Educational outcomes among children with type 1 diabetes: Whole-of-population linked-data study

Background: Challenges with type 1 diabetes (T1D) blood glucose management and illness-relatedschool absences potentially influence children's educational outcomes. However, evidence about the impact of T1D on children's education is mixed. The objectives were to estimate the effects of T1D on children's educational outcomes, and compare time since T1D diagnosis (recent diagnosis [<= 2 years] and 3 to 10 years long exposure) on educational outcomes. Methods: This whole-of-population study used de-identified, administrative linked-data from the South Australian Early Childhood Data Project. T1D was identified from hospital ICD-10-AM diagnosis codes (E10, ranging E101 to E109), from 2001 to 2014. Educational outcomes were measured in grade 5 by the National Assessment Program-Literacy and Numeracy (NAPLAN, 2008-2015) for children born from 1999 to 2005. Analyses were conducted using augmented inverse probability of treatment weighting. Multiple imputations was used to impute missing data. Results: Among 61 445 children born in South Australia who had undertaken NAPLAN assessments, 162 had T1D. There were negligible differences in the educational outcomes of children with and without T1D, and between recently diagnosed and those with longer exposure. For example, the mean reading score was 482.8 +/- 78.9 for children with T1D and 475.5 +/- 74.3 for other children. The average treatment effect of 6.8 (95% CI - 6.3-19.9) reflected one-tenth of a SD difference in the mean reading score of children with and without T1D. Conclusion: Children with T1D performed similarly on literacy and numeracy in grade 5 (age similar to 10-years) compared to children without T1D. This could be due to effective T1D management.