Table 4 The ratio of observed and estimated effect size in 113 randomized trials
From: Estimates of the mean difference in orthopaedic randomized trials: obligatory yet obscure
Ratio of observed effect size and estimated effect size in power calculation | Follow-up time not applicable | Follow-up under 3 months | Follow-up between 3 and 6 months | Follow-up between 6 and 12 months | Follow-up over 12 months | Total number of RCTs (%) |
|---|---|---|---|---|---|---|
Median (IQR) | 0.87 (0.7–1.1) | 0.51 (0.2–0.8) | 0.39 (0.2–0.6) | 0.21 (0.2–0.5) | 0.28 (0.1–0.6) | 0.41 (0.2–0.8) |
< 0.5 | 1 (8 %) | 15 (48 %) | 10 (63 %) | 12 (80 %) | 27 (71 %) | 65 (58 %) |
0.5-1.0 | 6 (46 %) | 9 (29 %) | 3 (19 %) | 1 (7 %) | 7 (18 %) | 26 (23 %) |
1.0-1.5 | 5 (38 %) | 4 (13 %) | 3 (19 %) | 2 (13 %) | 2 (5 %) | 16 (14 %) |
> 1.5 | 1 (8 %) | 3 (10 %) | 0 | 0 | 2 (5 %) | 6 (5 %) |