Meta-analysis of the statistical performance of validation studies of clinical decision rules for group A streptococcal pharyngitis in children (low vs intermediate and high risk)
| Initial CDR (first author) | Reference | Children /total | Se (95% CI) | Sp (95% CI) | PPV (95% CI) | NPV (95% CI) | LR+ (95% CI) | LR− (95% CI) | OR (95% CI) | Percentage of patients at low risk |
|---|---|---|---|---|---|---|---|---|---|---|
| Breese | 25–27 | 2419/2419 | 63 (60 to 66) | 83 (81 to 85) | 74 (71 to 77) | 76 (73 to 78) | 3.2 (2.8 to 3.6) | 0.7 (0.6 to 0.7) | 7.6 (6.2 to 9.3) | 64 |
| Centor | 28 29 | 640/1119 | 41 (34 to 48) | 85 (81 to 88) | 55 (47 to 62) | 76 (72 to 80) | 2.0 (1.6 to 2.7) | 0.8 (0.7 to 0.8) | 3.2 (2.1 to 4.8) | 77 |
| McIsaac | 32–35 | 3187/3652 | 95 (94 to 96) | 14 (13 to 15) | 33 (32 to 35) | 87 (83 to 90) | 1.1 (1.0 to 1.1) | 0.4 (0.3 to 0.5) | 3.2 (2.3 to 4.4) | 10 |
| WHO | 30 31 | 2261/2261 | 6 (4 to 8) | 96 (95 to 97) | 37 (28 to 46) | 73 (71 to 75) | 1.6 (1.1 to 2.4) | 1.0 (1.0 to 1.0) | 1.6 (1.1 to 2.5) | 95 |
| Attia | 36 | 545/545 | 99 (97 to 100) | 4 (3 to 7) | 39 (35 to 44) | 88 (66 to 97) | 1.0 (1.0 to 1.1) | 0.2 (0.1 to 0.9) | 4.9 (1.1 to 21.5) | 3 |
| Smeesters | 23 | 220/220 | 84 (73 to 91) | 41 (34 to 49) | 33 (26 to 41) | 88 (79 to 94) | 1.4 (1.2 to 1.7) | 0.4 (0.2 to 0.7) | 3.7 (1.7 to 8.1) | 35 |
| Joachim | 24 | 576/576* | 88 (82 to 92) | 35 (30 to 40) | 37 (33 to 42) | 87 (81 to 91) | 1.4 (1.2 to 1.5) | 0.3 (0.2 to 0.5) | 4.0 (2.4 to 6.6) | 28 |
The thresholds for low-risk groups were: Breese, score <29 (18–29); Centor, score <2 (0–2); McIsaac, score <1 (0–1); WHO, absence of ADP and exudate; Attia, 0 symptoms; Smeesters, score >8; Joachim, score <2 (0–2).
*Results that concerned the population of Smeesters et al and Joachim et al's study.
CDR, clinical decision rule; LR+, positive likelihood ratio; LR−, negative likelihood ratio; NPV, negative predictive value of the test; PPV, positive predictive value of the test; Se, sensitivity; Sp, specificity.