A retrospective epidemiological study using primary data sources and external validation was conducted to assess the incidence of childhood (0 to 14 years) Type 1 diabetes in Hungary (excluding the Budapest region) over a 10-year period. Hospital records and regional diabetes centres' registries served as primary data sources, and data were validated by using the records of summer camps for diabetic children. The degree of ascertainment was 96.2%. Age-specific incidence rates increased until puberty with peak incidence in girls at 10 yr, and in boys at 13 yr. There was a significant seasonal variation of onset with peaks in autumn and winter. The incidence was slightly lower amongst boys than amongst girls. A statistically significant increase in incidence was evident in the last decade, from 3.8 per 100,000 in 1978 to 8.2 per 100,000 in 1987.