Historically, clinical researchers have gauged the short-term effectiveness of antidepressants by response rates, which have been defined as a "significant" reduction of symptoms or a global impression of at least moderate benefit. However, increased focus over the past decade has led many researchers to suggest that remission, i.e., a virtual elimination of depressive symptoms and restoration of psychosocial functioning, should be the primary goal of the initial phase of therapy. This article examines the relative efficacy of various antidepressant therapies. There is some evidence that medications affecting multiple neurochemical systems, such as the tricyclics amitriptyline and clomipramine (in studies of hospitalized patients) and the more selective "dual reuptake inhibitor" venlafaxine, may result in higher rates of remission relative to other agents. Given the better tolerability of newer antidepressants relative to tricyclics, both logic and an increasing amount of data support a greater role for multiaction antidepressants.