Epilepsy is one of the commonest of the serious neurological disorders. The total economic burden of epilepsy in the United Kingdom has been estimated to be pound 1930 m, with around pound 32 m spent on antiepileptic drug therapy alone. Despite the high level of expenditure on drug therapy for epilepsy there is very little information regarding the relative cost-effectiveness of the different drugs available. It is important to establish the relative cost-effectiveness of therapies to provide decision makers with the information necessary to allocate resources in a rational manner and thus achieve the highest benefit for available resources. In this study the cost-effectiveness of lamotrigine, vigabatrin and gabapentin was estimated by a cost minimization analysis for the first year of drug therapy using data based on published studies. In general, there was little difference between the initial direct costs of treatment, however, the fewer side-effects associated with gabapentin is reflected in the lower total costs of treatment in the first year resulting in savings of pound 18.52 per patient compared with lamotrigine and pound 47.18 compared with vigabatrin. Based on incidence data estimates this translates to estimated direct cost savings to the UK of between pound 166 680 and pound 424 620 per annum.