AIM: To determine if there is any regional celiac disease (CD) risk variation in the Swedish childhood population. METHODS: Prospective nationwide Swedish incidence register of CD in children 0-15 years of age, with the present analysis covering the period from 1998 to 2003. ESPGHAN diagnostic criteria for CD were used. Regions were classified according to the Nomenclature of Territorial Units for Statistics. The incidence rate for each region, gender, age group and year of diagnosis was calculated. RESULTS: A regional variation in CD risk was demonstrated. The childhood populations in 'West Sweden' and 'Småland and the islands', situated in the southern part of the country, had a significantly higher incidence rate compared to in 'North Middle Sweden' and 'Stockholm', situated in the central part. This regional variation was not explained by variations in risk by gender, age at diagnosis or year of diagnosis. CONCLUSION: The Swedish regional variation in CD risk supports multifactorial disease aetiology. Continued efforts are warranted to define factors, besides gluten exposure, that modulate CD risk.