Aim: To probe into the effect of early intervention for post-stroke depression (PSD) on movement function after three months of stroke and the incidence rate of recurrent stroke within three years. Methods: Totally 180 patients hospitalized in Nanyang Central Hospital from January 2000 to August 2004, who were diagnosed as having PSD, were divided at random into a treatment group (90 cases) and a control group (90 cases) . In the severity of neurological function damage, there was no statistical difference between the two groups (P > 0. 05), which were given similar treatments of one month. At the same time when diagnosed as having PSD, the treatment group was given an anti-depression treatment: fluoxetine 20 mg once a day, continuously for three months. After that, evaluation severity of clinic extent of neurological function damage was used to assess the severity of patients' disease; and Hamilton rating scale for depression (HAMD) was taken to evaluate the severity of depression. Results: HAMD score of treatment group was 6. 4 ± 6. 2 after treatment, and 18. 5 ± 6. 0 before treatment, and there was significant difference (P < 0. 01). The score on the extent of neurological function damage was 10. 4 ± 8. 5 in the treatment group, and 22. 6 ± 8. 9 in the control group, and there was significant difference(P < 0. 05). The incidence rate of recurrent stroke within three years in the treatment group was 7. 1%, and 31% in the control group, and there was significant difference (P < 0. 05). Conclusion: Early intervention in post-stroke depression can lower the scores of neurological function defect and the incidence rate of recurrent stroke in three years, and reduce death rate.