The goal of this study was to collect prospective longitudinal information on the development of an epidemiologically defined cohort of patients with Tourette syndrome. These data may improve prognostic understanding of the condition. This information will also be important for specification of an adult phenotype for genetic marker studies. A prospective longitudinal cohort study was conducted. Fifty-four of 73 patients from our 1986 prevalence study of Tourette syndrome in North Dakota school-aged children were eligible for inclusion. The subjects were diagnosed in 1984 and 1985. We were able to interview 39 of 54 eligible patients for 507 person-years of follow-up. For the cohort, tic severity declined by 59%, global assessment of functioning improved by 50%, and the average number of comorbidities decreased by 42%. Forty-four percent of patients were essentially symptom free at follow-up. Only 22% were on medication as adults. Tourette syndrome is a developmental neuropsychiatric disorder with a long-term course that is favorable for most patients. Males demonstrated substantially more variability in improvement but overall demonstrated more improvement than females.