Table 3.
GRADE summary of findings table
| Summary of results | Quality of the evidence (GRADE) | ||||||
| Follow-up | Number of participants(trials) |
SMD
(95% CI) |
Design | Inconsistency | Indirectness | Imprecision | Quality |
| Short term | 385 (6 trials) | −0.28 (−0.49 to −0.08) | Limitations* | No inconsistency | No indirectness | Imprecision† |
Low
⨁⨁◯◯ |
| Medium term | 173 (3 trials) | −0.59 (−1.03 to −0.15) | Limitations* | No inconsistency | No indirectness | Imprecision† |
Low
⨁⨁◯◯ |
| Long term | 345 (5 trials) | 0.01 (−0.39 to 0.41) | Limitations* | Inconsistency‡ | No indirectness | Imprecision† |
Very low
⨁◯◯◯ |
| Sensitivity analysis | |||||||
| Short term | 215 (3 trials) | −0.27 (−0.54 to −0.05) | No limitations | No inconsistency | No indirectness | Imprecision† |
Moderate
⨁⨁⨁◯ |
| Medium term | 40 (1 trials) | −0.32 (−0.95 to 0.31) | No limitations | Inconsistency§ | No indirectness | Imprecision† |
Low
⨁⨁◯◯ |
| Long term | 215 (3 trials) | 0.13 (−0.14 to 0.40) | No limitations | No inconsistency | No indirectness | Imprecision† |
Moderate
⨁⨁⨁◯ |
*Lack of blinding of participants and personnel, attrition bias, unable to adequately assess selection bias risk.
†<400 participants for each outcome.
‡Large statistical heterogeneity; I2=70%.
§Only single trial available, <400 participants therefore downgraded for inconsistency and imprecision.
Short term, ≤3 months; medium term, >3 and <12 months; long term, ≥12 months.
High quality: further research is unlikely to change our confidence in the estimate of effect.
Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect.
Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect.
Very low quality: we are uncertain about the estimate.
GRADE, Grading of Recommendations Assessment, Development and Evaluation; SMD, standardised mean difference.