The aim was to explore familial aggregation of diabetes in genetically related and unrelated individuals.We included citizens from Danish nationwide registries between 1995 and 2018 and calculated rate ratios (RR) of diabetes based on family relation using Poisson regression.Of 7.3 million individuals eligible for inclusion, we identified 343,237 (4.7%) with diabetes. The RR of diabetes was 2.02 (95% CI: 1.99-2.05; p < 0.0001) if any relative had diabetes, 1.79 (95% CI: 1.76-1.83) if a father had diabetes, and 2.06 (95% CI: 2.02-2.10) if a mother had diabetes. If both parents had diabetes, the RR was 3.40 (95% CI: 3.24-3.56). Among full siblings, the RR for developing diabetes was 2.77 (95% CI: 2.71-2.84) and 5.76 (95% CI: 5.00-6.63) for twins. For second-degree relatives, half siblings with a common mother had a RR of 2.35 (95% CI: 2.15-2.56), and with a common father 1.99 (95% CI: 1.81-2.17). Furthermore, the RR was 1.60 (95% CI: 1.56-1.64) if a wife had diabetes, and 1.41 (95% CI: 1.38-1.44) if a husband had diabetes. A subgroup analysis of individuals receiving insulin only treatment (N = 23,054) demonstrated a similar risk pattern, although with slightly higher risk estimates.Family aggregation of diabetes is associated with genetic disposition with maternal status being the predominant factor. Furthermore, we observed increased risk of diabetes in second-degree relatives, and between unrelated spouses, indicating that environmental factors influence diabetes risk substantially.