Table 3.
GRADE assessments for the certainty of evidence.
| Outcome | Study Design | RoB 1 | Publication Bias | Inconsistency | Indirectness | Imprecision | Quality of Evidence | Recommendation |
|---|---|---|---|---|---|---|---|---|
| ROM | 11 RCTs, 452 participants in meta-analysis. | Randomization—low in four articles, moderate in one, and high in six. | No publication bias. | 9 RCTs showed improvements in ROM in both groups. 2 RCTs showed no changes in ROM in either group. 11 RCTs showed effects of equal magnitude for ST and stretching. 2 |
No serious indirectness. | Moderate. 3 | Moderate. ⨁⨁⨁ |
Moderate recommendation for either strength training or stretching 4. No recommendation for choosing one of the protocols over the other, as their efficacy in ROM gains was statistically not different. |
| Deviations from intended interventions—Low. | ||||||||
| Missing outcome data—Low. | ||||||||
| Measurement of the outcome—low in six articles and high in five. | ||||||||
| Selection of the reported results—Low. |
1—Meta-analyses moderated by RoB showed no differences between studies with low and high risk. 2—Because ROM is a continuous variable, high heterogeneity was expected. However, this heterogeneity is mostly between small and large beneficial effects. No adverse effects were reported. 3—Expected because ROM is a continuous variable. Furthermore, imprecision referred to small to large beneficial effects. 4—Both strength training and stretching presented benefits without reported adverse effects.