Objective: To assess the 24-month cost-effectiveness of supervised treatment compared to written advice in knee osteoarthritis (OA). Design: 100 adults with moderate-severe OA not eligible for total knee replacement (TKR) randomized to a 12-week individualized, supervised treatment (exercise, education, diet, insoles and pain medication) or written advice. Effectiveness was measured as change in quality-adjusted life years (QALYs) from baseline to 24 months, including data from baseline, 3, 6, 12 and 24 months, while healthcare costs and transfer payments were derived from national registries after final follow-up. Incremental costs, quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratios (ICERs) were calculated. A sensitivity analysis resampling existing data was conducted and the probability of cost-effectiveness was estimated using a 22,665 Euros/QALY threshold. In a sensitivity analysis, cost-effectiveness was calculated for different costs of the supervised treatment (actual cost in study; cost in private practice; and in-between cost). Results: Average costs were similar between groups (6,758 Euros vs 6,880 Euros), while the supervised treatment were close to being more effective (incremental effect (95% CI) of 0.075 (−0.005 to 0.156). In the primary analysis excluding deaths, this led the supervised treatment to be cost-effective, compared to written advice. The sensitivity analysis demonstrated that the results were sensitive to changes in the cost of treatment, but in all scenarios the supervised treatment was cost-effective (ICERs of 6,229 to 20,688 Euros/QALY). Conclusions: From a 24-month perspective, a 12-week individualized, supervised treatment program is cost-effective compared to written advice in patients with moderate-severe knee OA not eligible for TKR. Trial registration: ClinicalTrials.gov number: NCT01535001.