Objective: To evaluate the efficacy and safety of duloxetine enteric coated tablets in patients with depressive disorders. Methods: A double-blind, double-dummy, parallel randomized controlled study was carried out for 51 patients who met the DSM-IV criteria of depression and depressed episode. 25 of the total patients were treated with duloxetine (40-60 mg·d-1) and the rest were treated with paroxetine (20mg·d-1) for 8 weeks. Efficacy was then assessed by Hamilton Depression Rating Scale (HAMD), Hamilton Anxiety Scale(HAMA), Montgomery Depression Rating Scale (MADRS), Sheehan Disability Scale (SDS), visual analogue scale (VAS-PI) and Clinical Global Impression (CGI). Safety assessments included physical examinations, laboratory evaluations, and electrocardiographic findings of adverse events. Assessment time point were evaluated at baseline and 1,2,4,6,8 weeks after starting treatment. Results: After 8 weeks of treatment, the total effective rates of duloxetine group and paroxetine group were 72.0% and 73.1%, respectively, with no significant difference (t=0.465, P=0.612). The clinical curing rates of duloxetine group and paroxetine group were 20.0% and 23.1%, respectively, with no significant difference (t=0.547, P=0.590). The scores of HAMD17, HAMA, MADRS, SDS, VAS-PI and CGI in both groups decreased significantly, with statistical difference between the baseline and other observation time points (P < 0.001). The adverse event rates of the duloxetine group and paroxetine group were 36.0% and 34.6% respectively, with no significant difference (P > 0.05). The main adverse events of two groups were dry mouth, nausea, dizziness, and stomach discomfort. Conclusion: Duloxetine enteric-coated tablet is an effective antidepressant with less side effects, better safety, and more suitable for the treatment of depression patients.