Nineteen studies (717 participants) were included. Sample sizes ranged from 14 to 312. Five studies were classified as high quality, three of moderate quality and 11 of low quality.
Nine studies compared aquatic exercise to no treatment. Three studies were of sufficient quality and had appropriate data to include in the meta-analysis. There was a small but statistically significant reduction in pain levels in favour of aquatic exercise (SMD -0.17; 95% CI, -0.33, 0.01, p=0.04). Overall results of the nine studies were not consistent.
Ten studies compared aquatic exercise to land exercise. Two studies were of sufficient quality and had appropriate data to included in the meta-analysis. There was no difference between the two interventions in pain outcome (SMD 0.11; 95% CI, -0.27, 0.50, p=0.56); all but one of the remaining studies were consistent with this finding. The authors stated that the statistical tests for heterogeneity were not significant for these analyses.
Two studies compared aquatic exercise to immersion (one study was low quality). Neither study found a difference between the two interventions in pain outcome.