Outcome was studied of all children and adolescents with pseudoseizures without epilepsy, who were referred and treated as in-patients and/or day patients in the tertiary psychiatric ward at Birmingham Children's Hospital, UK between 1988 and 1994. Information was derived from case-note data. Freedom from seizures for six months was defined as 'cure', as no recurrences after this period were noted. Kaplan-Meier survival analysis was used. Seventeen patients were identified: 15 females and two males; mean age at presentation to the tertiary psychiatric service was 12 years 9 months (SD 26 months; range 8 years 3 months to 15 years 9 months). Fourteen participants recovered and resumed regular school attendance. Three were lost to follow-up due to being referred on to other agencies before recovery because they became too old for the service. Recovery followed an exponential distribution, with a mean symptom survival time following treatment of 1.5 years. These results are consistent with a treatment effect. Younger age at presentation, female sex, having more types of seizures, and not receiving both in-patient and outpatient treatment were associated with better outcome. The deteriorating prognosis with age at treatment suggests prompt identification may be important in ensuring a good outcome.