The question "Has your child ever had wheezing or whistling in the chest at any time in the past?" is a simple and widely used proxy measure for the lifetime prevalence of asthma. Our aim was to test its validity in a longitudinal survey, comparing retrospective recall with prospective assessment of lifetime prevalence. A population-based cohort of 1,422 children, surveyed twice previously, was studied again at age 8-13 yrs by postal questionnaire using standardized questions from the International Study of Asthma and Allergies in Childhood (ISAAC). Of those traced (1,190) questionnaires were returned by 89%. The prevalence of current wheeze was higher than in the previous surveys (20.5% versus 12.4% and 12.5%). Reported "wheeze ever" increased significantly from survey 1 (15.6%) to survey 2 (22.4%) and survey 3 (39.2%) and was very similar to the cumulative lifetime prevalence assessed prospectively over three surveys (42.8%). The retrospective question had a good negative predictive value (97%) and a reasonable positive predictive value (65%) compared to prospective assessment. Children reporting "wheeze ever" (but not current wheeze) in surveys 1 and 2 had at survey 3 an asthma prevalence higher than never-wheezers but lower than current-wheezers. It is concluded that retrospective recall of wheeze at age 8-13 yrs is a valid proxy measure for the lifetime prevalence of wheeze.