It is still unclear whether the reported increase in the prevalence of asthma is real or due to changes in diagnostic criteria. The objectives of this study were to compare the prevalence of diagnosed asthma with the prevalence of respiratory symptoms, and to compare the association between asthma and other atopic diseases in 1981 and 1994. The study populations comprised randomly selected school classes in Oslo in 1981 (n=1,772) and 1994 (n=2,577). The main outcomes in these comparable cross-sectional studies of children, 6-16 yrs of age, were parent-reported prevalence of diagnosed asthma, respiratory symptoms, eczema and hay fever. The questionnaire was identical in 1981 and 1994. The response rates were 94% (1,674 out of 1,772) in 1981 and 85% (2,188 out of 2,577) in 1994. The lifetime prevalence of asthma increased from 3.4% in 1981 to 9.3% in 1994; odds ratio (OR) 2.9 (95% confidence interval (95% CI) 2.1-4.0) comparing 1994 to 1981. The prevalence of occasional wheezing increased from 9.0 to 10.8%; OR 1.2 (95% CI 1.0-1.5), and attacks of wheezing from 3.7 to 6.8%; OR 1.8 (95% CI 1.3-2.5). Survival analyses for 3 year birth cohorts showed that asthma was more readily diagnosed in the latest birth cohort (1985-1988). The association between asthma and other atopic diseases decreased during the period under study. The increase in diagnosed asthma and respiratory symptoms supports a true increase in asthma. However, the larger increase in diagnosed asthma than wheezing and a reduced association between asthma and other atopic diseases suggest that the increase in asthma may be explained, in part, by changes in diagnostic criteria.