OBJECTIVE: To assess health care costs among patients with multiple sclerosis (MS) in the Florida Medicaid program. METHODS: We employed a retrospective matched cohort design using administrative claims data for Florida Medicaid recipients. Patients were included if they had a diagnosis of MS between July 1, 2001 and June 30, 2002, and were eligible for Medicaid as of July 1, 2001. Those covered in capitated plans or dually eligible for Medicare and Medicaid were excluded. The comparison cohort consisted of MS-free patients matched on age, gender, and race to the MS group. The excess cost of MS (in 2002 US dollars) was estimated as the difference in mean Medicaid payments between the MS group and matched controls. Patients were followed for 12 months unless eligibility terminated earlier. RESULTS: A total of 951 patients with MS met cohort selection criteria, a prevalence rate of approximately 8 per 10,000 Medicaid eligibles. About one-third of these patients were dispensed glatiramer acetate or beta-interferon. MS patients and matched controls (n = 951) averaged 43 years of age; 78% were female, and 61% were white. Both cohorts had similar mean Charlson comorbidity scores and low mortality rates (<3%). Compared to their matched controls, MS patients were more likely to be hospitalized (33% vs. 20%, respectively), use nursing home services (15% vs. 3%), or use home health care services (37% vs. 11%). The annual per-patient excess cost of MS was estimated to be $11,383 ($20,264 vs. $8,881 for matched controls). Long-term care accounted for 44% of the excess costs, followed by medications (28%), and hospitalizations (14%). CONCLUSIONS: Although MS is relatively rare in this Medicaid population, the per-patient costs are high. Long-term care costs are the largest portion of Medicaid payments, highlighting the disease burden.