To estimate the 11-year incidence trend of diabetic ketoacidosis (DKA) at and after the diagnosis of type 1 diabetes.

A retrospective cohort study using a population-based administrative cohort diagnosed with type 1 diabetes at <20 years of age from 2002 to 2012 in British Columbia, Canada. DKA at (1 episode per individual) and DKA after (multiple episodes per individual) the diagnosis of diabetes were defined as DKA occurring ≤14 days or >14 days, respectively, from diagnosis, identified using International Classification of Diseases, 9th and 10th editions codes. Incidence rate ratios were estimated using Poisson regression and DKA trends using Joinpoint regression analyses.

There were 1519 individuals (mean age at first-DKA, 12.6 ± 5.9 years; 50% male) with ≥1 DKA episode identified. Of 2615 incident cases of type 1 diabetes, there were 847 (32.4%; mean age, 9.9 ± 4.8 years; 52% male) episodes of DKA at the diagnosis of diabetes. Among prevalent cases of type 1 diabetes (1790 cases in 2002 increasing to 2264 in 2012), there were 1886 episodes of DKA after the diagnosis of diabetes (mean age at first DKA, 15.7 ± 5.2 years). The rates per 100 person-years of DKA at diabetes diagnosis (ranging from 24.1 in 2008 to 37.3 in 2006) and DKA after diabetes diagnosis (ranging from 4.9 in 2002 to 7.7 in 2008) remained stable. Females showed a 67% higher rate of incidence of DKA after the diagnosis of diabetes compared with their male counterparts (incidence rate ratio, 1.67; 95% CI, 1.50-1.86; P < .001), adjusted for the temporal trend by fiscal year. Younger age at diagnosis (<5 years) was associated with a greater risk of DKA at the time of diabetes diagnosis and older children (≥10 years) had a greater risk of DKA after the diagnosis of diabetes.

The risk of DKA at the time of diagnosis of diabetes was greater with younger age and the risk of DKA after the diagnosis of diabetes was higher in females and older children and youth.