Table 7.
Outcome comparisons in effect sizes: Sensitivity analyses (SA)
| Analysis | Median | Interquartile range | Ranges: Number (percentage) of comparisons | N (total) | N | ||||||
| ≤ -0.8 | -0.79 to -0.5 | -0.49 to -0.2 | -0.19 to 0.19 | 0.2 to 0.49 | 0.5 to 0.79 | ≥ 0.8 | Comparisons | Cohorts | |||
| Main analysis: any design, setting, intervention, and baseline status | 0.11 | -0.11 to 0.35 | 13 (3) | 20 (4) | 58 (11) | 214 (41) | 142 (27) | 53 (10) | 17 (3) | 517 (100) | 84 |
| SA1: Design: only observational studies (excluding randomised controlled trials) | 0.08 | -0.14 to 0.28 | 6 (2) | 11 (4) | 40 (13) | 140 (47) | 77 (26) | 21 (7) | 6 (2) | 301 (100) | 44 |
| SA2: Setting: only primary care/health maintenance organization | 0.24 | 0.03 to 0.38 | 3 (2) | 4 (3) | 9 (7) | 42 (30) | 60 (43) | 16 (12) | 4 (3) | 138 (100) | 27 |
| SA3: Intervention: only drugs, physiotherapy, other physical therapy, or mixed | 0.08 | -0.09 to 0.30 | 3 (1) | 7 (2) | 36 (11) | 164 (50) | 86 (26) | 21 (6) | 8 (2) | 325 (100) | 49 |
| SA4: Baseline status: only comparisons with small baseline difference (< 0.50 standard deviation) | 0.08 | -0.13 to 0.28 | 5 (1) | 14 (4) | 41 (12) | 161(47) | 81 (24) | 29 (9) | 9 (3) | 340 (100) | 81 |
| SA1 + SA2 + SA3 + SA4 | 0.07 | -0.03 to 0.19 | 0 (0) | 0 (0) | 2 (13) | 10 (63) | 3 (19) | 0 (0) | 1 (6) | 16 (100) | 5 |