STUDY OBJECTIVE: To determine the frequency of relapse of asthma in young adults in remission at 18 years of age, during a follow-up period of 8 years, and to identify possible prognostic markers for relapse. DESIGN: Longitudinal study of birth cohort (n = 1,037) born in New Zealand in 1972-1973. SETTING: University hospital research clinic. MEASUREMENTS: Participants were assessed at 9, 11, 13, 15, 18, 21, and 26 years of age using a respiratory questionnaire (all ages), spirometry (all ages), bronchodilator response (18 years and 26 years of age), methacholine challenge (9, 11, 13, 15, and 21 years of age), and allergen skin-prick testing (13 years and 21 years of age). RESULTS: Approximately one third of study members (35%) with asthma in remission at 18 years of age relapsed by 21 years or 26 years of age. Atopy and lower FEV(1)/FVC ratio at 18 years of age were significant independent prognostic factors for relapse in multiple logistic regression analyses. Increased responsiveness to methacholine (provocative concentration < 8 mg/mL) or bronchodilator (improvement in FEV(1) >/= 10%) at 21 years of age were more common among those with relapse, but the positive and negative predictive values for a previous positive methacholine challenge test result at 15 years of age were low. Asthma after relapse was generally mild (mean FEV(1) 97.1% predicted). Totally new adult asthma developed by 26 years of age in 9% of study members who had no asthma or wheezing at any time up to 18 years of age. CONCLUSIONS: Subsequent relapse of previously diagnosed asthma in remission at 18 years of age occurs in one in three young adults. Such relapse is not easily predicted, especially by measurements of airway responsiveness. A history of asthma currently in remission should not be used to prejudice employment opportunities for young adults.