Antiepileptic medication should be administered to patients after two or more seizures, although there is disagreement on how to handle individuals following their first unprovoked seizure. Patients who experienced their first tonic-clonic seizure were randomized to either immediate treatment (carbamazepine, phenytoin, phenobarbital, or sodium valproate) or wait until they experienced another seizure before receiving therapy in a multicenter, randomized, open trial. During follow-up, seizure recurrence occurred in 52 (24%) of the 215 patients who were randomly assigned to immediate therapy and 85 (42%) of the 204 patients who were randomly assigned to delayed treatment. Relapse was significantly predicted by age, benzodiazepine immediate treatment of the seizure, remote causative variables, and abnormalities in the EEG. 87% of the patients who were treated right away had no seizures for a year, and 68% had none for two years. However, these objectives were only marginally more successful in originally untreated individuals (83% and 60%). Patients who were treated after their first seizure and those who were treated following a seizure relapse both had the same time-dependent likelihood of living one or two years seizure-free. None of the factors that were predictive relapse predictors were significantly correlated with the likelihood of achieving one or two year’s seizure control. Anticonvulsants lessen the likelihood of relapse in patients who have had their first tonic-clonic seizure, yet 50% of untreated people never have another seizure. Furthermore, the management of the initial seizure has no impact on the likelihood of long-term remission.