Objective Investigating demographic and socioeconomic factors as intersecting rather than as separate dimensions may improve our understanding of the heterogeneous distribution of type 2 diabetes in the population. However, this complexity has scarcely been investigated and we still do not know the accuracy of these factors for predicting type 2 diabetes. Improved understanding of the demographic and socioeconomic disparities predicting type 2 diabetes risk in the population would contribute to more precise and effective public health interventions.

Research design and methods We analyzed the risk of type 2 diabetes among 4 334 030 individuals aged 40–84 years who by 2010 had resided in Sweden for at least 5 years. We stratified the study population into 120 strata defined by categories of age, gender, income, education, and immigration status. We calculated measures of absolute risk (prevalence) and relative risk (prevalence ratio), and quantified the discriminatory accuracy of the information for predicting type 2 diabetes in the population.

Results The distribution of type 2 diabetes risk in the population was highly heterogeneous. For instance, immigrated men aged 70–79 years with low educational achievement and low income had a risk around 32 times higher than native women aged 40–49 years with high income and high educational achievement (ie, 17.6% vs 0.5%). The discriminatory accuracy of the information was acceptable.

Conclusion A more detailed, intersectional mapping of socioeconomic and demographic distribution of type 2 diabetes can assist in public health management aiming to reduce the prevalence of the disease.