ESPE Yearbook of Paediatric Endocrinology

To read the full abstract: JAMA 2019; 32:484–492

Type 1 diabetes (T1DM) is clearly associated with higher risks for cardiovascular disease and late complications, such as retinopathy and nephropathy. A putative impact of T1DM on neurocognitive development is less clear. For example, it is unclear whether T1DM is associated with school performance in children.

This study compared standardized reading and mathematics test scores of school children with and without T1DM, using population-based retrospective data from January 1, 2011, to December 31, 2015 on Danish public school children attending grades 2, 3, 4, 6, and 8. Mathematics test scores were accessible in 524 764 children, and reading performance scores in 1 037 006 (mean (SD) age: 10.31 (2.42) years; 51% male). In linear regression models outcomes with and without adjustment for socioeconomic characteristics were analyzed. Primary outcomes were defined as pooled test scores in mathematics and reading.

T1DM was diagnosed in 2031 children. They had similar combined mathematics and reading scores (mean 56.56) to children without diabetes (56.11). No significant difference in combined scores was found between groups in a linear regression model fully adjusted for grade, test topic, year, and socioeconomic status (mean difference: 0.45 (95% CI, −0.58 to 1.49)).

In summary, there was no difference in standardized reading and mathematics test scores of children with T1DM compared to children without diabetes. These data are truly important, considering the plausible risks of both glycopenia on the brain during hypoglycemic events as well as chronic and short-term hyperglycemia on cognitive performance. Some previous studies had shown lower academic performance in children with T1DM, while others had shown no difference. Therefore, the issue remains controversial. Differences between study findings could be explained by effects of hypoglycemia on working memory and language processing. Furthermore, in the current large retrospective Danish cohort, while there was no difference between children with and without T1DM, there was an albeit nonsignificant association of lower test scores with an HbA1c above 7.5%. Also, children with HbA1c <7.5% scored higher in the tests than the population mean. There was no association with DKA at onset nor with diabetes duration of more than 4 years. However, children with diabetes had missed more tests than their healthy school peers and this points to a possible susceptibility of not reaching their true potential simply through lower school/test attendance rather than through direct neuroglycopenic events.