Background: Computer-assisted-telephone-interviewing (CATI), widely used in market research, could be a useful alternative for conducting diagnostic interviews in epilepsy epidemiology. Methods: We administered a diagnostic seizure questionnaire by CATI, interpreting the responses with standardized classification guidelines, compared against an epilepsy specialist's assessment, for agreement [Kappa statistic (κ)], sensitivity, specificity, positive predictive value, negative predictive value and Youden's Index (YI). Results: 99 outpatients with 382 lifetime events participated: 22 generalized-onset epilepsy [16 Idiopathic Generalized Epilepsy (IGE)], 59 partial-onset epilepsy, 12 non-epileptic and 6 uncertain. We observed almost perfect agreement in diagnosing epilepsy (κ=0.94), seizure-onset types (κ=0.84), simple or complex partial seizures (κ=0.87), any generalized non-convulsive seizure (κ=0.82), and IGE (κ=0.82). Although substantial, agreement was not as close for secondarily generalized seizures (κ=0.74), and generalized tonic-clonic seizures (κ=0.79). This related more to under-recognition of individual generalized non-convulsive seizures rather than misinterpretation of partial seizures. Discussion: Epilepsy diagnostic questionnaires administered by CATI and interpreted with standardized diagnostic guidelines can effectively classify epilepsy, most seizure types and IGE in outpatients with suspected seizures. Applying this diagnostic method in 'field' settings will allow firmer conclusions to be drawn on its wider epidemiological utility.