OBJECTIVE To compare the effectiveness of aquatic exercises with patient-education in individuals with knee osteoarthritis. METHODS Randomized controlled trial with blinded assessor and intention-to-treat analysis. METHODS Aquatic Physiotherapy Centre and Primary Health Care Unit. METHODS A total of 60 patients, aged 68.3 (SD = 4.8) with clinical symptoms and radiographic grading (Kellgren-Lawrence 1-4) of knee osteoarthritis were included. METHODS An eight-week treatment protocol of aquatic exercise ( n = 31) (16 individual sessions, twice a week) and an educational program (group sessions, once a week) ( n = 29). METHODS Before, after eight-week intervention, and a three-month follow-up with results for the following outcome measures: pain, function, quality of life, functional mobility, and depression. RESULTS At the end of treatment, the WOMAC (Western Ontario and McMaster Universities Osteoarthritis Index) functional capacity values reduced in favour of the aquatic exercise group for both the total score MD (mean difference) = -14.2; 95% CI (confidence interval) (-18; -10.5), P = 0.04 and the pain domain MD = -3.8 points; 95% CI (-8.71; -1), P = 0.021. The total score also reduced in the follow-up: MD = -12.3 points; 95% CI (-24.7; -6.1), P = 0.017. No differences were found for the outcomes functional mobility or depression. CONCLUSIONS Aquatic exercise improved pain and function after eight weeks, and function at the three-month follow-up compared to the patient-education program.