In the mid 1980s the incidence of coeliac disease in Swedish children below 2 years of age increased threefold within a few years, and after a 10-year high incidence period returned equally rapidly to the previous level. Analysing the epidemic with respect to any change in female to male ratio over time, or shift in age at diagnosis, may increase the understanding of coeliac disease aetiology. In a population-based incidence study of childhood coeliac disease, 2151 cases (811 boys/1340 girls) were diagnosed from 1973 to 1997. Incidence rates and relative risks (RRs) were calculated by gender, age at diagnosis and calendar time. Cumulative incidences by age and gender were calculated for different birth cohorts. A twofold higher risk (RR: 1.9, 95% confidence interval (CI) 1.7-2.1) for coeliac disease in girls as compared to boys prevailed throughout the epidemic. Further, during the post-epidemic period there was an upward shift in age at diagnosis. So far, however, a majority of the cases diagnosed at older ages belong to birth cohorts of the epidemic period, i.e. cohorts that already had a high coeliac disease risk before 2 years of age. Our results suggest that girls as compared to boys may be genetically more vulnerable to environmental exposures influencing the immunological processes towards coeliac disease. Further, an increased risk for coeliac disease during the first years of life due to, for example, unfavourable infant dietary habits, may result in an increased total childhood risk for coeliac disease. A longer follow-up, even into adulthood, is needed to determine whether or not the lifetime risk has changed.