We designed a multicenter case control study to evaluate whether chronic alcoholism and alcohol are risk factors for developing a first generalized tonic-clonic epileptic seizure. Cases were 278 patients (92 women, 186 men), with a first generalized tonic-clonic seizure (either idiopathic or symptomatic), matched to 556 controls for center, sex, age, and weekday of the seizure. Information on risk factors was obtained through a questionnaire. This paper focuses on the study design (selection of cases and choice of controls), the validity of the instruments for data collection, and the strategies of study conduction (interviews, use of proxy respondents). Inter-rater agreement was excellent for drinking status, patterns of drinking, and broad indicators of consumption (yes/no). Agreement was variable for the number of servings/week, but was excellent (r = 0.91) for a summary of daily alcohol intake.