Depression is associated with marked suffering, morbidity, and high risk of recurrence and/or chronicity. As a result, the disorder represents a considerable public health problem and economic burden on society. Treatment of patients with depression to a state of remission is associated with a significantly improved long-term outcome, including a reduced risk of relapse and improved functioning. Thus, remission (which can be defined as attainment of a total score < or = 7 on the Hamilton Rating Scale for Depression) should be the goal of the acute phase of pharmacotherapy. Although it is widely assumed that available antidepressants are comparably effective, comparisons of the efficacy of antidepressants in clinical trials are flawed by a number of factors. Most notable problems are higher-than-expected placebo effects and low statistical power. Double-blind, comparative studies are also compromised by relatively high attrition rates and the often underestimated effects of patient nonadherence. Large study groups (in the order of 300 patients per treatment group) are needed to differentiate between a good and a potentially better antidepressant. The statistical technique of meta-analysis has been used to examine the results of studies of various antidepressants; these meta-analyses have shown that the selective serotonin reuptake inhibitors (SSRIs) have overall "within group" comparability and, overall, an efficacy profile comparable to the previous standard, tricyclic antidepressants (TCAs). However, in studies of hospitalized patients, TCAs affecting both serotonergic and noradrenergic systems (eg, amitriptyline or clomipramine) have been found to have greater efficacy when compared with SSRIs. Results of some individual studies, as well as a pooled analysis of the outcomes of more than 2,000 depressed patients, indicate that venlafaxine, a selective serotonin and norepinephrine reuptake inhibitor, may have a similar advantage relative to SSRIs across a broader range of patients. These findings suggest antidepressants that affect both serotonergic and noradrenergic neurotransmission may be more likely to accomplish the goal of remission. Compared with TCAs, the better safety profile of venlafaxine has established the drug as a more appropriate first-line treatment option.