GRADE criteria | Risk of bias | Publication bias | Imprecision | Inconsistency | Indirectness | ||
Item assessed for algorithm | Blinding of outcome assessor | Allocation concealment | Differential attrition | ROBIS domain 2 | CIs + sample sizea | Heterogeneity, using I2 statisticb | Clinical heterogeneity |
No downgrade | < 50% of participants are from trials with high risk of bias for this domain. | < 50% of participants are from trials with high risk of bias for this domain. | Chi2 testc, P ≥ 0.05 OR (if baseline participant numbers not available) < 50% participants are from trials with high risk of bias for differential attrition/loss to follow‐up. |
Low or unclear risk of bias on ROBIS domain 2 | If 95% CI excludes an RR of 1.0, or an MD of 0 | I2 ≤ 50% | Data from participants with SUI and UUI is not combined. AND data from women who are antenatal and postnatal are not combined. |
1 downgrade | ≥ 50% of participants are from trials with high risk of bias for this domain. | ≥ 50% of participants are from trials with high risk of bias for this domain. | YES differential attrition – Chi2 testc, P < 0.05 OR (if baseline participant numbers not available) ≥ 50% participants are from trials with high risk of bias for differential attrition/loss to follow‐up. |
High risk of bias on ROBIS domain 2 | If 95% CI includes an RR of 1.0, or an MD of 0 AND sample size < 344 |
I2 > 50% | If data from participants with SUI and participants with UUI are combined OR if data from women who are antenatal and postnatal (as defined by review authors) are combined OR both. |
2 downgrades | — | — | — | — | If 95% CI includes an RR of 1.0, or an MD of 0 AND sample size < 258 |
I2 > 75% | — |
CI: confidence interval; MD: mean difference; ROBIS: Risk of Bias In Systematic Reviews assessment tool; RR: risk ratio; SUI: stress urinary incontinence; UUI: urgency urinary incontinence. |