OBJECTIVE: Antidepressant therapy is highly effective in patients with major depressive disorder (MDD). Evidence has shown that most patients stay on pharmacotherapy for less than 6 months even though clinical guidelines recommend treatment for longer periods of time. The objective of this study was to assess the impact of premature discontinuation of antidepressant therapy on costs and outcomes in MDD patients. METHODS: We created a simulation model to compare the costs and outcomes associated with patients who respond to treatment with a selective serotonin reuptake inhibitor (SSRI) and discontinue treatment prematurely to those who respond to SSRIs and complete the recommended course of treatment. Patients are outpatients and are assumed to follow treatment as recommended by the clinical guidelines except when early discontinuation occurs. The model considers medication, primary care physician visit, specialist (i.e., psychiatrist, hospital days, suicide, etc.), and adverse event costs. Treatment efficacy was taken from published meta-analyses, and early discontinuation was estimated from the published literature. Resource use was estimated from the clinical guidelines and published literature. Unit costs were drawn from standard published sources and inflated to 2003 dollars. RESULTS: Over the course of 5 years, we observe that continuation patients (i.e. patients who complete a recommended course of treatment) have 155 fewer symptom days, 19 fewer disability days, and lower costs by $2133 than discontinuation patients (i.e. patients who discontinue early) when having relapses/recurrences. In the index episode, continuation patients incur more costs than discontinuation patients due to increase usage of drugs and physician. However, patients who discontinue incur more costs later due to higher relapse/recurrence rates. CONCLUSION: By encouraging patients to complete a full course of drug therapy, patients will incur fewer costs and fewer symptom and disability days.