Table 4.
Quality of the included studies
| Study | Selection bias | Study design | Confounders | Blinding | Data collection | Drop-outs | Global rating |
|---|---|---|---|---|---|---|---|
| [26] | Weak | Moderate | Weak | Weak | Strong | Strong | Weak |
| [27] | Moderate | Moderate | Moderate | Weak | Strong | Strong | Moderate |
| [28] | Strong | Strong | Moderate | Weak | Strong | Strong | Moderate |
| [29] | Strong | Strong | Strong | Moderate | Strong | Strong | Strong |
| [30] | Moderate | Moderate | Moderate | Weak | Moderate | Strong | Moderate |
| [36] | Strong | Strong | Moderate | Moderate | Strong | Strong | Strong |
| [31] | Moderate | Moderate | Moderate | Weak | Strong | Strong | Strong |
| [33] | Moderate | Moderate | Moderate | Weak | Moderate | Moderate | Moderate |
| [34] | Moderate | Moderate | Weak | Weak | Weak | Moderate | Weak |
| [35] | Moderate | Moderate | Weak | Weak | Moderate | Moderate | Weak |
| [37] | Moderate | Moderate | Moderate | Weak | Strong | Strong | Moderate |
| [23] | Weak | Moderate | Moderate | Weak | Strong | Moderate | Weak |
| [24] | Moderate | Moderate | Weak | Moderate | Moderate | Moderate | Moderate |
| [25] | Moderate | Moderate | Moderate | Weak | Moderate | Strong | Moderate |
| [32] | Weak | Moderate | Weak | Moderate | Moderate | Moderate | Moderate |
| [38] | Moderate | Moderate | Moderate | Moderate | Strong | Strong | Strong |
| [39] | Moderate | Moderate | Weak | Weak | Moderate | Strong | Weak |