A major component of the population of people who have epilepsy are people with a learning disability. As a group, such individuals often have complex epilepsy which is refractory to treatment. Current available measures to assess the outcomes of therapeutic interventions in epilepsy are based on seizure frequency, seizure severity and quality-of-life measures, but have not been validated in people with a learning disability. Thus, we do not know if such measures of outcome serve the needs of this group. This review examines how able we are to assess the efficacy of our interventions to control epilepsy in people with learning disability. It is suggested that a standard data set is necessary as the basis of the assessment of any therapeutic intervention. Central components of this data set would encompass a definition of important characteristics of an individual, a description of their epilepsy and an assessment of the impact of their condition on both their own and their carer's health. The approach to obtaining this information should employ a methodology which can allow for environmental influences.