Data from the National Ambulatory Medical Care Survey for the period 1990 through 1995 were used to discern the population-adjusted rate of office-based physician-patient encounters at which the prescribing or continuation of antidepressant pharmacotherapy (tricyclic antidepressants [TCAs], selective serotonin reuptake inhibitors [SSRIs], or others), a diagnosis of depression (International Classification of Diseases, 9th Revision, Clinical Modification codes 296.2 through 296.36, 300.4, or 311), or both were documented. National estimates of the number of office-based visits resulting in a prescription for or continuation of antidepressant pharmacotherapy for any purpose escalated from 16,534,268 in 1990 to 28,664,796 in 1995, a 73.4% increase. Although the number of office-based visits at which a diagnosis of depression was documented increased 23.2% during this period, the proportion of patients with a diagnosis of depression who were prescribed or continued antidepressant pharmacotherapy increased only 14.9%, from 52.1% in 1990 to 67.0% in 1995. Among patients with a diagnosis of depression, use of a TCA declined from 42.1% in 1990 to 24.9% in 1995. In contrast, use of an SSRI for the treatment of depression increased from 37.1% in 1990 to 64.6% in 1995. The rate of office-based visits at which the use of antidepressant pharmacotherapy for any purpose was documented increased from 6.7 per 100 US population in 1990 to 10.9 in 1995, a 62.7% increase; documentation of a diagnosis of depression increased from 6.1 per 100 US population in 1990 to 7.1 in 1995, a 16.4% increase; and the recording of a diagnosis of depression in concert with the prescribing or continuation of antidepressant pharmacotherapy increased from 3.2 per 100 US population in 1990 to 4.8 in 1995, a 50.0% increase. Further research is required to elucidate the effect of observed trends on clinical and financial outcomes.