Objective.

—To compare the clinical, functional, and economic outcomes of initially prescribing fluoxetine with outcomes of initially selecting imipramine or desipramine.

Design.

—Randomized controlled trial.

Setting.

—Primary care clinics of a Seattle, Wash, area staff-model health maintenance organization from 1992 through 1994.

Patients.

—A total of 536 adults beginning antidepressant treatment for depression.

Intervention.

—Random assignment of initial antidepressant prescription (desipramine, fluoxetine, or imipramine). Subsequent antidepressant treatment (doses, medication changes or discontinuation, specialty referral) was managed by the primary care physician.

Main Outcome Measures.

—Assessments after 1, 3, and 6 months examined clinical outcomes (Hamilton Depression Rating Scale and the depression subscale of the Hopkins Symptom Checklist) and quality-of-life outcomes (Medical Outcomes Study SF-36). Medication use and health care costs were assessed using the health maintenance organization's computerized data.

Results.

—Patients assigned to receive fluoxetine reported fewer adverse effects, were more likely to continue the original medication, and were more likely to reach adequate doses than patients beginning treatment with either tricyclic drug. The fluoxetine group reported marginally better clinical outcomes after 1 month, but these differences were not statistically significant and disappeared by the 3-month assessment. Quality-of-life outcomes in the 3 groups did not differ. Total health care costs over 6 months were approximately equal for the 3 groups, with higher antidepressant costs in the fluoxetine group balanced by lower outpatient visit and inpatient costs.

Conclusions.

—Clinical outcomes, quality-of-life outcomes, and overall treatment costs provide no clear guidance on initial selection of fluoxetine or tricyclic drugs. Thus, patients' and physicians' preferences are an appropriate basis for treatment selection. (JAMA. 1996;275:1897-1902)

	Year	Citations

Abstract

References