Population density determines the direction of the association between ambient ultraviolet radiation and type 1 diabetes incidence

Type 1 diabetes incidence has increased rapidly over the last 20 years, and ecological studies show inverse latitudinal gradients for both incidence and prevalence. Some studies have found season of birth or season of diagnosis effects. Together these findings suggest an important role for environmental factors in disease etiology.

To examine whether type 1 diabetes incidence varies in relation to ambient ultraviolet radiation (UVR) in Australian children.

We used case records of 4773 children aged 0-14 yr from the Australian National Diabetes Register to estimate type 1 diabetes incidence in relation to residential ambient UVR, both as a continuous variable and in four categories. We examined season of birth and season of diagnosis and variation in these parameters and in age at diagnosis, in relation to ambient UVR.

Overall incidence was 20 per 100 000 population with no sex difference. There was a statistically significant trend toward winter diagnosis (adjusted RR = 1.22, 95% CI 1.13-1.33, p<0.001) but no apparent season of birth effect. Incidence in the highest UVR category was significantly lower than in the lowest UVR category (RR = 0.85, 95% CI 0.75-0.96). We found an inverse association between incidence and ambient UVR that was present only at low population densities; at high population densities type 1 diabetes incidence increased with increasing ambient UVR.

In low population density, largely rural environments, ambient UVR may better reflect the personal UV dose, with the latter being protective for the development of type 1 diabetes. This effect is lost or reversed in high population density, largely urban, environments.