| Abstract: | SummaryA recent review found that economic assessment of epilepsy treatment relies largely on hypothetical modelling of outcomes and combining these with resource and cost data from different sources. Prospective evaluations combining cost studies with outcome assessments are lacking. However, such a prospective observational study has been carried out previously, but only partially reported. We present a comprehensive cost-utility analysis of adjunctive newer antiepileptic drugs (AEDs) based on observational data from that study, and assess the uncertainty of the results using bootstrapping.A total of 125 patients with intractable epilepsy were recruited. Each patient was about to start treatment with a new adjunctive AED [clobazam, (non-proprietary) gabapentin (Neurontin®, Parke-Davis, UK), lamotrigine (Lamictal®, GlaxoSmithKline, UK), topiramate (Topamax®, Janssen-Cilag, UK), or vigabatrin (Sabril®, Aventis Pharma, UK)]. Patients completed semi-structured interviews on resource use, side effects, and the EuroQol EQ-5D. Patients were followed up for 6 months. Patient-specific cost and utility data were analysed separately for each AED on an intent-to-treat basis. Uncertainty in the estimated incremental cost-utility ratios was quantified using the non-parametric bootstrap method, and cost-effectiveness acceptability curves were calculated.At 6 months, 78 patients were still on their prescribed drug. Only topiramate and vigabatrin patients showed an increase in EQ-5D scores, and therefore dominatedother AEDs. Topiramate had an incremental cost-effectiveness ratio of £7,869/QALY compared with vigabatrin, and had more than a 50% chance of being optimal if the ceiling ratio was above £10,000/QALY.Observational studies provide a valuable source of information for the economic evaluation of AEDs. In this study non-parametric bootstrapping was used to confirm the cost-effectiveness of adjunctive topiramate for patients with refractory epilepsy. |
