In the JournalsPerspective

Type 1 diabetes status not associated with reading, math scores in children

Danish public school children with and without type 1 diabetes performed similarly on standardized reading and math tests, with minor score differences observed only in children with an HbA1c of at least 8.6%, according to findings published in JAMA.

Niels Skipper

“It has been unclear if the chronically elevated — and fluctuating — glucose levels common in type 1 diabetes affect school performance in children,” Niels Skipper, PhD, associate professor in the department of economics and business economics at Aarhus University, Denmark, told Endocrine Today. “In this study, we used national data on all Danish children enrolled in public schools. With this large data material, we did not find any differences in the test scores when comparing children with diabetes to children without diabetes.”

In a population-based, retrospective registry study, the researchers analyzed test scores from 608,655 second- to eighth-grade children attending Danish public schools who took at least one standardized test between 2011 and 2015 (mean age, 10 years; 51% boys). Students were identified through Statistics Denmark and linked with background information from administrative registries. The DanDiabKids database was used to identify children with diabetes who took at least one test (n = 1,878; mean diabetes duration, 4.5 years) and their yearly HbA1c levels. Primary outcome was pooled reading and math test scores (scores ranged from 1 to 100 for both reading and math). Researchers used linear regression models to estimate the relationship between diabetes status and test scores.

Student, sibling comparisons

Researchers identified 1,037,006 reading test scores and 524,764 math test scores, including 4,234 test scores for children with diabetes. Children with diabetes were more likely to miss a test vs. children without diabetes (6.3% vs. 3.8%; difference, 2.5 percentage points; 95% CI, 1.8-3.2).

The overall pooled test score was 56.11, and no statistically significant difference in test scores was observed when comparing children with vs. without diabetes (mean, 56.56 vs. 56.11; difference, 0.45; 95% CI, –0.31 to 1.22). The difference fell to 0.24 after adjusting for grade, topic and year-specific effects (95% CI, –0.9 to 1.39).

In a case-sibling analysis, researchers identified 2,966 test scores from children with diabetes (1,257 children) and 4,378 test scores from siblings without diabetes (1,606 children), and again observed no between-group differences in test scores.

Influence of HbA1c

When comparing children with diabetes with children without diabetes, researchers found that having an HbA1c of at least 7.5% was associated with lower test scores (mean, 53.84 vs. 56.11; difference, –2.26; 95% CI, –3.27 to –1.26); however, the association did not persist after adjustment for socioeconomic status. Having an HbA1c of at least 8.6% was also associated with poorer test scores (472 children; 867 test scores) with the difference still significant after adjustment for socioeconomic status (mean scores, 49.95 vs. 56.11; difference, –6.15; 95% CI, –7.8 to –4.5).

“There are a lot of things to worry about when living with type 1 diabetes; however, our study is suggestive that school performance should not be one of them,” Skipper said. “Good metabolic control in affected children is of course still warranted, and it is not clear that our results would apply to other countries.”

Additionally, diabetic ketoacidosis, diabetes onset at age 6 years or younger and a diabetes duration of at least 4 years were not associated with test scores when compared against the test scores of children without diabetes, according to the researchers.

“It is possible that advances in treatment modalities over recent decades — 64% of children with diabetes in this study used an insulin pump — have improved not only gaps in mortality and morbidity between individuals with diabetes and the overall population, but also have improved gaps in school performance,” the researchers wrote. “This study also documented that it is important to adjust for confounders such as socioeconomic status when assessing the association between diabetes control and test scores or school performance.” – by Regina Schaffer

For more information:

Niels Skipper, PhD, can be reached at Aarhus University, Department of Economics and Business Economics, Fuglesangs Allé 4, DK 8210 Aarhus V; email: nskipper@econ.au.dk.

Disclosures: One of the study authors reports he received speaking fees from Novo Nordisk; another author reports he has served as an adviser to Janssen, Medtronic and Novo Nordisk, owns shares in Novo Nordisk and has received speaking fees from Bayer, Medtronic, Novo Nordisk and Sanofi.

Danish public school children with and without type 1 diabetes performed similarly on standardized reading and math tests, with minor score differences observed only in children with an HbA1c of at least 8.6%, according to findings published in JAMA.

Niels Skipper

“It has been unclear if the chronically elevated — and fluctuating — glucose levels common in type 1 diabetes affect school performance in children,” Niels Skipper, PhD, associate professor in the department of economics and business economics at Aarhus University, Denmark, told Endocrine Today. “In this study, we used national data on all Danish children enrolled in public schools. With this large data material, we did not find any differences in the test scores when comparing children with diabetes to children without diabetes.”

In a population-based, retrospective registry study, the researchers analyzed test scores from 608,655 second- to eighth-grade children attending Danish public schools who took at least one standardized test between 2011 and 2015 (mean age, 10 years; 51% boys). Students were identified through Statistics Denmark and linked with background information from administrative registries. The DanDiabKids database was used to identify children with diabetes who took at least one test (n = 1,878; mean diabetes duration, 4.5 years) and their yearly HbA1c levels. Primary outcome was pooled reading and math test scores (scores ranged from 1 to 100 for both reading and math). Researchers used linear regression models to estimate the relationship between diabetes status and test scores.

Student, sibling comparisons

Researchers identified 1,037,006 reading test scores and 524,764 math test scores, including 4,234 test scores for children with diabetes. Children with diabetes were more likely to miss a test vs. children without diabetes (6.3% vs. 3.8%; difference, 2.5 percentage points; 95% CI, 1.8-3.2).

The overall pooled test score was 56.11, and no statistically significant difference in test scores was observed when comparing children with vs. without diabetes (mean, 56.56 vs. 56.11; difference, 0.45; 95% CI, –0.31 to 1.22). The difference fell to 0.24 after adjusting for grade, topic and year-specific effects (95% CI, –0.9 to 1.39).

In a case-sibling analysis, researchers identified 2,966 test scores from children with diabetes (1,257 children) and 4,378 test scores from siblings without diabetes (1,606 children), and again observed no between-group differences in test scores.

Influence of HbA1c

When comparing children with diabetes with children without diabetes, researchers found that having an HbA1c of at least 7.5% was associated with lower test scores (mean, 53.84 vs. 56.11; difference, –2.26; 95% CI, –3.27 to –1.26); however, the association did not persist after adjustment for socioeconomic status. Having an HbA1c of at least 8.6% was also associated with poorer test scores (472 children; 867 test scores) with the difference still significant after adjustment for socioeconomic status (mean scores, 49.95 vs. 56.11; difference, –6.15; 95% CI, –7.8 to –4.5).

“There are a lot of things to worry about when living with type 1 diabetes; however, our study is suggestive that school performance should not be one of them,” Skipper said. “Good metabolic control in affected children is of course still warranted, and it is not clear that our results would apply to other countries.”

Additionally, diabetic ketoacidosis, diabetes onset at age 6 years or younger and a diabetes duration of at least 4 years were not associated with test scores when compared against the test scores of children without diabetes, according to the researchers.

“It is possible that advances in treatment modalities over recent decades — 64% of children with diabetes in this study used an insulin pump — have improved not only gaps in mortality and morbidity between individuals with diabetes and the overall population, but also have improved gaps in school performance,” the researchers wrote. “This study also documented that it is important to adjust for confounders such as socioeconomic status when assessing the association between diabetes control and test scores or school performance.” – by Regina Schaffer

For more information:

Niels Skipper, PhD, can be reached at Aarhus University, Department of Economics and Business Economics, Fuglesangs Allé 4, DK 8210 Aarhus V; email: nskipper@econ.au.dk.

Disclosures: One of the study authors reports he received speaking fees from Novo Nordisk; another author reports he has served as an adviser to Janssen, Medtronic and Novo Nordisk, owns shares in Novo Nordisk and has received speaking fees from Bayer, Medtronic, Novo Nordisk and Sanofi.

    Perspective

    Author Name

    This is a large retrospective cohort study using standardized test scores in reading and math to compare the school performance of Danish public school children with diabetes to students without diabetes. The mean test scores were not different between groups. Diabetes duration and age of onset were assessed in post hoc subgroup analyses and were not associated with test score differences. Diabetes control was assessed using an average of three HbA1c measurements, if available. However, it is unclear how many students had multiple HbA1c readings, as an isolated HbA1c value has less utility. Higher HbA1c may lead to poorer standardized test performance; however, adjustment for socioeconomic status was shown to attenuate this risk. Episodes of diabetic ketoacidosis did not contribute to lower scores after controlling for HbA1c and socioeconomic status.

    Several studies have assessed structural changes in the brain of children with type 1 diabetes in addition to cognitive functioning, mostly near diagnosis. With disparate results in the literature, this large retrospective study does lend support to a lack of type 1 diabetes impact on school performance; however, the role of metabolic control for students with type 1 diabetes requires further study. 

    Laura Jacobsen

    Assistant Professor, Pediatric Endocrinology
    University of Florida

     

    Disclosure: Jacobsen reports no relevant financial disclosures.