Table 2. Assessment of the certainty of evidence using GRADE approach.
| Outcomes (no. of trials) | Certainty assessment | Effect | Certainty | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Study design | Risk of bias | Inconsistency | Indirectness | Imprecision | Publication bias | No. of patients | SMD (95%CI) | ||
| BBS (5) | RCT | Serious a | Serious b | Not serious | Not serious | None | 326/324 | 0.28 (0.03, 0.53) | ⊕⊕◯◯ Low |
| TUG (5) | RCT | Serious a | Not serious | Not serious | Serious c | None | 170/177 | -0.28 (-0.50, -0.07) | ⊕⊕◯◯ Low |
| Fast gait speed (5) | RCT | Not serious | Serious b | Not serious | Serious c | Potential bias | 172/168 | 0.29 (-0.08, 0.66) | ⊕◯◯◯ Very low |
| Usual gait speed (10) | RCT | Serious a | Not serious | Not serious | Not serious | None | 420/411 | 0.07 (-0.07, 0.21) | ⊕⊕⊕◯ Moderate |
| SPPB (5) | RCT | Not serious | Serious b | Not serious | Not serious | None | 298/289 | 0.30 (-0.10, 0.69) | ⊕⊕⊕◯ Moderate |
| 6MWT (4) | RCT | Not serious | Serious b | Not serious | Serious c | None | 97/81 | 0.37 (-0.19, 0.92) | ⊕⊕◯◯ Low |
| Barthel’s ADL (5) | RCT | Not serious | Serious b | Not serious | Not serious | None | 202/202 | 0.05 (-0.29, 0.38) | ⊕⊕⊕◯ Moderate |
| IADL (5) | RCT | Not serious | Not serious | Not serious | Serious c | None | 172/176 | 0.15 (-0.06, 0.36) | ⊕⊕⊕◯ Moderate |
| Knee extensor strength (7) | RCT | Serious a | Not serious | Not serious | Not serious | Potential bias | 417/385 | 0.23 (0.09, 0.37) | ⊕⊕◯◯ Low |
| Falls efficacy scale (6) | RCT | Not serious | Serious b | Not serious | Not serious | None | 260/253 | 0.32 (-0.07, 0.71) | ⊕⊕⊕◯ Moderate |
| SF-36 PCS (6) | RCT | Not serious | Not serious | Not serious | Serious c | None | 255/139 | 0.49 (0.28, 0.70) | ⊕⊕⊕◯ Moderate |
6MWT: 6-minute walking test; BBS: Berg balance score; GRADE: Grading of Recommendations Assessment, Development and Evaluation; IADL: instrumental activities of daily living; RCT: randomized controlled trial; SF-36 PCS: Short Form-36 questionnaire physical component score; SMD: standard mean difference; SPPB: short physical performance battery; TUG: timed-up-and-go test;
a Including >25% of patients from studies with low methodological quality.
b Wide variance of estimates across studies or large between-study heterogeneity (I2>50%)
c Total number of patients <400.