Abstract
Background and Objective
The impact of type 1 diabetes mellitus (T1D) on academic performance is inconclusive. This study aims to compare scholastic performance and high‐school completion in young people hospitalized with T1D compared to matched peers not hospitalized with diabetes.
Research Design
Retrospective case‐comparison cohort study.
Method
A population‐level matched case‐comparison study of people aged ≤18 hospitalized with T1D during 2005–2018 in New South Wales, Australia using linked health‐related and education records. The comparison cohort was matched on age, gender, and residential postcode. Generalized linear mixed modeling examined risk of school performance below the national minimum standard (NMS) and generalized linear regression examined risk of not completing high school for young people hospitalized with T1D compared to peers. Adjusted relative risks (ARR) were calculated.
Results
Young females and males hospitalized with T1D did not have a higher risk of not achieving the NMS compared to peers for numeracy (ARR: 1.19; 95%CI 0.77–1.84 and ARR: 0.74; 95%CI 0.46–1.19) or reading (ARR: 0.98; 95%CI 0.63–1.50 and ARR: 0.85; 95%CI 0.58–1.24), respectively. Young T1D hospitalized females had a higher risk of not completing year 11 (ARR: 1.73; 95%CI 1.19–2.53) or 12 (ARR: 1.65; 95%CI 1.17–2.33) compared to peers, while hospitalized T1D males did not.
Conclusions
There was no difference in academic performance in youth hospitalized with T1D compared to peers. Improved glucose control and T1D management may explain the absence of school performance decrements in students with T1D. However, females hospitalized with T1D had a higher risk of not completing high school. Potential associations of this increased risk, with attention to T1D and psycho‐social management, should be investigated.
Keywords: high‐school completion, school performance, type 1 diabetes, young people
1. INTRODUCTION
Type 1 diabetes mellitus (T1D) is one of the most common chronic health conditions experienced by young people. 1 Worldwide, there are an estimated 2.58 million young people aged ≤19 years with T1D, 2 with a further 132,000 young people diagnosed with T1D annually. 3 In Australia, there are an estimated 6,500 children aged ≤14 years living with T1D. 4 T1D can influence the lives of young people in a variety of ways, including by affecting their general health, 1 , 5 family, peer and community interactions 1 and both their school performance 2 , 6 and attendance. 7
T1D can be challenging to manage and without metabolic control, young people may experience acute complications, such as hypoglycaemia or ketoacidosis, 2 or more long‐term chronic complications, such as retinopathy, neuropathy, or nephropathy. 1 There is also evidence that some young people with T1D experience mild deficits in their cognitive, motor, or visuospatial skills. 8 , 9 , 10 , 11 Less ideal glycaemic control leading to frequent hypoglycaemic or keto‐acidotic episodes could affect neuropsychological development, and thus, cognitive performance, including attention and memory skills. 12 , 13 Cognitive deficits have the potential to influence academic performance among young people with T1D. 2 , 6
There have been mixed findings on the impact of T1D on academic performance in young people. 12 , 14 , 15 , 16 Systematic reviews and meta‐analyses have identified that young people with T1D can experience some cognitive impairment. 10 , 17 , 18 Two studies conducted in Sweden and Arizona have identified that young people with T1D have poor academic performance compared to their peers or siblings without T1D, 11 , 19 while other studies conducted in Australia, Scotland, and Washington have all found no difference in academic performance compared to peers. 14 , 15 , 20 , 21 Three studies have identified differences in academic performance between sub‐groups of young people with T1D, depending on their level of metabolic control, with young people with better metabolic control performing better academically. 12 , 14 , 15 As previous research has indicated mixed findings in relation to whether T1D has an adverse impact on academic performance, the authors hypotheses that there is a difference between the academic performance of young people with T1D compared to their matched peers. Specifically, this study aims to compare scholastic performance and high school completion in young people hospitalized with T1D compared to matched peers not hospitalized with diabetes.
2. METHOD
This is a retrospective population‐based, case‐comparison cohort study of young people hospitalized with T1D aged ≤18 years in New South Wales (NSW), Australia using linked birth, health, education and mortality data collections from 1 January 2005 to 31 December 2018. 22 Ethical approval and a waiver of consent was obtained from the NSW Population and Health Services Research Ethics Committee (2018HRE0904).
2.1. Data sources
Health service use information was obtained from emergency department (ED) visit and hospital admission data collections. ED visits to public hospitals in NSW included information on arrival and departure dates, visit type, and provisional diagnosis. Hospital admissions were to public or private hospitals and contained information on patient demographics, diagnoses, separation type (e.g., hospital transfer, death), and clinical procedures. Information on mortality was obtained from the NSW Registry of Births, Deaths, and Marriages and young people who died during the study period were excluded from analyses (Supplementary Figure 1).
School performance and parental demographic information was obtained from the National Assessment Plan for Literacy and Numeracy (NAPLAN) assessments conducted annually in May at government, Catholic, and independent schools from 2008 to 2018. NAPLAN assessments were conducted for young people in primary school grade 3 (age 7–9 years) and 5 (age 9–11 years), and high‐school grades 7 (age 11–13 years) and 9 (age 13–15 years), and included assessments in numeracy and reading. Each assessment score is translated into proficiency bands that indicate whether performance was above, at, or below the national minimum standard (NMS). Inability to achieve the NMS indicates that a young person will have difficulty making progress in school without assistance. 23 For each NAPLAN assessment, scores represent the same level of achievement over time. 24
A young person's attendance, absence, withdrawal (e.g., philosophical objections to testing, or religious beliefs) or exemption due to significant disability for the NAPLAN assessments was obtained (Supplementary Table 1). Young people exempt from sitting an assessment, due to a severe disability or language difficulties were rated as achieving below the NMS as per technical guidelines. 25
A young person was identified as having a language background other than English (LBOTE) if either they or their parents or guardians spoke a language other than English at home. 24 Where there were multiple records of the parents' level of education and occupation over time, the highest level of education and occupation of either parent was identified. Information on high‐school completion at years 10 (15–16 years of age), 11 (17–18 years of age), and 12 (17–18 years of age) were obtained for each young person from the Record of School Achievement and the Higher School Certificate.
The Center for Health Record Linkage (CHeReL) randomly identified the population comparison group and linked the birth, health, education, and mortality records using probabilistic record linkage. Upper and lower probability cut‐offs for a link were 0.75 and 0.25 and record groups with probabilities between the cut‐offs were clerically reviewed.
2.2. Case inclusion criteria
Cases included young people with a year of birth ≥1997 who were aged ≤18 years at their index hospitalization who had a principal or additional diagnosis of T1D in their principal diagnosis or up to 50 additional diagnosis classifications (International Classification of Diseases, 10th Revision, Australian Modification (ICD‐10‐AM: E10) during 1 January 2005 to 31 December 2018. Cases were included if both the numeracy and reading NAPLAN assessments in a school grade were completed by the young person.
2.3. Comparison group criteria
The population‐based comparison group included young people not hospitalized with any form of diabetes (ICD‐10‐AM: E10‐E14) from 1 July 2001 to 31 December 2018. Comparison group members were randomly selected from NSW birth records matched 1:1 on age, gender, and residential postcode to their counterpart. The timeframe for selection of comparisons included a 3.5 year wash out period prior to the case selection timeframe to avoid the potential selection of comparisons who may have been hospitalized with any form of diabetes prior to the case criteria timeframe.
2.4. Identification of chronic health conditions
Common chronic health conditions experienced by young people were identified from prior studies of pediatric comorbidities 26 , 27 , 28 and were health conditions reasonably expected to last 12 months or need ongoing healthcare. 26 For this study, comorbidities were identified using diagnosis classifications from ICD‐10‐AM and a 3‐year lookback period (to 1 January 2002) (Supplementary Table 2).
2.5. Socioeconomic status and geographical location
The young person's postcode of residence was used to classify socioeconomic disadvantage into quintiles from most (i.e., 1) to least (i.e., 5) disadvantaged. 29 The Australian Statistical Geographical Standard (ASGS), 30 based on distance to service centers, was used to classify the residential postcode of the young person as either urban (i.e., major cities) or rural (i.e., inner and outer regional, remote, and very remote). The remoteness area of the school was obtained from the NAPLAN data collection and was categorized as major city, inner regional, outer regional, or remote. 30
2.6. Data organization and analysis
Data analysis was conducted using SAS 9.4 (SAS Institute, Cary NC). All hospital episodes of care related to the same event were linked to form a period of care. Chi‐square tests of independence and Wilcoxon Mann–Whitney tests were used to examine characteristics of young people hospitalized for T1D and their matched counterparts. The number of ED visits, hospitalisations, and hospital length of stay (LOS) before and during, and after, the index hospital admission were identified for cases and their matched peers. The calculation of hospital LOS was cumulative and included hospital transfers.
To examine school performance, generalized linear regression using PROC GENMOD assessed the difference in proportions of performance below the NMS for each assessment at each school grade (i.e., 3, 5, 7, and 9) for young people hospitalized with T1D and their matched counterparts (Supplementary Table 3). Unadjusted and adjusted relative risks (ARR) and 95% confidence intervals (CIs) were calculated.
For the numeracy and reading assessments, each model was fitted using generalized estimating equations (GEE) with binomial distribution and a log function. Purposeful forward selection was used 31 to sequentially add covariates to the models 22 and significance was assessed using p‐values (p < 0.05) to examine overall effect in the models. The final models examined predictors of performance below the NMS and included diabetes status (Y/N), gender, socioeconomic status, highest level of education for any parent (i.e., ≤Year 12; certificate I‐IV, trade or diploma; bachelor or higher degree; or other), and highest level of occupation of any parent (i.e., senior manager/qualified professional; business manager/associate professional; trades/clerks/skilled office/sales and service; machine operators/ hospitality/assistants/laborers; or not in paid work in last 12 months). Matching variables were included in each model to control for any possible confounding from the matching variables. 32
Generalized linear mixed modeling (GLMM) was conducted to perform multi‐level modeling of school performance below the NMS for each of the numeracy and reading assessments for cases and their matched peer who had completed assessments in different school grades. For each NAPLAN assessment, PROC GLIMMIX was used with a binary distribution, log link function, and Kenward and Roger denominator degrees of freedom. The residual option of the random statement was used to model R‐side covariance and data were analyzed to account for within student correlation in the longitudinal data and repeated measurements using an autoregressive covariance structure. Unadjusted RR and ARRs and 95% CIs were generated. The final models included: diabetes status (Y/N), gender, comorbidity status (Y/N), NAPLAN grade (i.e., 3, 5, 7, 9), school sector (i.e., government, Catholic, independent), LBOTE, socioeconomic status and highest level of education and occupation for any parent.
Factors associated with high‐school completion at either year 10, 11, or 12 for cases compared to their matched peers were examined using generalized linear regression using PROC GENMOD. For each grade, a model was fitted using GEE with binomial distribution and a log function. ARR and 95% CIs were calculated. The final models examined predictors of not completing high school and included: diabetes status (Y/N), gender, comorbidity status (Y/N), LBOTE, socioeconomic status, highest level of education for any parent/guardian, and geographic location of residence.
3. RESULTS
There were 833 young people hospitalized with T1D for whom a matched comparison was identified and who completed NAPLAN assessments in Grade 3; 673 in Grade 5; 516 in Grade 7; and 324 in Grade 9 during 2015–2018. There were 843 young people with T1D hospitalized with a matched peer identified who could have completed year 10; 775 in year 11; and 682 in year 12 of high school.
Across the school grades 3–9, there was a higher proportion of females (between 52.2% and 54.5%) compared to males and a higher proportion of young people in urban areas (between72.4% and 74.4%) compared to rural areas hospitalized with T1D. Most young people (≥79%) were from English‐speaking households and almost all (≥97%) had no other chronic comorbidities identified. There were no significant differences between parental education or occupation for young people with T1D and their matched peers in each school grade. Young people hospitalized with T1D had higher mean number of ED visits, hospital admissions, and hospital LOS than their matched counterparts in each school grade (Table 1).
TABLE 1.
Demographic and healthcare use characteristics of young people hospitalized with type 1 diabetes mellitus and their matched comparison by grade, linked health, and school performance data NSW, 2005–2018
| Grade 3 a | Grade 5 b | Grade 7 c | Grade 9 d | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Characteristics | Case (n = 833) | Comparison (n = 833) | Case (n = 673) | Comparison (n = 673) | Case (n = 516) | Comparison (n = 516) | Case (n = 324) | Comparison (n = 324) | ||||||||
| n | % | n | % | n | % | n | % | n | % | n | % | n | % | n | % | |
| Gender | ||||||||||||||||
| Male | 379 | 45.5 | 379 | 45.5 | 306 | 45.5 | 306 | 45.5 | 243 | 47.1 | 243 | 47.1 | 155 | 47.8 | 155 | 47.8 |
| Female | 454 | 54.5 | 454 | 54.5 | 367 | 54.5 | 367 | 54.5 | 273 | 52.9 | 273 | 52.9 | 169 | 52.2 | 169 | 52.2 |
| Location of residence e | ||||||||||||||||
| Urban | 606 | 72.8 | 606 | 72.8 | 487 | 72.4 | 487 | 72.4 | 378 | 73.3 | 378 | 73.3 | 241 | 74.4 | 241 | 74.4 |
| Rural | 226 | 27.1 | 226 | 27.1 | 186 | 27.6 | 186 | 27.6 | 138 | 26.7 | 138 | 26.7 | 83 | 25.6 | 83 | 25.6 |
| Socioeconomic status e | ||||||||||||||||
| Most disadvantaged | 164 | 19.7 | 164 | 19.7 | 132 | 19.6 | 132 | 19.6 | 94 | 18.2 | 94 | 18.2 | 53 | 16.4 | 53 | 16.4 |
| 2 | 187 | 22.5 | 187 | 22.5 | 151 | 22.4 | 151 | 22.4 | 112 | 21.7 | 112 | 21.7 | 66 | 20.4 | 66 | 20.4 |
| 3 | 190 | 22.8 | 190 | 22.8 | 157 | 23.3 | 157 | 23.3 | 129 | 25.0 | 129 | 25.0 | 78 | 24.1 | 78 | 24.1 |
| 4 | 87 | 10.4 | 87 | 10.4 | 64 | 9.5 | 64 | 9.5 | 49 | 9.5 | 49 | 9.5 | 33 | 10.2 | 33 | 10.2 |
| Least disadvantaged | 204 | 24.5 | 204 | 24.5 | 169 | 25.1 | 169 | 25.1 | 132 | 25.6 | 132 | 25.6 | 94 | 29.1 | 94 | 29.1 |
| LBOTE f | ||||||||||||||||
| Non‐LBOTE | 656 | 78.8 | 678 | 81.4 | 541 | 80.4 | 553 | 82.2 | 423 | 82.0 | 439 | 85.1 | 270 | 83.3 | 274 | 84.6 |
| LBOTE | 171 | 20.5 | 151 | 18.1 | 130 | 19.3 | 117 | 17.4 | 92 | 17.8 | 76 | 14.7 | 53 | 16.4 | 49 | 15.1 |
| Health condition | ||||||||||||||||
| 0 | 818 | 98.2 | 818 | 98.2 | 659 | 97.9 | 660 | 98.1 | 503 | 97.5 | 504 | 97.7 | 315 | 97.2 | 315 | 97.2 |
| ≥1 | 15 | 1.8 | 15 | 1.8 | 14 | 2.1 | 13 | 1.9 | 13 | 2.5 | 12 | 2.3 | 9 | 2.8 | 9 | 2.8 |
| Parent highest level of education | ||||||||||||||||
| Year 11 or equivalent | 52 | 6.2 | 63 | 7.6 | 46 | 6.8 | 60 | 8.9 | 34 | 6.6 | 52 | 10.1 | 23 | 7.1 | 29 | 9.0 |
| Year 12 or equivalent | 49 | 4.9 | 32 | 3.8 | 41 | 6.1 | 28 | 4.2 | 28 | 5.4 | 25 | 4.8 | 16 | 4.9 | 17 | 5.3 |
| Certificate I–IV or trade | 255 | 30.6 | 263 | 31.6 | 205 | 30.5 | 216 | 32.1 | 166 | 32.2 | 158 | 30.6 | 102 | 31.5 | 91 | 28.1 |
| Advanced diploma/diploma | 127 | 15.3 | 141 | 16.9 | 109 | 16.2 | 115 | 17.1 | 85 | 16.5 | 91 | 17.6 | 45 | 13.9 | 59 | 18.2 |
| Bachelor degree or higher | 319 | 38.3 | 300 | 36.0 | 261 | 38.8 | 242 | 36.0 | 196 | 38.0 | 183 | 35.5 | 134 | 41.4 | 121 | 37.44 |
| Not stated/not known | 31 | 3.7 | 34 | 4.1 | 11 | 1.6 | 12 | 1.8 | 7 | 1.4 | 7 | 1.4 | 4 | 1.2 | 7 | 2.2 |
| Parent highest occupation | ||||||||||||||||
| Senior manager/qualified professional | 275 | 33.0 | 261 | 31.3 | 236 | 35.1 | 217 | 32.2 | 181 | 35.1 | 175 | 33.9 | 125 | 38.6 | 119 | 36.7 |
| Business management/associate professional | 196 | 23.5 | 208 | 25.0 | 158 | 23.5 | 168 | 25.0 | 120 | 23.3 | 130 | 25.2 | 78 | 24.1 | 74 | 22.8 |
| Trades, clerks, skilled office, sales and service | 189 | 22.7 | 187 | 22.5 | 152 | 22.6 | 152 | 22.6 | 119 | 23.1 | 107 | 20.7 | 66 | 20.4 | 72 | 22.2 |
| Machine operators, hospitality, assistants, laborers | 98 | 11.8 | 101 | 12.1 | 79 | 11.7 | 81 | 12.0 | 56 | 10.9 | 65 | 12.6 | 35 | 10.8 | 35 | 10.8 |
| Not in paid work in last 12 months | 50 | 6.0 | 46 | 5.5 | 31 | 4.6 | 35 | 5.2 | 27 | 5.2 | 27 | 5.2 | 13 | 4.0 | 13 | 4.0 |
| Not known | 25 | 3.0 | 30 | 3.6 | 17 | 2.5 | 20 | 3.0 | 13 | 2.5 | 12 | 2.3 | 7 | 2.2 | 11 | 3.4 |
| Health care use | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ED visits | 4.0 | 6.9 | 1.8 | 3.6 | 4.3 | 6.6 | 2.1 | 3.8 | 5.4 | 6.4 | 2.3 | 3.8 | 5.5 | 5.5 | 2.4 | 3.8 |
| Hospital admissions | 2.5 | 3.8 | 0.5 | 1.0 | 2.7 | 3.6 | 0.5 | 1.0 | 3.6 | 4.4 | 0.6 | 1.1 | 3.9 | 4.8 | 0.6 | 1.0 |
| Hospital length of stay | 5.6 | 9.9 | 1.5 | 4.0 | 6.2 | 9.8 | 1.5 | 3.8 | 8.5 | 12.3 | 1.5 | 3.6 | 9.3 | 13.1 | 1.4 | 3.2 |
Abbreviation: LBOTE, language background other than English.
Grade 3 chi‐square test: LBOTE p = 0.4; health conditions; p = 1.0; parent highest level of education p = 0.3; parent highest occupation p = 0.9; and Wilcoxon Mann–Whitney tests: ED visits p < 0.0001; hospital admissions p < 0.0001; hospital length of stay p < 0.0001.
Grade 5 chi‐square test: LBOTE p = 0.6; health conditions p = 0.8; parent highest level of education p = 0.4; parent highest occupation p = 0.9; and Wilcoxon Mann–Whitney tests: ED visits p < 0.0001; hospital admissions p < 0.0001; hospital length of stay p < 0.0001.
Grade 7 chi‐square test: LBOTE (excluding not known) p = 0.2; health conditions p = 0.8; parent highest level of education p = 0.4; parent highest occupation p = 0.9; and Wilcoxon Mann–Whitney tests: ED visits p < 0.0001; hospital admissions p < 0.0001; hospital length of stay p < 0.0001.
Grade 9 chi‐square test: LBOTE (excluding not known) p = 0.7; health conditions p = 1.0; parent highest level of education p = 0.5; parent highest occupation p = 0.9; and Wilcoxon Mann–Whitney tests: ED visits p < 0.0001; hospital admissions p < 0.0001; hospital length of stay p < 0.0001.
Excludes not known location of residence and socioeconomic status for 1 case and 1 comparisons in grade 3.
Language background other than English and LBOTE was not known for 10 cases and 9 comparisons.
Young people predominantly attended government schools in major cities and there were no significant differences by school sector or location of the school for young people hospitalized with T1D, compared to their matched counterparts, in each grade. The proportion of students hospitalized with T1D who did not achieve the NMS for their numeracy or reading assessment in grades 3–9, did not differ significantly from their matched peers (Table 2).
TABLE 2.
School and NAPLAN assessment characteristics of young people hospitalized with type 1 diabetes mellitus and their matched comparison by grade, linked health, and school performance data NSW, 2005–2018
| Grade 3 a | Grade 5 b | Grade 7 c | Grade 9 d | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Characteristics | Case (n = 833) | Comparison (n = 833) | Case (n = 673) | Comparison (n = 673) | Case (n = 516) | Comparison (n = 516) | Case (n = 324) | Comparison (n = 324) | ||||||||
| n | % | n | % | n | % | n | % | n | % | n | % | n | % | n | % | |
| School sector | ||||||||||||||||
| Government | 571 | 68.6 | 576 | 69.2 | 457 | 67.9 | 457 | 67.9 | 296 | 57.4 | 281 | 54.5 | 191 | 59.0 | 169 | 52.2 |
| Catholic | 174 | 20.9 | 168 | 20.2 | 145 | 21.6 | 134 | 19.9 | 125 | 24.2 | 136 | 26.4 | 82 | 25.3 | 89 | 27.5 |
| Independent | 87 | 10.4 | 89 | 10.7 | 70 | 10.4 | 82 | 12.2 | 94 | 18.2 | 99 | 19.2 | 51 | 15.7 | 66 | 20.4 |
| Remoteness area of school | ||||||||||||||||
| Major city | 599 | 71.9 | 574 | 68.9 | 479 | 71.2 | 465 | 69.1 | 371 | 71.9 | 365 | 70.7 | 234 | 72.2 | 229 | 70.7 |
| Inner regional | 181 | 21.7 | 203 | 24.4 | 151 | 22.4 | 170 | 25.3 | 118 | 22.9 | 121 | 23.5 | 70 | 21.6 | 80 | 24.7 |
| Outer regional/remote | 52 | 6.2 | 56 | 6.7 | 42 | 6.2 | 38 | 5.7 | 26 | 5.0 | 30 | 5.8 | 20 | 6.2 | 15 | 4.6 |
| NAPLAN assessment e | ||||||||||||||||
| Numeracy (below NMS) | 32 | 3.8 | 33 | 4.0 | 18 | 2.7 | 27 | 4.0 | 16 | 3.1 | 17 | 3.3 | 7 | 2.2 | 3 | 0.9 |
| Reading (below NMS) | 28 | 3.4 | 32 | 3.8 | 34 | 5.1 | 40 | 5.9 | 18 | 3.5 | 17 | 3.3 | 10 | 3.1 | 15 | 4.6 |
Grade 3 chi‐square test: school sector (excluding 1 home‐schooled) p = 0.9; remoteness area of school (excluding 1 missing) p = 0.4; numeracy p = 0.9; reading p = 0.6.
Grade 5 chi‐square test: school sector (excluding 1 home‐schooled) p = 0.5; remoteness area of school (excluding 1 missing) p = 0.5. numeracy p = 0.2; reading p = 0.5.
Grade 7 chi‐square test: school sector (excluding 1 home‐schooled) p = 0.6; remoteness area of school (excluding 1 missing) p = 0.8. numeracy p = 0.9; reading p = 0.9.
Grade 9 chi‐square test: school sector p = 0.2; remoteness area of school p = 0.5. numeracy p = 0.2; reading p = 0.3.
NAPLAN: National Assessment Plan for Literacy and Numeracy; NMS: National Minimum Standard.
After adjustment for covariates, young females hospitalized with T1D did not have a higher risk of not achieving the NMS compared to matched peers on school assessments for numeracy (ARR: 1.19; 95%CI 0.77–1.84) or reading (ARR: 0.98; 95%CI 0.63–1.50). Young males hospitalized with T1D similarly had no higher risk of not achieving the NMS for numeracy (ARR: 0.74; 95%CI 0.46–1.19) or reading (ARR: 0.85; 95%CI 0.58–1.24), compared to their matched counterparts (Table 3).
TABLE 3.
Multilevel model of type 1 diabetes mellitus associated with a below NMS NAPLAN assessment for young people with an index hospitalization for during 2005–2018 compared to a matched comparison, linked health and school performance data NSW
| Numeracy | Reading | |||||||
|---|---|---|---|---|---|---|---|---|
| RR a | 95%CI | ARR b | 95%CI | RR a | 95%CI | ARR b | 95%CI | |
| All persons b | 0.90 | 0.65–1.24 | 0.96 | 0.70–1.31 | 0.88 | 0.66–1.17 | 0.91 | 0.69–1.20 |
| Male c | 0.69 | 0.43–1.11 | 0.74 | 0.46–1.19 | 0.82 | 0.57–1.19 | 0.85 | 0.58–1.24 |
| Female c | 1.14 | 0.73–1.77 | 1.19 | 0.77–1.84 | 0.97 | 0.61–1.54 | 0.98 | 0.63–1.50 |
Abbreviations: ARR, adjusted relative risks; CI, confidence interval.
Unadjusted relative risk. Analyses exclude 1 (0.04%) case with missing socioeconomic status, 10 (0.4%) with missing LBOTE and 3 (0.1%) who were home schooled.
Adjusted for diabetes status, gender, NAPLAN grade, LBOTE, socioeconomic status, parental education and occupation, and school sector.
Adjusted for diabetes status, NAPLAN grade, LBOTE, socioeconomic status, parental education and occupation, and school sector.
Young females hospitalized with T1D showed no difference in completion of year 10, but had a higher risk for not completing year 11 (ARR: 1.73; 95%CI 1.19–2.53) or year 12 (ARR: 1.65; 95%CI 1.17–2.33) compared to their matched peers. Young males hospitalized with T1D had no higher risk for not completing high school in years 10, 11, or 12, compared to their matched counterparts (Table 4).
TABLE 4.
Analysis of characteristics associated with not completing high school for young people with an index hospitalisation with type 1 diabetes mellitus compared to a matched comparison by grade, linked health, and school performance data NSW, 2005–2018
| Year 10 n = 843 in each cohort | Year 11 n = 775 in each cohort | Year 12 n = 682 in each cohort | |||||||
|---|---|---|---|---|---|---|---|---|---|
| No school completion | n | % a | n | % a | n | % a | |||
| Case | 23 | 2.7 | 136 | 17.6 | 164 | 21.1 | |||
| Comparison | 17 | 2.0 | 104 | 13.4 | 126 | 18.5 | |||
| Gender | n b | ARR c | 95%CI | n b | ARR d | 95%CI | n b | ARR e | 95%CI |
| All persons | 23 | 1.29 | 0.68–2.42 | 136 | 1.31** | 1.02–1.69 | 164 | 1.31** | 1.04–1.65 |
| Male | 8 | 0.79 | 0.31–2.01 | 63 | 1.04 | 0.73–1.47 | 79 | 1.10 | 0.80–1.50 |
| Female | 15 | 2.00 | 0.80–4.97 | 73 | 1.73* | 1.19–2.53 | 85 | 1.65* | 1.17–2.33 |
Abbreviations: ARR, adjusted relative risks; CI, confidence interval.
Percent calculated for young people in case and comparison cohorts not completing the school grade.
Number of students hospitalized with diabetes not completing the year.
Year 10 all person adjusted relative risk for diabetes status, gender, socioeconomic status (excluding 5 missing), and parental education. Both males and females adjusted for diabetes status, socioeconomic status, and parental education.
Year 11 all person adjusted relative risk for diabetes status, gender, LBOTE, socioeconomic status (excluding 4 missing), residential geographic location, and parental education. Males and females adjusted for diabetes status, LBOTE, socioeconomic status, and parental education.
Year 12 all person adjusted relative risk for diabetes status, gender, LBOTE, socioeconomic status (excluding 4 missing), and parental education. Males and females adjusted for diabetes status, LBOTE, socioeconomic status, and parental education.
p < 0.004; **p < 0.03.
4. DISCUSSION
This study found no association between young people hospitalized with T1D and the risk of not achieving minimum standards for numeracy and reading in standardized school‐based assessments, compared to matched peers for young females or males not hospitalized with diabetes. It also found that young females with T1D had a higher risk of not completing year 11 or 12, but not year 10, compared to matched peers. However, it did not find a higher risk of not completing high school for young males hospitalized with T1D, compared to their matched counterparts. This study has confirmed that academic performance is not adversely affected by T1D, although females with T1D were less likely to complete high school.
The impact of T1D on the lives and abilities of young people is complex and does not appear to be uniform. While T1D has been shown in previous research to negatively affect facets of academic skill development, 11 the current study found no discernible difference in academic performance compared to matched peers. Similar to the current study, comparisons of academic performance between young people with T1D and non‐diabetic peers in Scotland 14 and Denmark 15 found no significant differences in performance. In addition, two studies conducted in Australia, one in Western Australia 20 and another in South Australia among grade 5 students, 16 similarly found no differences in the academic performance of young people with T1D compared to non‐diabetic peers.
Differences in the findings on the impact of T1D on academic performance may be explained by the fact that previous studies that found students with T1D had worse academic performance than peers, included student cohorts from the 1970s and 1980s 19 or used subjective teacher ratings of performance, rather than standardized assessments. 11 The non‐objective assessment of student performance may have resulted in biased reporting of academic performance. It is also likely that significant advances in treatment regimens and modalities, as well as the increased availability of diabetes insulin pump and continuous glucose monitoring (CGM) technologies, which have been associated with improved outcomes in severe hypoglycaemia 33 and diabetic ketoacidosis, 13 as well as overall diabetes control, 34 have improved T1D outcomes 20 and subsequently reduced the potential for negative consequences of T1D on student academic performance.
One of the studies that identified decrements in academic performance for young people with T1D compared to healthy peers, did not find consistent performance across each subject. 19 This Swedish study examined performance in mathematics, Swedish, English and sports in young people with T1D compared to peers and found an increased likelihood of not passing, or receiving a lower mark, for mathematics and Swedish, but no significant difference in performance in sport or English for young people with T1D 19 compared to peers. However, the research did identify that young people with T1D had a reduced likelihood of achieving a high mark (e.g., distinction) in sport or English, compared to peers. 19 That consistent performance decrements for young people with T1D are not evident across school subjects illustrates the complexity of the impact of T1D on academic performance. It is also possible that annual standardized school assessments, such as NAPLAN, may not adequately capture everyday difficulties that could be experienced by young people with T1D in the school environment, 1 , 16 and that more nuanced, frequent, and regular monitoring of academic performance is required.
The current study was unable to take into account age at onset of T1D and level of metabolic control (i.e., glycosylated hemoglobin AIc (HbA1c). One prior study found that young people who had early onset of T1D (i.e., <7 years of age) had poorer school performance compared to young people with later onset T1D. 19 There is evidence of structural changes in the brains of young people diagnosed with T1D at a younger age, 35 thought to be due to episodes of significant hypoglycaemia, and it is possible that a longer disease duration of T1D, because of early onset age, may be associated with difficulties with cognition, memory, and attention. 36 Yet other studies did not identify any impact of age of onset for T1D on academic achievement. 11 , 12 , 15 , 21 It is probable that once an intensive treatment regimen has commenced for a young person, the age of onset could become less of an issue for academic performance, 11 particularly where insulin pump and integrated CGM technologies can ameliorate the incidence of profound hypo‐ and hyper‐glycaemia. 33 However, most research‐lacked information regarding adherence by young people with T1D to a treatment regimen.
In comparing academic performance on numeracy and literacy for sub‐groups of young people with T1D, previous research found that young people who experienced metabolic control (i.e., HbA1c levels <8%) performed better academically than those with less ideal metabolic control (i.e., HbA1c levels >10%). 12 Similarly, research in Scotland 14 identified that young people with T1D with better metabolic control performed better academically than students with poor metabolic control. Whereas, a study in Denmark 15 found that young people with metabolic control performed worse on assessments of literacy and numeracy than students with poor metabolic control. The authors speculated that the negative association between metabolic control and school performance could be the result of the provision of increased learning support services to students with T1D. 15
The current study found that young females hospitalized with T1D had a higher risk of not completing high school than matched peers. Similarly, Fletcher and Richards 37 identified that young people with T1D were 5% to 7% more likely to drop out of high school than their peers. Adolescent females with T1D experience more issues with metabolic control, 38 more distress, 39 and have a higher risk of developing an eating disorder than young males. 40 , 41 For females with T1D, transitioning from childhood to adolescence may have wider psychosocial implications that contribute to young females having higher risk of early school leaving. The additional psychological distress and a higher prevalence of eating disorders among young people with T1D 42 attests to the need for regular psychological screening of young people with T1D. 43
Other studies did not find a similar association with T1D and high‐school dropout. In Scotland, there was no increase in the proportion of young people with T1D leaving school before age 16 compared to peers. 14 Likewise, in Finland there was no difference in young people with T1D leaving school at age 16 compared to peers, including no difference by sex, 44 nor in Sweden. 19 However, in Sweden, once the age of onset of T1D was taken into account, children diagnosed with T1D before 2 years of age had a higher likelihood of dropping out of high school. 19 The differences found between studies could relate to case identification criteria, with the current study only including young people hospitalized with T1D, therefore was more likely to include more poorly controlled cases. Other studies identified cases from clinical registers or from prescribed medications to identify any young person with T1D, 44 which could also include young people who had not been hospitalized with T1D.
Advances in treatment modalities, 15 , 20 compared to some of the earlier studies of the impact of T1D on academic performance, better glycaemic control through improved diabetes management, 11 , 20 such as using insulin pumps rather than insulin injection modalities, 16 use of CGM technology, especially where CGM is linked to pump technology, and better communication and training regarding T1D management strategies, including with school staff 1 , 16 are all likely to have contributed to obtaining the similar educational outcomes for young people with T1D compared to non‐diabetic peers. Further support for students with T1D is being provided in Australia with the launch in 2020 of the national diabetes in schools program in Australia. 45 The program aims to provide information, training, and support for educators, families, and health professionals to better support young people with T1D. 45
Further research examining T1D and academic performance in Australia could link health and education performance data to primary care attendances, diabetes registry and pharmaceutical dispensing data, as this would allow confirmation of age at T1D diagnosis, diabetes management, and outcomes. Additional research is also required to further tease out the impact of recent versus latent T1D diagnosis on a young person's academic performance and also identify if there are any later health and social outcomes, including on employment opportunities.
This study's strengths were that it was a large population‐level cohort study, linking health and standardized academic assessment records over a 13‐year period and was able to adjust for key factors that may influence academic performance, such as socioeconomic status and parental education. However, the study had some limitations. The study only included young people who had been hospitalized with T1D, so did not include young people presenting solely to tertiary center diabetes outpatient departments, primary care or other health professionals for treatment. Although, hospitalized young people with T1D were more likely to have less perfect diabetes control. There was no information in the hospitalization data regarding metabolic control (i.e. HbA1c levels, time in range on CGM or the proportion of time high, very high, or low), number of severe hypo/hyperglycaemic episodes, frequency of diabetic ketoacidosis, prescribed insulin doses/kg, age of T1D onset for the young person, or access to specialist care from endocrinologists. Only chronic health conditions relevant to a hospital admission are recorded in diagnosis classifications and it is likely that some comorbidities were not identified. This is particularly the case for the comparison cohort, where 68.6% had not been hospitalized, leaving no opportunity to identify comorbid conditions, despite the 3‐year lookback period.
A small number of residential postcodes were not able to be provided to the authors and socioeconomic status for these records was not able to be identified. The recency of postcode of residence may vary between data collections, which may affect estimates of socioeconomic status. Data validity was not able to be assessed and it is possible that there could be some data misclassification in hospital or education records. Visits to private hospital EDs were not able to be accessed for this study. A higher proportion of young people with T1D were absent for NAPLAN assessments compared to their matched counterparts. The current study was not able to take into account school clustering and no information was available concerning any tutoring or supplementary education services that a young person may have received.
5. CONCLUSION
There was no difference in the academic performance of either young females or males with T1D compared to matched peers. Better glucose control and awareness of strategies to manage T1D, including diabetes education programs for young people in schools, may explain the absence of school performance decrements for students with T1D. However, young females hospitalized with T1D did have a higher risk of leaving high school than their matched peers. Additional research is needed to explore psychological and social factors, as well as T1D control factors, that may contribute to this increased risk, and suggests the need for routine psychological screening as recommended in best practice guidelines.
CONFLICT OF INTEREST
The authors declare they have no conflict of interest.
AUTHOR CONTRIBUTIONS
RM, CC, and AM were all involved in study concept and design. RM acquired and organized the data, conducted the analysis and wrote the first draft of the manuscript. TB‐P provided statistical assistance for the GLMM analysis. All authors were involved in interpretation of data and critical revision of the manuscript.
PEER REVIEW
The peer review history for this article is available at https://publons.com/publon/10.1111/pedi.13317.
Supporting information
Appendix S1: Supporting information
ACKNOWLEDGMENTS
The authors wish to thank the NSW Ministry of Health for providing access to the ED visit, hospitalization, and mortality data, NSW Department of Education for providing access to school enrolment and completion information, the NSW Education Standards Authority for providing access to the NAPLAN data, and the Centre for Health Record Linkage for conducting the data linkage. This study was funded by a philanthropic donor to Macquarie University. Open access publishing facilitated by Macquarie University, as part of theWiley ‐ Macquarie University agreement via the Council of Australian University Librarians.
Mitchell RJ, McMaugh A, Woodhead H, et al. The impact of type 1 diabetes mellitus in childhood on academic performance: A matched population‐based cohort study. Pediatr Diabetes. 2022;23(3):411‐420. doi: 10.1111/pedi.13317
Funding information Macquarie University
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the NSW Health Department, NSW Department of Education, and NSW Education Standards Authority. Restrictions apply to the availability of these data, which were used under licence for the current study, so are not publicly available.
REFERENCES
- 1. Kise SS, Hopkins A, Burke S. Improving school experiences for adolescents with type 1 diabetes. J School Health. 2017;87(5):363‐375. [DOI] [PubMed] [Google Scholar]
- 2. Patterson CC, Karuranga S, Salpea P, et al. Worldwide estimates of incidence, prevalence and mortality of type 1 diabetes in children and adolescents: results from the international diabetes federation diabetes atlas. Diabetes Res Clin Pract. 2019;157:107842. [DOI] [PubMed] [Google Scholar]
- 3. International Diabetes Federation . Diabetes in Children and Adolescents. International Diabetes Federation; 2019. [Google Scholar]
- 4. Australian Institute of Health and Welfare . Australia's Children. AIHIW. https://www.aihw.gov.au/reports/children-youth/australias-children/contents/health/children-diabetes, 2021.
- 5. Murillo M, Bel J, Pérez J, et al. Health‐related quality of life (HRQOL) and its associated factors in children with type 1 diabetes mellitus (T1DM). BMC Pediatr. 2017;17(1):1‐9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Knight MF, Perfect MM. Glycemic control influences on academic performance in youth with type 1 diabetes. School Psychol. 2019;34(6):646‐655. [DOI] [PubMed] [Google Scholar]
- 7. Glaab LA, Brown R, Daneman D. School attendance in children with type 1 diabetes. Diabet Med. 2005;22(4):421‐426. [DOI] [PubMed] [Google Scholar]
- 8. Omladič JŠ, Ozimič AS, Vovk A, et al. Acute hyperglycemia and spatial working memory in adolescents with type 1 diabetes. Diabetes Care. 2020;43(8):1941‐1944. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Cato MA, Mauras N, Mazaika P, et al. Longitudinal evaluation of cognitive functioning in young children with type 1 diabetes over 18 months. J Int Neuropsychol Soc. 2016;22(3):293‐302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Blasetti A, Chiuri RM, Tocco AM, et al. The effect of recurrent severe hypoglycemia on cognitive performance in children with type 1 diabetes: a meta‐analysis. J Child Neurol. 2011;26(11):1383‐1391. [DOI] [PubMed] [Google Scholar]
- 11. Parent KB, Wodrich DL, Hasan KS. Type 1 diabetes mellitus and school: a comparison of patients and healthy siblings. Pediatr Diabetes. 2009;10(8):554‐562. [DOI] [PubMed] [Google Scholar]
- 12. McCarthy AM, Lindgren S, Mengeling MA, Tsalikian E, Engvall J. Factors associated with academic achievement in children with type 1 diabetes. Diabetes Care. 2003;26(1):112‐117. [DOI] [PubMed] [Google Scholar]
- 13. Cameron FJ, Scratch SE, Nadebaum C, et al. Neurological consequences of diabetic ketoacidosis at initial presentation of type 1 diabetes in a prospective cohort study of children. Diabetes Care. 2014;37(6):1554‐1562. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Fleming M, Fitton CA, Steiner MF, et al. Educational and health outcomes of children treated for type 1 diabetes: Scotland‐wide record linkage study of 766,047 children. Diabetes Care. 2019;42(9):1700‐1707. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Skipper N, Gaulke A, Sildorf SM, Eriksen TM, Nielsen NF, Svensson J. Association of type 1 diabetes with standardized test scores of Danish school children. JAMA. 2019;321(5):484‐492. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Begum M, Chittleborough C, Pilkington R, et al. Educational outcomes among children with type 1 diabetes: whole‐of‐population linked‐data study. Pediatr Diabetes. 2020;21(7):1353‐1361. [DOI] [PubMed] [Google Scholar]
- 17. Naguib JM, Kulinskaya E, Lomax CL, Garralda ME. Neuro‐cognitive performance in children with type 1 diabetes—a meta‐analysis. J Pediatr Psychol. 2009;34(3):271‐282. [DOI] [PubMed] [Google Scholar]
- 18. Gaudieri PA, Chen R, Greer TF, Holmes CS. Cognitive function in children with type 1 diabetes: a meta‐analysis. Diabetes Care. 2008;31(9):1892‐1897. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Dahlquist G, Källén B, Group SCDS . School performance in children with type 1 diabetes—a population‐based register study. Diabetologia. 2007;50(5):957‐964. [DOI] [PubMed] [Google Scholar]
- 20. Cooper MN, McNamara KA, de Klerk NH, Davis EA, Jones TW. School performance in children with type 1 diabetes: a contemporary population‐based study. Pediatr Diabetes. 2016;17(2):101‐111. [DOI] [PubMed] [Google Scholar]
- 21. Semenkovich K, Patel P, Pollock A, et al. Academic abilities and glycaemic control in children and young people with type 1 diabetes mellitus. Diabet Med. 2016;33(5):668‐673. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Mitchell R, Cameron C, Lystad R, et al. Impact of chronic health conditions and injury on school performance and health outcomes in New South Wales, Australia: a retrospective record linkage study protocol. BMJ Paediatr Open. 2019;3(1):e000530. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. ACARA . National Assessment Program: Literacy and Numeracy. Sydney: ACARA; 2017.
- 24. Australian Curriculum Assessment and Reporting Authority . National Assessment Program–Literacy and Numeracy (NAPLAN) 2018: Technical Report. Sydney: ACARA; 2019.
- 25. Australian Curriculum Assessment and Reporting Authority . National Assessment Program–Literacy and Numeracy Achievement in Reading, Persuasive Writing, Language Conventions and Numeracy: National Report for 2014. Sydney: ACARA; 2014.
- 26. Miller C, Shi J, Wheeler K, et al. Chronic conditions and outcomes of pediatric trauma patients. J Trauma Acute Care. 2013;75(2):250‐257. [DOI] [PubMed] [Google Scholar]
- 27. Edwards J, Houtrow A, Vasilevskis E, et al. Chronic conditions among children admitted to US pediatric intensive care units: their prevalence and impact on risk for mortality and prolonged length of stay. Crit Care Med. 2012;40(7):2196‐2203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Mitchell R, Curtis K, Braithwaite J. Health outcomes and costs for injured young people hospitalised with and without chronic health conditions. Injury. 2017;48(8):1776‐1783. [DOI] [PubMed] [Google Scholar]
- 29. Australian Bureau of Statistics . Census of Population and Housing: Socio‐Economic Indexes for Areas (SEIFA), Australia. Catalogue no: 2033.0.55.001. Canberra: Australian Bureau of Statistics; 2011.
- 30. Australian Bureau of Statistics . 1270.0.55.005–Australian Statistical Geography Standard (ASGS): Volume 5—Remoteness Structure, July 2011. Australian Bureau of Statistics. http://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/1270.0.55.005July%202011?OpenDocument. 2014.
- 31. Bursac Z, Gauss C, Williams D, Hosmer D. Purposeful selection of variables in logistic regression. Source Code Biol Med. 2008;3:17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Pearce N. Analysis of matched case‐control studies. BMJ. 2016;352:i969. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Battelino T, Phillip M, Bratina N, Nimri R, Oskarsson P, Bolinder J. Effect of continuous glucose monitoring on hypoglycemia in type 1 diabetes. Diabetes Care. 2011;34(4):795‐800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Poolsup N, Suksomboon N, Kyaw AM. Systematic review and meta‐analysis of the effectiveness of continuous glucose monitoring (CGM) on glucose control in diabetes. Diabetol Metab Syndr. 2013;5(1):1‐14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Ho MS, Weller NJ, Ives FJ, et al. Prevalence of structural central nervous system abnormalities in early‐onset type 1 diabetes mellitus. J Pediatr. 2008;153(3):385‐390. [DOI] [PubMed] [Google Scholar]
- 36. Patiño‐Fernández AM, Delamater AM, Applegate EB, et al. Neurocognitive functioning in preschool‐age children with type 1 diabetes mellitus. Pediatr Diabetes. 2010;11(6):424‐430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Fletcher JM, Richards MR. Diabetes's ‘health shock' to schooling and earnings: increased dropout rates and lower wages and employment in young adults. Health Aff. 2012;31(1):27‐34. [DOI] [PubMed] [Google Scholar]
- 38. Hochhauser CJ, Rapaport R, Shemesh E, Schmeidler J, Chemtob CM. Age at diagnosis, gender, and metabolic control in children with type 1 diabetes. Pediatr Diabetes. 2008;9(4 pt 1):303‐307. [DOI] [PubMed] [Google Scholar]
- 39. Forsander G, Bøgelund M, Haas J, Samuelsson U. Adolescent life with diabetes—gender matters for level of distress. Experiences from the national TODS study. Pediatr Diabetes. 2017;18(7):651‐659. [DOI] [PubMed] [Google Scholar]
- 40. McCarvill R, Weaver K. Primary care of female adolescents with type 1 diabetes mellitus and disordered eating. J Adv Nurs. 2014;70(9):2005‐2018. [DOI] [PubMed] [Google Scholar]
- 41. Wisting L, Bang L, Skrivarhaug T, Dahl‐Jørgensen K, Rø Ø. Adolescents with type 1 diabetes–the impact of gender, age, and health‐related functioning on eating disorder psychopathology. PLoS One. 2015;10(11):e0141386. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Wagner G, Karwautz A. Eating disorders in adolescents with type 1 diabetes mellitus. Curr Opin Psychiatry. 2020;33(6):602‐610. [DOI] [PubMed] [Google Scholar]
- 43. Delamater AM, de Wit M, McDarby V, et al. ISPAD clinical practice consensus guidelines 2018: psychological care of children and adolescents with type 1 diabetes. Pediatr Diabetes. 2018;19:237‐249. [DOI] [PubMed] [Google Scholar]
- 44. Mikkonen J, Moustgaard H, Remes H, Martikainen P. The population impact of childhood health conditions on dropout from upper‐secondary education. J Pediatr. 2018;196(283–290):e284. [DOI] [PubMed] [Google Scholar]
- 45. Diabetes Australia . New Diabeetes in Schools Program Launched. Diabetes Australia. https://diabetesnsw.com.au/news/new-diabetes-in-schools-program-launched/. 2021.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Appendix S1: Supporting information
Data Availability Statement
The data that support the findings of this study are available from the NSW Health Department, NSW Department of Education, and NSW Education Standards Authority. Restrictions apply to the availability of these data, which were used under licence for the current study, so are not publicly available.