Aims/hypothesis: Our study aimed to describe the incidence of type 1 diabetes in children below 15 years of age in Norway during the period 1989-2012 and to assess the regional variation during 2004-2012. We further set out to estimate the completeness of ascertainment in the Norwegian Childhood Diabetes Registry (NCDR).
Methods: Incident cases of type 1 diabetes were registered in the NCDR and incidence rates were modelled using Poisson regression. Ascertainment for 2005-2008 was estimated using capture-recapture methodology by using data from the Norwegian Prescription Database (NorPD), a nationwide register established in 2004, which included insulin prescribed and dispensed at pharmacies to individual patients. Population data were obtained from Statistics Norway.
Results: Observed incidence rates for 1989-2012 suggested three distinct time segments: in 1989-1996, the average incidence rate was 22.6 per 100,000 person-years (95% CI 21.4, 23.7); in 1996-2004, the average incidence rate was 28.4 per 100,000 person-years (95% CI 27.3, 29.6); and from 2004 to 2012, the average incidence rate per 100,000 person-years was 32.7 (95% CI 31.5, 34.0). After adjustment for age and sex, the estimated change per year was 1.8% for 1989-1996 (95% CI -0.07, 3.6; p = 0.059), 3.4% for 1996-2004 (95% CI 2.2, 4.7; p < 0.0001) and 0.3% for 2004-2012 (95% CI -0.9, 1.6; p = 0.64). The highest incidence was in the age group 10-14 years for both sexes. A significant regional variation in incidence was observed (p < 0.001). Completeness of ascertainment in the NCDR was estimated to be 91%.
Conclusions/interpretation: The previously observed increase in incidence of type 1 diabetes has levelled off and remained essentially constant at 32.7 per 100,000 person-years during 2004-2012. There is a significant variation in type 1 diabetes incidence within Norway.