Table 3. Primary outcome comparisons between and within groups (PP population 3rd–25th percentile)
| Variable | Observed value | Changed from baseline |
| CG (n = 38) | EG (n = 37) | p** | CG (n = 38) | p* | EG (n = 37) | p* | p** |
| Height (cm) | | | | | | | | |
| visit 2 | 125.07 ± 9.31 | 127.47 ± 10.20 | 0.3481) | | | | | |
| visit 3 | 125.96 ± 9.18 | 129.18 ± 10.70 | 0.2391) | 0.89 ± 0.67 | < 0.0012) | 1.71 ± 1.48 | < 0.0013) | 0.0041) |
| visit 4 | 126.67 ± 9.23 | 129.96 ± 10.68 | 0.2031) | 1.60 ± 0.59 | < 0.0012) | 2.48 ± 1.54 | < 0.0013) | 0.0021) |
| visit 5 | 127.28 ± 9.19 | 130.67 ± 10.90 | 0.2171) | 2.21 ± 0.66 | < 0.0012) | 3.19 ± 1.70 | < 0.0013) | < 0.0011) |
| visit 6 | 128.06 ± 9.35 | 131.69 ± 10.91 | 0.1941) | 2.99 ± 0.72 | < 0.0012) | 4.22 ± 1.68 | < 0.0013) | < 0.0011) |
p-values were compared within each group.
p-values were compared between groups.
p-values were derived from Mann-Whitney U test.
p-values were derived from paired t-test.
p-values were derived from Wilcoxon’s signed rank test.
PP, per-protocol; CG, control group; EG, experimental group.