To assess the associations between birth weight or gestational age and risk of type 1 diabetes.
Population based cohort study by record linkage of the medical birth registry and the National Childhood Diabetes Registry.
Two national registries in Norway.
All live births in Norway between 1974 and 1998 (1 382 602 individuals) contributed a maximum of 15 years of observation, a total of 8 184 994 person years of observation in the period 1989 to 1998. 1824 children with type 1 diabetes were diagnosed between 1989 and 1998.
Main outcome measures
Estimates of rate ratios with 95% confidence intervals for type 1 diabetes from Poisson regression analyses.
The incidence rate of type 1 diabetes increased almost linearly with birth weight. The rate ratio for children with birth weights 4500 g or more compared with those with birth weights less than 2000 g was 2.21 (95% confidence interval 1.24 to 3.94), test for trend P=0.0001. There was no significant association between gestational age and type 1 diabetes. The results persisted after adjustment for maternal diabetes and other potential confounders.
There is a relatively weak but significant association between birth weight and increased risk of type 1 diabetes consistent over a wide range of birth weights.
What is already known on this topicResults of case-control studies of birth weight and risk of type 1 diabetes have been inconsistentIt is possible that a relatively weak association exists, and large studies are needed to find out if this is the caseWhat this study addsThis is the largest study of birth weight and type 1 diabetes published to date, and the first one to use a cohort designThe incidence of type 1 diabetes increased almost linearly with increasing birth weight over a wide range of birth weights, independent of gestational age, maternal diabetes, and other potential confoundersThe trend was highly significant, but the increment in risk with increasing birth weight was still relatively low