liers in the meta-analyses, which raised the concern that standard
errors had been inadvertently reported instead of standard deviations.
With the original data, the results were SMD -2.04 (95%
CI -3.03 to -1.04) for pain and SMD -1.86 (95% CI -4.63 to
0.91) for disability. Removing them from the analyses resulted in
smaller estimates of effect: SMD -0.75 (95% CI -1.10 to -0.40)
for pain and SMD -0.49 (95% CI -0.87 to -0.12) for disability;
re-calculating the standard deviation and using the new values in
the meta-analyses resulted in less heterogeneity and more modest
estimates of effect that were closer to the sensitivity analyses (SMD
-0.80; 95% CI -1.07 to -0.53 for pain and SMD -0.56; 95% CI
-0.89 to -0.23 for disability).
Considering the quality of the evidence in this review, theseresults
must be considered with caution and generalising the results to all
pregnant women is likely premature.
Incorporating the evidence into clinical practice may be challenging
since ’usual prenatal care’ and ’standard physiotherapy’ are not
described in sufficient detail in the trials and are likely to vary
across jurisdictions. Similarly, there were insufficient details provided
about other interventions that would make it difficult to
replicate in another clinical setting