Good evidence supports the efficacy of pharmacologic treatment and brief psychotherapy for depression, but the trials included primarily white, upper-middle-class populations. Miranda and colleagues considered whether guideline-based care for depression improves symptoms and function in depressed, low-income minority women.

The researchers screened women in Women, Infants, and Children programs and Title X family-planning clinics. Of 532 black, 71 white, and 408 Latina patients meeting inclusion criteria and with a diagnosis of major depressive disorder, 427 completed a structured psychiatric diagnostic telephone interview. Of these, 267 completed a clinical interview and were randomized to antidepressant medications (n = 88), cognitive behavior therapy (n = 90), or referral to community mental health services (n = 89), with 117 black women, 134 Latina women, and 16 white women participating.

Antidepressant medication was given for six months, with paroxetine used initially. If paroxetine had no effect despite dosage adjustment or was not tolerated, bupropion was used. Patients in the behavior therapy group were treated by experienced psychotherapists in eight weekly group or individual sessions. Women assigned to community health center referral were offered help with making their appointments and were contacted frequently to encourage them to attend the intake appointment for care. During the six-month study, patients were screened at monthly intervals with the Hamilton Depression Rating Scale (HDRS) and at baseline and three-month intervals for functional outcomes and social function.

Patients in this study were poor and mostly black and Latina, and many had suffered rape or abuse. Fewer than one half were living with a partner or married, and the average patient had two or more children. Of the women offered community referral, 74 (83 percent) failed to attend even one session, and only eight (9 percent) attended four or more sessions. Of the women randomized to medication, 67 (76 percent) received nine or more weeks of medication, and 40 (45 percent) received guideline-concordant medication for 24 or more weeks. Of the patients randomized to behavior therapy, 48 (53 percent) received four or more sessions, and 32 (36 percent) received six or more sessions.

According to the HDRS, women who received medication experienced a significant decrease in depressive symptoms (P < .001), as did women participating in behavior therapy (P = .006) when compared with the community referral group. Medication, but not behavior therapy, significantly improved instrumental role function, and both treatments improved social function compared with community referral. By month 6, 44.4 percent of the medication group, 32.2 percent of the behavior therapy group, and 28.1 percent of the community referral group had achieved HDRS scores of 7 or less. Further analyses replicated the results in the medication group, but results showed less benefit in the behavior therapy group compared with the referral group.

It is likely that outreach, transportation, and child care for the intervention groups enhanced access to treatment. This study showed that without this support, women were unlikely to use resources that were available to them. Evidence-based treatments appear to be effective in minority women if they are given support to overcome barriers to care. Medication interventions may be more effective in this population than psychotherapy interventions.