Objective: To evaluate the literature on the effectiveness of aquatic exercise in relieving pain in adults with neurologic or musculoskeletal disease. Data sources: A systematic literature search of 14 databases was examined for research on aquatic exercise over the period January 1980 to June 2006. Study selection: Randomized controlled trials (RCTs) that included adults with neurologic or musculoskeletal disease, pain as an outcome measure, and exercise in water were included. Data extraction: Information on the participants, interventions, and outcomes was extracted from the included studies. Quality appraisal was assessed using the Scottish Intercollegiate Guidelines Network criteria for RCTs. Data synthesis: Nineteen studies met the inclusion criteria; 8 were of moderate to low risk of bias, and 5 of these had data suitable for meta-analyses. This showed that aquatic exercise has a small posttreatment effect in relieving pain compared with no treatment (P=.04; standardized mean difference [SMD], -.17; 95% confidence interval [CI], -.33 to -.01), but it is not possible to draw a firm conclusion because of the lack of consistency of evidence across studies. Comparable pain-relieving effects were found between aquatic and land-based exercise (P=.56; SMD=.11; 95% CI, -.27 to .50). Conclusions: There is sound evidence that there are no differences in pain-relieving effects between aquatic and land exercise. Compared with no treatment, aquatic exercise has a small pain-relieving effect; however, the small number of good-quality studies and inconsistency of results means that insufficient evidence limits firm conclusions. Future studies should aim for focused research questions on specific aquatic exercise techniques, using robust methodologic designs and detailed reporting of temperature, depth, and care setting.