Objective To investigate the efficacy and its relevant influence factors of paroxetine combined with mirtazapine in the treatment of patients with depression without response in early stage. Methods A multicenter， block randomized， double-blind， double-dummy， controlled trial was designed. A total of 525 patients with depression were selected from five general hospitals and psychiatric hospitals in China. Participants received paroxetine during a 2-weeks open-label phase， with 475 cases completed. A total of 204 patients who showed non-response in early stage （Hamilton Rating Scale score reduction rate ＜ 20%） were divided into paroxetine group （n=68）， mirtazapine group （n=68）， and paroxetine combined with mirtazapine group （n=68） according to the ratio of 1∶1∶1 by block randomization method. After 6 weeks of treatment， the curative effect was evaluated by the reduction rate of Hamilton Depression Scale at the time of randomization and before and after treatment. The treatment was effective if the reduction rate was more than 50%. The patients were divided into effective group and ineffective group according to the results. Univariate and multivariate Logistic regression analysis were used to analyze the influencing factors of efficacy. Results Univariate analysis showed that there was no significant correlation between each variable and antidepressant efficacy. Multivariate logistic regression analysis showed that age ≤ 37 years old （OR=4.694， 95%CI=1.061-20.779， P＜0.05） and combination treatment of paroxetine and mirtazapine （OR=2.836， 95%CI=1.035-7.770， P＜0.05） were independent efficacy factors after controlling for the confounding factors. Conclusions The efficacy of paroxetine combined with mirtazapine in patients with acute depression is superior to monotherapy. Age under 37 years old is an independent risk factor of efficacy.