## Table 4 |
||||

Results of the meta-analyses taking the between-group standardized mean differences
as the effect size |
||||

Outcome measure |
k |
d_{+} |
95% C. I. |
d_{T}- d_{C} |

d_{l}d_{u} |
||||

Pain | 5 | 1.019 | 0.478 1.561 | 0.994 |

Disability | 2 | 0.366 | −0.345 1.077 | 0.742 |

Flexibility | 2 | 0.703 | −0.024 1.430 | 0.711 |

Endurance | 1 | 0.870 | −0.212 1.952 | 0.777 |

Mental health | 1 | −0.243 | −1.235 0.749 | −0.115 |

Self-reports | 5 | 0.656 | 0.127 1.184 | 0.737 |

Clinicians | 2 | 0.548 | −0.176 1.273 | 0.641 |

Combined | 5 | 0.664 | 0.139 1.189 | 0.730 |

*k*: number of studies for each outcome measure. *d*_{+}: mean effect size for each outcome. *d*_{l} and *d*_{u}: lower and upper limits of the 95% confidence interval for the mean effect size in
each outcome. *d*_{T} – *d*_{C}: difference between the treatment and control within-group means (calculated from
Table 3). All of the heterogeneity *Q* statistics were not statistically significant (*p* > .05) and all *I*^{2} indices were equal to 0%.

Calvo-Muñoz * et al.*

Calvo-Muñoz * et al.* *BMC Musculoskeletal Disorders* 2013 **14**:55 doi:10.1186/1471-2474-14-55