Few prospective, population-based, long-term follow-up studies exist on people with epilepsy. Still fewer reports cover social outcome. Overall mortality is two to three times higher than expected. The contribution of epilepsy is variable. Importantly, the type of epilepsy syndrome and gender must be considered in the estimation of mortality rates in epilepsy. Sudden unexpected death and its mechanisms also need further consideration. Approximately. two thirds of surviving patients will be in terminal remission twenty years after onset of epilepsy and half of them are seizure-free without medication. The best independent predictors of remission are absence of organic brain damage, low intensity seizure propensity and good early effect of drug therapy. The long-term outcome is often predictable by observation of the early outcome of seizures. One third of children with epilepsy are mentally retarded. Poor social outcome is related to associated neurological disabilities, drug resistant seizures and polytherapy. However, even patients with uncomplicated epilepsy, idiopathic etiology and terminal remission without medication do less favourably than their matched controls in basic and vocational education, and reproductive activity. The employability of this subgroup, however, does not differ significantly from that of controls, compared with approximately 60% of all people with epilepsy. Further research is needed particularly to enable a better determination of predictors of long-term outcome, recurrence of seizures after drug withdrawal and the role of drug therapy in long-term prognosis.