high-risk patients. qSOFA (2a): original qSOFA + oxygen saturation. qSOFA (2b): adapted (see qSOFA [1]) qSOFA + oxygen saturation, using a cutoff ≥1 to identify high-risk patients. qSOFA (3): original qSOFA + 1 extra point for altered GCS. UVA (1): UVA score without incorporating HIV status.
In our previous study in Malawi, data on temperature and oxygen saturation were not collected, so we limited our external validation to amalgamated qSOFA (1) and qSOFA (3). Part of the data in this table have been previously published.8