Table 5.
Summary of findings and quality of evidence assessment for physical activity interventions vs control
| Outcome | Summary of findings | Quality of evidence assessment (GRADE) | |||||
|---|---|---|---|---|---|---|---|
| Effect sizea (95% CI) | n (trials) | Study limitations | Imprecision | Inconsistency | Publication bias | Quality | |
| Main outcome | 0.15 (0.05 to 0.25) | 1560 (20) | -1b | None | None | None | Moderate |
| Femoral neck BMD | 0.09 (− 0.03 to 0.21) | 1032 (14) | -1b | None | None | -1c | Low |
| Lumbar spine BMD | 0.17 (0.04 to 0.30) | 874 (11) | -1b | None | None | None | Moderate |
BMD bone mineral density, n number of participants
apooled standardised effect size and 95% confidence intervals (positive value favours physical activity interventions)
b > 50% of studies in the meta-analysis had a PEDro score < 6/10
cSerious small study effects suggested by visual inspection of funnel plot or sensitivity analysis investigating the impact of removal of small studies on pooled estimate