Presentation to primary care during the prodrome of type 1 diabetes in childhood: A case-control study using record data linkage

Objective To evaluate primary care presentations during the prodrome (12 months prior to onset type-1 diabetes (T1D), with or without diabetic ketoacidosis [DKA]), to identify opportunities for earlier diagnosis. Methods This was a case-control study, linking 16 years of data from children (<= 15 years) registered at diagnosis of T1D, and routinely collected primary care records in Wales (United Kingdom). Controls (without T1D) were matched on a 3:1 ratio. Conditional logistic regression modeling was used to compare characteristics occurring in cases (children with T1D) and controls; and cases that presented with/without DKA. Results A total of 1345 children with T1D (19% DKA) and 4035 controls were identified. During the 12 months prior to diagnosis, cases were 6.5 times more likely to have at least one primary care contact (P < 0.001). One to 30 days prior to diagnosis, contacts relating to blood tests, fungal conditions, respiratory tract infections (RTIs), urinary conditions, vomiting, and weight were independently associated with T1D, as were contacts relating to blood tests, between 91 and 180 days prior to diagnosis. Children with a contact up to a month prior to diagnosis, relating to RTIs, antibiotic prescriptions, and vomiting, were more likely to present in DKA, as were boys (P = 0.047). Conclusion There are opportunities in primary care for an earlier diagnosis of T1D in childhood. These data could be used to create a predictive diagnostic tool, as a potential aid for primary care health professionals, to prevent presentation in DKA.