OBJECTIVES: To compare the health care costs and hospitalization days of elderly major depressive disorder (MDD) patients treated with escitalopram to those treated with other SSRI/SNRI. METHODS: Elderly MDD patients (age °Ý65) initiated on escitalopram or other SSRI/SNRI were identified in the IHCIS National Managed Care Database (2003 ¨C 2005). Health care costs and hospitalization days were compared between patients on escitalopram to those on other SSRI/SNRI over 6 months following therapy initiation (study period). Wilcoxon tests were used to compare the costs descriptively. Diff-in-diff analyses were conducted to control for baseline (the 6 months prior to therapy initiation) costs. GLM regression with log link function was used for total health care cost comparison. Negative binomial regression was used to compare the number of hospitalization days. Both regressions adjusted for patient demographics, comorbidities, and medical resource use at baseline. Costs were inflation adjusted to 2005 US dollars. RESULTS: The study sample included 459 elderly MDD patients initiated on escitalopram and 1517 patients initiated on other SSRI/SNRI. Descriptive analyses showed that patients on escitalopram had $2143 lower six-month total health care costs than patients on other SSRI/SNRI (p=0.398), and about half of the difference was from lower hospitalization costs ($1271, p=0.145). After adjusting for baseline costs using the diff-in-diff method, the difference in total health care costs became significant ($2319, p=0.037), and about one third of the difference was from lower hospitalization costs ($806, p=0.036). The GLM regression showed that the six-month total health care costs of patients on escitalopram was $3120 less than patients on other SSRI/SNRI (p=0.001). Negative binomial regression showed that patients on escitalopram had 39% less hospitalization days (p=0.004). CONCLUSION: Compared to elderly MDD patients initiated on other SSRI/SNRI, patients who initiated escitalopram have lower health care costs and less hospitalization.