When attempting to demonstrate a purely antidepressive effect of new antidepressants the HAMD depression factor has been found adequate in placebo-controlled trials. When attempting to demonstrate an early onset of action of amitriptyline the HAMD item of depressed mood has previously been found sufficient, using effect size as outcome statistic. Therefore, the HAMD depression factor as well as the HAMD item of depressed mood have been used separately in this meta-analysis to evaluate the pure antidepressive effect and early onset of action of mirtazapine when compared to placebo or amitriptyline. The results showed that in all placebo-controlled trials mirtazapine obtained an effect size of 0.42 on the HAMD depression factor subscale and 0.49 on the full HAMD. In the trials in which mirtazapine was compared to amitriptyline the effect sizes for the HAMD depression factor subscale were 0.40 and 0.57, respectively. This difference was not statistically significant. An early onset of action was found for the HAMD item of depressed mood as well as the total HAMD both for mirtazapine and amitriptyline when compared to placebo. As early as after 1 wk of therapy both drugs were significantly better than placebo. In conclusion, a purely antidepressive effect of mirtazapine has been demonstrated concerning both improvement after the acute therapy of major depression and early onset of action.