Objective: To evaluate the long-term benefits of constraint-induced movement therapy in chronic stroke.Design: A four-year follow-up after constraint-induced group therapy assessing arm and hand function and self-reported daily hand use.Subjects: Fourteen post-stroke individuals (six women and eight men; mean age 59.6 ± 12.7 years, range 23-75 years) with mild to moderate impairments of hand function.Outcome measures: The Sollerman hand function test and the Motor Activity Log test.Results: Four years after constraint-induced group therapy the participants had maintained their hand function, as measured by the Sollerman hand function test. The self-reported use and quality of movements of the more affected hand, as measured by the Motor Activity Log test, had decreased compared to post-treatment and three months follow-up (P < 0.01), but was still significantly higher than pre-treatment (P < 0.05).Conclusion: There seems to be a long-term benefit of constraint-induced group therapy. Hand function was maintained over time and daily hand use had increased compared to pre-treatment. To provide guidelines about the clinical use of constraint-induced movement therapy further, larger and controlled studies are needed.