Table 2.
Risk of bias
| Randomised trials: bias from randomisation | Non-randomised trials: bias from confounding | Non-randomised trials: bias from participant selection | Non-randomised trials: bias in intervention classification | All: bias from deviation from intended intervention | All: bias from missing outcome data | All: bias in outcome measurement | All: bias in selection of reported results | Overall risk of bias judgment | |
|---|---|---|---|---|---|---|---|---|---|
| Bavinton et al (2018)18 | NA | Low | Low | Low | Low | Low | Low | Low | Low |
| Cohen et al (2016)19 | Low | NA | NA | NA | Low | Low | Low | Low | Low |
| Fideli et al (2001)25 | NA | Low | Some | Low | Low | Low | Low | Low | Low |
| Mujugira et al (2016)26 | Low | NA | NA | NA | Low | Low | Low | Low | Low |
| Quinn et al (2000)20 | Low | NA | NA | NA | Low | Low | Low | Low | Low |
| Rodger et al (2016)21 | NA | Low | Low | Low | Low | Low | Low | Low | Low |
| Rodger et al (2019)22 | NA | Low | Low | Low | Low | Low | Low | Low | Low |
| Tovanabutra et al (2002)27 | NA | Some | Some | Some | Low | Low | Low | Low | Some |
NA=not applicable.