Current evidence suggests that epilepsy outcome for people with a learning disability is poor, with the majority remaining refractory to treatment. There is very little evidence from randomised controlled trials in this population and despite the many uncontrolled variables it is suggested that an outcome audit may be one method of adding to the evidence base. This audit reports on the outcome for 37 patients with learning disability and refractory epilepsy. All patients were seen for the first time before March 2001 and the mean number of seizures for the sample was 10.4 per month. The exit audit included all patients at a date 2 years after their initial visit following a programme of medication changes. The mean seizure frequency had reduced to 5.9 per month. Ten patients had become seizure-free and 76% had experienced an improvement in seizure frequency overall. A simple questionnaire was administered to carers and relatives at the exit audit in an attempt to establish an impression of global changes in alertness, assertiveness and challenging behaviour following interventions. Sixty-five percent of patients were regarded as being more aware and interactive with their surroundings following medication changes and 49% were reported to be more assertive. Thirty percent presented with an increase in behaviours regarded as challenging and 22% were reported to present with less challenging behaviour. The results of this audit suggest that the outcome for the majority of patients with learning disability and refractory epilepsy may be better than that has been previously reported.