A previously published decision-analytic model assessing the clinical and economic consequences of topiramate versus no preventive treatment in migraineurs was updated with new published literature and unpublished clinical trial data. The model captured baseline migraine days, treatment discontinuation, treatment response (i.e., > or = 75%, 50%-74%, and < 50% reduction in migraine frequency), hours of disability, cost of preventive therapy, cost of acute treatment (pharmacy and medical service), and wages. Topiramate was associated with 29 fewer migraine-days and 78 fewer hours of disability per year, compared with no preventive treatment. The incremental cost per migraine-day averted for topiramate versus no preventive treatment was dollar 29 when only direct medical costs were considered and dollar 2 when total costs were included. Model results were sensitive to baseline migraine-days, response probability, and probability of an attack being treated with a triptan. Topiramate may be a cost-effective treatment for the prevention of migraine.