TABLE 4.
GRADE evidence profile.
| Certainty assessment | No of patient | Effect estimate | Certainty | Importance | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No of studies | Study design | Risk of bias | Inconsistency | Indirectness | Imprecision | Other considerations | E | C | Relative (95% CI) | Absolute (95% CI) | ||
| Clinical efficacy | ||||||||||||
| 20 | RCT | Serious | Not serious | Not serious | Not serious | None | 809/869 (93.1%) | 650/872 (74.5%) | RR 1.25 (1.20, 1.30) | 186 per 1,000 (149, 224) | Moderate | CRITICAL |
| Time for defervescence | ||||||||||||
| 15 | RCT | Seriousa | Very seriousb | Not serious | Not serious | None | 670 | 670 | - | MD −1.55 (−1.91, −1.18) | Very low | IMPORTANT |
| Time for disappearance of cough | ||||||||||||
| 9 | RCT | Seriousa | Very seriousb | Not serious | Not serious | None | 396 | 390 | - | MD −1.97 (−2.63, −1.31) | Very low | IMPORTANT |
| Time for disappearance of lung rales | ||||||||||||
| 14 | RCT | Seriousa | Very seriousb | Not serious | Not serious | None | 631 | 631 | - | MD −2.39 (−3.09, −1.70) | Very low | IMPORTANT |
| Improvement time of chest X-ray | ||||||||||||
| 10 | RCT | Seriousa | Not serious | Not serious | Not serious | None | 443 | 443 | - | MD −2.73 (−2.96, −2.49) | Moderate | IMPORTANT |
| The hospitalization time | ||||||||||||
| 8 | RCT | Seriousa | Very seriousb | Not serious | Not serious | None | 363 | 369 | - | MD −3.11 (−3.80, −2.41) | Very low | IMPORTANT |
| APACHEⅡ | ||||||||||||
| 4 | RCT | Seriousa | Very seriousb | Not serious | Seriousc | None | 158 | 154 | - | MD −2.45 (−6.07, 1.17) | Very low | IMPORTANT |
Abbreviations: CI, confidence interval; MD, mean difference; RR, risk ratio; E, Experimental; C, Control. Explanations: a. The results of the COCHRANE, bias risk assessment show that most (2/3) of the information comes from moderate bias; b. Heterogeneity test shows a high I2 value, with the forest plot indicating I2 > 75%; c. 95% confidence interval CI, over equivalent line.