Objective: To determine the recent trends in incidence, to analyze the age and geographic distribution, as well as the seasonal pattern of type 1 diabetes in Hungarian children aged 0-14 yr for the period from 1978 to 1998.
Methods: Primary ascertainment of cases was by retrospective (1978-88) and by prospective (1989-98) registration using hospital notifications. The level of ascertainment was estimated by the capture-recapture method. The temporal trend was estimated by fitting Poisson regression models to the yearly incidence data. Roger's test was used to investigate possible seasonal variation in time of diagnosis. Heterogeneity between geographic areas was assessed by Poisson regression.
Results: A total of 2616 patients (1214 in the first 11 yr, 1402 in the remaining 10 yr) were identified; the male:female ratio was 0.93. The overall standardized incidence rate was 7.87 (95% CI = 7.57-8.18) per 100,000 person-yr, the lowest in the youngest (0-4 yr) and highest in the oldest (10-14 yr) age group. There was an increasing trend in incidence with a largest relative annual increase in the youngest age group. Seasonal and regional variations in incidence were also observed.
Conclusion: Our 21-yr study shows dynamic changes in incidence of childhood type 1 diabetes in Hungary, probably reflecting changes in the environment.