Our aim was to study the association between serum 25-hydroxyvitamin D (25OHD) concentration and islet autoimmunity and type 1 diabetes in children with an increased genetic risk of type 1 diabetes.Serum samples for 25OHD measurements were obtained in the Trial to Reduce IDDM in the Genetically at Risk (TRIGR) ancillary study (Divia) from children in 15 countries. Case children (n = 244) were defined as having positivity for at least two out of four diabetes-associated autoantibodies measured at any one sample. For each case child, two control children were selected matched for country and date of birth (±1 year) (n = 488). Of the case children, 144 developed type 1 diabetes. Serum 25OHD was measured repeatedly in infancy and childhood and was compared according to age at the first seroconversion (at 6, 12 and 18 months prior to and at seroconversion) and calendar age (0, 6, 12 and 18 months).In children with islet autoimmunity, mean serum 25OHD concentration was lower 18 months prior to the age of first seroconversion of the case children compared with the control children (57.7 vs 64.8 nmol/l, p = 0.007). In children with type 1 diabetes (n = 144), mean serum 25OHD concentration was lower 18 months prior to the age of the first seroconversion (58.0 vs 65.0 nmol/l, p = 0.018) and at the calendar age of 12 months (70.1 vs 75.9 nmol/l, p = 0.031) than in their control counterparts. Analyses were adjusted for month of sample collection, human leucocyte antigen genotype, maternal type 1 diabetes and sex.The results suggest that early postnatal vitamin D may confer protection against the development of type 1 diabetes.ClinicalTrials.gov NCT00179777.