School performance in Danish children exposed to maternal type 1 diabetes in utero: A nationwide retrospective cohort study.

BACKGROUND: Conflicting results have been reported concerning possible adverse effects on the cognitive function of offspring of mothers with type 1 diabetes (O-mT1D). Previous studies have included offspring of parents from the background population (O-BP), but not offspring of fathers with type 1 diabetes (O-fT1D) as the unexposed reference group. METHODS AND
FINDINGS: This is a population-based retrospective cohort study from 2010 to 2016. Nationally standardized school test scores (range, 1 to 100) were obtained for public school grades 2, 3, 4, 6, and 8 in O-mT1D and compared with those in O-fT1D and O-BP. Of the 622,073 included children, 2,144 were O-mT1D, and 3,474 were O-fT1D. Multiple linear regression models were used to compare outcomes, including the covariates offspring with type 1 diabetes, parity, number of siblings, offspring sex, smoking during pregnancy, parental age, and socioeconomic factors. Mean test scores were 54.2 (standard deviation, SD 24.8) in O-mT1D, 54.4 (SD 24.8) in O-fT1D, and 56.4 (SD 24.7) in O-BP. In adjusted analyses, the mean differences in test scores were -1.59 (95% CI -2.48 to -0.71, p < 0.001) between O-mT1D and O-BP and -0.78 (95% CI -1.48 to -0.08, p = 0.03) between O-fT1D and O-BP. No significant difference in the adjusted mean test scores was found between O-mT1D and O-fT1D (p = 0.16). The study's limitation was no access to measures of glycemic control during pregnancy.
CONCLUSIONS: O-mT1D achieved lower test scores than O-BP but similar test scores compared with O-fT1D. Glycemic control during pregnancy is essential to prevent various adverse pregnancy outcomes in women with type 1 diabetes. However, the present study reduces previous concerns regarding adverse effects of in utero hyperglycemia on offspring cognitive function.

Introduction

The influence of maternal diabetes during pregnancy on offspring cognition has been widely explored [1-10]. Unlike insulin, glucose crosses the placenta, and, therefore, maternal hyperglycemia leads to intrauterine hyperglycemia and subsequently fetal hyperinsulinemia, which increases fetal growth [11]. In addition, maternal hyperglycemia serves as a nonspecific teratogenic factor. In animal studies, impaired dendritic development is described in offspring born to rats with streptozotocin-induced hyperglycemia in pregnancy—possibly through abnormal insulin/insulin-like growth factor 1 (IGF1) signaling in the fetal brain [10]. Other studies have found an up-regulation of IGF-1 receptors in the cerebellum and structural changes in the hypothalamus in the offspring of diabetic rats [12,13]. However, few human studies on cognitive function in offspring exposed to maternal diabetes during pregnancy have distinguished between different types of maternal diabetes (e.g., gestational diabetes and type 1 and type 2 diabetes). These studies were based on small sample sizes and a wide variety of cognitive tests, and they reached varying conclusions [1-6]. Furthermore, to our knowledge, previous studies have included offspring of parents from the background population (O-BP), but not offspring of fathers with type 1 diabetes (O-fT1D) as the unexposed reference group.

All Danish children in public schools are tested in reading and math using nationally standardized tests. This provides a unique opportunity to assess the potential association between in utero exposure to maternal type 1 diabetes and school performance in offspring of mothers with type 1 diabetes (O-mT1D). Therefore, the primary objective of this study was to compare test scores in O-mT1D with test scores in O-fT1D and O-BP. We hypothesized that O-mT1D would achieve lower test scores compared with O-fT1D and O-BP due to intrauterine exposure of hyperglycemia.

Methods

This population-based retrospective cohort study included all singletons attending public schools in Denmark from January 1, 2010 to December 31, 2016 (S1 Fig). The analysis plan was drafted prospectively in June 2020 (S1 Text). This study is reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (S1 Checklist).

Offspring were initially identified through the national Medical Birth Register to ensure inclusion of offspring born in Denmark in whom there was valid information on maternal and paternal diabetes diagnosed before pregnancy. Subsequently, maternal, obstetric, and perinatal outcomes from the Medical Birth Register were cross-linked to in- and outpatient diagnoses from the National Patient Register and cross-linked to nationally standardized school test scores and socioeconomic factors from different administrative registers from Statistics Denmark. This cross-linking of registers is possible due to the unique personal identification number allocated to all residents in Denmark either at birth or on immigration. All data were deidentified prior to access.

Exposure

Type 1 diabetes was defined according to the International Classification of Diseases (ICD) 8th revision (ICD-8) 249 code and the 10th revision (ICD-10) E10 code assigned before childbirth.

Offspring were grouped by parental diabetes status: O-mT1D, O-fT1D, or O-BP. Offspring in whom both the mother and the father were diagnosed with type 1 diabetes were excluded (S1 Fig).

Outcomes

During the follow-up period, nationally standardized school tests were mandatory in Danish public schools, but not private schools. Consequently, outcomes based on test scores were restricted to offspring attending public schools.

The children were tested in math in grades 3 and 6 and in reading in grades 2, 4, 6, and 8. The tests evaluate the students’ abilities in math and reading within different profile areas. In math, the profile areas include algebra, geometry, statistics, and probability, while the reading profile areas include language and text comprehension and decoding. The tests are computer based, and the students are presented with questions of varying difficulty to assess their individual level of skills. At the end of the test, a score is automatically calculated, ranging from 1 to 100 [14,15]. The scores from these tests are highly correlated with the grade point average on compulsory school exit exams, and the tests have been used for research in several previous studies [14,16-18].

The primary outcome was the test scores in both math and reading. Secondary outcomes were test scores analyzed separately in math or reading and test scores in math or reading stratified by grade. Third, the likelihood of missing a test despite attending a public school during follow-up was assessed. Finally, the likelihood of attending a private school was assessed to check if the studied population was representative.

Covariates

Covariates were included to compare outcomes across grades (categorical: 2, 3, 4, 6, or 8), topic (math/reading), and year of test (categorical: 2010, 2011, 2012, 2013, 2014, 2015, or 2016).

Basic covariates were offspring sex (male/female), parity (categorical: 1, 2, 3, 4, or ≥5), number of siblings (categorical: 0, 1, 2, or ≥3), offspring with type 1 diabetes (yes/no) defined as ICD-10 E10 code assigned before a test, and maternal smoking during pregnancy (yes/no).

Parental socioeconomic factors were parental highest educational level (categorical: primary school, high school, vocational, short higher education, middle secondary education, master’s degree, or higher), income (categorical: percentile rank Q1, Q2, Q3, or Q4), immigrant or descendent status (yes/no), and age (continuous). Furthermore, parents living together (yes/no) was included as a socioeconomic variable. All socioeconomic factors in the offspring were recorded at age 5 (before starting school).

Obstetric and perinatal covariates acting as potential mediators were hypertensive disorders of pregnancy (yes/no) defined as the ICD-10 codes O10 to 16 assigned during the 280 days before date of birth, cesarean section (yes/no), low Apgar score at 5 minutes (<7 or ≥7), gestational age (categorical: <32 + 0, 32 + 0 to 36 + 6, 37 + 0 to 3 9 +6, or ≥40 + 0 weeks), and birth weight according to expected sex-specific birth weight for the given gestational age, calculated according to Marsal and colleagues (categorical: small, average, or large for gestational age) [19]. Small, average, and large for gestational age were defined as the <10%, 10% to 90%, and >90% percentile, respectively.

Statistical analyses

Descriptive statistics regarding maternal, paternal, and offspring characteristics were presented as proportions or means with standard deviations (SDs; Table 1).

Table 1

Descriptive statistics of study sample.

	%
	O-mT1D		O-fT1D		O-BP		F-test for no differences across groups
Characteristics	(n = 2,144)	% missing obs.	(n = 3,474)	% missing obs.	(n = 616,455)	% missing obs.
Child
Male	52.7	0%	50.2	0%	51.2	0%	0.21
Female	47.3	0%	49.8	0%	48.7	0%	0.20
Firstborn	44.9	0%	42.4	0%	43.2	0%	0.18
Number of siblings, mean (SD)	1.2 (0.78)	0%	1.4 (0.82)	0%	1.5 (0.8)	0%	<0.001
Parents living together	64.5	0%	61.5	0%	67.3	0%	<0.001
Mother						1%	<0.001
Age, mean (SD)	35.5 (4.67)	0%	35.2 (4.83)	0%	35.0 (4.75)	0%	<0.001
Master’s degree or higher	6.0	1%	7.0	1%	9.1	1%	<0.001
Income percentile rank, mean (SD)	0.60 (0.23)	1%	0.63 (0.22)	1%	0.6 (0.22)	1%	<0.001
Immigrant or descendanta	6.4	2%	10.8	3%	11.3	2%	<0.001
Father						0%	<0.001
Age, mean (SD)	38.0 (5.67)	2%	38.3 (6.01)	3%	37.6 (5.62)	2%	<0.001
Master’s degree or higher	8.2	2%	7.7	3%	10.8	2%	<0.001
Income percentile rank, mean (SD)	0.74 (0.24)	0%	0.7 (0.26)	0%	0.7 (0.24)	0%	<0.001
Immigrant or descendanta	7.8	2%	9.5	3%	11.5	2%	<0.001
Clinical
Offspring with type 1 diabetes	2.1	0%	4.0	0%	0.4	0%	0.03
Maternal smoking during pregnancy	30.5	16%	26.9	12%	28.2	14%	0.03
Hypertensive disorders of pregnancy	16.8	0%	4.5	0%	4.2	0%	<0.001
Cesarean section	59.6	0%	16.5	6%	16.1	0%	<0.001
Low Apgar score at 5 minutes (<7)	1.2	2%	0.6	1%	0.6	1%	0.006
Gestational age <37+0	31.5	10%	4.2	9%	4.4	11%	<0.001
Large for gestational age	59.2	10%	13.6	9%	13.5	11%	<0.001
Test
Math and reading test score, mean (SD) (n = 1,704,447)	54.2 (24.8)		54.4 (24.8)		56.4 (24.7)		<0.001
Math test score, mean (SD) (n = 574,160 test scores)	53.2 (25.1)		53.9 (25.2)		56.5 (25.1)		<0.001
Reading test score, mean (SD) (n = 1,130,287 test scores)	54.7 (24.7)		54.7 (24.5)		56.4 (24.5)		<0.001
Missing test scores	5.5		4.4		4.1		<0.001

Notes: Data reported as proportions or means with SD in parentheses. p-Value from F-test for no differences in means across the groups O-mT1D (maternal diabetes), O-fT1D (paternal diabetes), and O-BP (background population).

aImmigrants are Danish residents who were not born in Denmark and had patents who were not born in Denmark. Descendants are Danish residents who were born in Denmark and had parents who were not born in Denmark.

O-BP, offspring of parents from the background population; O-fT1D, offspring of fathers with type 1 diabetes; O-mT1D, offspring of mothers with type 1 diabetes; SD, standard deviation.

The association between O-mT1D, O-fT1D, and O-BP and test scores was estimated in multiple regression models with O-BP as reference. As a child could take multiple tests, hence, contributing with multiple test scores (math or reading, but also at different grade levels), test scores were analyzed in a pooled analysis, with each test score as a unique observation and cluster-robust standard errors were used to adjust for within-individual correlations. To accommodate clustering of data at the school level, a school fixed effect regression model was estimated. To assess whether test scores in O-mT1D were significantly different from those in O-fT1D, we performed a Wald test for the equality of coefficients in the 2 groups in the regression model. The mean difference in test scores was estimates in a linear regression model. Selected mean differences in test scores were also presented as Cohen’s d to assess the magnitude of the differences. The risk of missing a test score and the odds ratio (OR) of attending a private school were estimated as an OR in logistic regression models. Covariates were chosen based on current knowledge about factors associated with intelligence and school performance [14,20].

Crude analyses of the pooled test scores in math and reading included only adjustment for grade, topic (reading/math), and year effects (grade–topic–year–specific effects) (Model 1). In Model 2, the multiple regression was further adjusted for the following basic covariates: offspring sex, parity, number of siblings, offspring with type 1 diabetes, and maternal smoking during pregnancy. Model 3 was further adjusted for socioeconomic factors including parental age. Finally, to evaluate a potential mediating role of obstetric and perinatal complications, these covariates were added to Model 3 one by one in separate analyses (Models S1–S5) and in a combined analysis (Model S6).

In multiple regression analyses that included the covariates in Model 3, we estimated the OR of attending a private school (logistic regression analyses), the mean differences in test scores stratified separately by math or reading and test scores in math or reading stratified by grade (linear regression analyses), and the OR of missing a test score (logistic regression analyses). To investigate the importance of missing data on test scores conditional on being enrolled in a public school, 2 different sensitivity analyses were performed on the main outcome: (1) all children with missing test scores were assigned the lowest possible score, and the main analysis was reestimated to see whether conclusions changed; and (2) inverse probability weighting was undertaken using the predictions from the logistic regression of missing a test score to reweight for nonresponse in the main analysis [21].

Missing values on independent variables in the regression models were addressed by adding dummy variables for missing values (Table 1).

A p-value <0.05 was considered statistically significant. Because of the potential for type I errors due to multiple comparisons, results for analyses of secondary outcomes should be interpreted as exploratory. All statistical analyses were performed using Stata 15 (Stata).

Ethical approval

The study was approved by the Danish Data Protection Agency. This approval constitutes the necessary legal requirement, and informed consent is not required.

Results

Of 731,455 children initially identified via Statistics Denmark, 622,073 attended public schools (85%), and 1,704,447 test scores were obtained. Among children in public schools, 2,144 O-mT1D, 3,474 O-fT1D, and 616,455 O-BP were eligible for the study (S1 Fig). The adjusted odds ratios (aORs) for attending a private school were 1.02 (95% CI 0.89 to 1.17, p = 0.85) for O-mT1D versus O-BP and 0.94 (95% CI 0.84 to 1.05, p = 0.17) for O-fT1D versus O-BP. There were no significant different in the aOR of attending a private school for O-mT1D versus O-fT1D (p = 0.31). Among the offspring attending 1 or more grades, 96% O-mT1D, 97% O-fT1D, and 97% O-BP were registered with at least 1 test score. More O-mT1D had at some point missed a test score than O-BP (aOR 1.30 [95% CI 1.16 to 1.47, p < 0.001]), but no difference was observed for O-fT1D versus O-BP (aOR 1.02 [95% CI 0.92 to 1.13, p = 0.80]). O-mT1D were significantly more likely to have a missing test score compared with O-fT1d (p = 0.003).

In total, 2,144 offspring of 1,597 unique mothers and 3,474 offspring of 2,419 unique fathers with type 1 diabetes and 616,455 offspring of 401,406 unique mothers and 402,785 unique fathers from the background population had offspring included in the cohort. Offspring of both O-mT1D and O-fT1D were excluded (n = 30).

Characteristics of the population

The mean age of the study population was 10.1 years (SD 2.39) (at first test). Of the study population, 98.3% were born between 1996 and 2007. Comparisons of background characteristics showed that O-mT1D had fewer siblings, their mothers were older and less likely to have a master’s degree or higher educational level, and their parents were less likely to be immigrants or descendants than O-BP and O-fT1D. Obstetric and perinatal complications were more often present in pregnancies in O-mT1D.

Offspring living with both parents accounted for 64.5% in O-mT1D, 61.5% in O-fT1D, and 67.3% in O-BP. Mean paternal age was higher in O-mT1D, and their fathers were less likely to have a master’s degree or higher educational level than O-BP. Mean paternal age was lower in O-mT1D, and their fathers were more likely to have a master’s degree or higher educational level than O-fT1D. Diabetes was diagnosed in 2.1% of O-mT1D, 4.0% of O-fT1D, and 0.4% of O-BP (Table 1).

Test scores

The mean of test scores were 54.2 (SD 24.8) in O-mT1D, 54.4 (SD 24.8) in O-fT1D, and 56.4 (SD 24.7) in O-BP, corresponding to a mean difference of −1.90 [95% CI −2.87 to −0.94, p < 0.001] (Cohen’s d = –0.08) for O-mT1D versus O-BP and −1.57 [95% CI −2.33 to −0.81, p < 0.001] (Cohen’s d = −0.06) for O-fT1D versus O-BP after adjusting for the grade–topic–year–specific effects (Table 2, Model 1). After adjusting for basic and socioeconomic covariates, O-mT1D and O-fT1D had lower mean test scores than O-BP (mean differences −1.59 [95% CI −2.48 to −0.71, p < 0.001] (Cohen’s d = −0.06) and −0.78 [95% CI −1.48 to −0.08, p = 0.03] (Cohen’s d = −0.03), respectively) (Table 2, Model 3). The mean difference in test score for O-mT1D versus O-fT1D was insignificant in both Model 1 and 3 (p = 0.60 and p = 0.16, respectively) (Table 2). Analyses that included further adjustments for all potentially mediating obstetric and perinatal covariates did not change the results (S1 Table, Model S1–S6).

Table 2

Multiple linear regression analyses comparing test scores in offspring of mothers (O-mT1D) and fathers (O-fT1D) with type 1 diabetes compared with O-BP, mean test score difference (95% CI).

	Model (n = 1,704,447 test scores)
Explanatory variables	Model 1	Model 2	Model 3
	Mean difference (95% CI), p-value	Mean difference (95% CI), p-value	Mean difference (95% CI), p-value
Diabetes status
O-mT1D	−1.90 (−2.87 to −0.94), p < 0.001	−1.96 (−2.91 to −1.01), p < 0.001	−1.59 (−2.48 to −0.71), p < 0.001
O-fT1D	−1.57 (−2.33 to −0.81), p < 0.001	−1.49 (−2.23 to −0.74), p < 0.001	−0.78 (−1.48 to −0.08), p = 0.03
O-BP	(ref.)	(ref.)	(ref.)
Basic covariates
Male		−2.88 (−2.99 to −2.77), p < 0.001	−2.98 (−3.08 to −2.87), p < 0.001
Parity
1		(ref.)	(ref.)
2		−2.67 (−2.80 to −2.54), p < 0.001	−2.96 (−3.09 to −2.84), p < 0.001
3		−3.96 (−4.16 to −3.76), p < 0.001	−4.65 (−4.85 to −4.45), p < 0.001
4		−4.20 (−4.59 to −3.80), p < 0.001	−5.46 (−5.85 to −5.08), p < 0.001
≥5		−9.02 (−9.58 to −8.46), p < 0.001	−8.00 (−8.55 to −7.44), p < 0.001
Number of siblings
0		(ref.)	(ref.)
1		1.95 (1.73 to 2.18), p < 0.001	0.51 (0.29 to 0.73), p < 0.001
2		1.85 (1.61 to 2.10), p < 0.001	0.68 (0.44 to 0.92), p < 0.001
≥3		−2.19 (−2.51 to −1.88), p < 0.001	−0.22 (−0.52 to 0.09), p = 0.17
Offspring with type 1 diabetes		−0.09 (−0.97 to 0.80), p = 0.84	−0.16 (−0.98 to 0.67), p = 0.71
Maternal smoking during pregnancy		−6.29 (−6.44 to −6.14), p < 0.001	−2.08 (−2.23 to −1.93), p < 0.001
Socioeconomic covariates
Mother’s highest educational level
Primary school			(ref.)
High school			7.36 (7.12 to 7.60), p < 0.001
Vocational			3.27 (3.09 to 3.44), p < 0.001
Short higher education			7.1 (6.82 to 7.38), p < 0.001
Middle secondary education			8.75 (8.55 to 8.94), p < 0.001
≥Master’s degree			12.59 (12.34 to 12.84), p < 0.001
Mother’s income, by quartile in income distribution
Q1			(ref.)
Q2			0.13 (−0.17 to 0.43), p = 0.40
Q3			0.84 (0.67 to 1.02), p < 0.001
Q4			2.27 (2.07 to 2.46), p < 0.001
Mother is immigrant or descendant*			−3.42 (−3.67 to −3.18), p < 0.001
Father’s highest educational level
Primary school			(ref.)
High school			7.87 (7.61 to 8.12), p < 0.001
Vocational			2.76 (2.60 to 2.91), p < 0.001
Short higher education			6.29 (6.05 to 6.54), p < 0.001
Middle secondary education			8.9 (8.69 to 9.10), p < 0.001
≥Master’s degree			11.35 (11.12 to 11.58), p < 0.001
Father’s income, by quartile in income distribution
Q1			(ref.)
Q2			0.12 (−0.22 to 0.46), p = 0.49
Q3			−0.03 (−0.23 to 0.17), p = 0.74
Q4			1.18 (1.00 to 1.36), p < 0.001
Father is immigrant or descendant*			−2.96 (−3.21 to −2.72), p < 0.001
Parents living together			1.73 (1.61 to 1.85), p < 0.001
Test: (p-value)
O-mT1D = O-fT1D	0.60	0.44	0.16

Notes: All regression models are adjusted for grade-, topic-, and year-specific fixed effects.

Model 1 is with no further adjustment.

Model 2 is adjusted for offspring sex, parity, number of siblings, offspring with type 1 diabetes, and maternal smoking during pregnancy

Model 3 is adjusted for offspring sex, parity, number of siblings, offspring with type 1 diabetes, and maternal smoking during pregnancy, parental highest educational level, income, immigrant or descendant status, age (coefficients not reported), and parents living together.

All covariates are dichotomous (0/1).

p-Value from Wald test (F-test) of equality of the regression coefficients to O-mT1D (maternal diabetes) and O-fT1D (paternal diabetes) is reported.

*Immigrants are Danish residents not born in Denmark, with neither of their parents born in Denmark. Descendants are Danish residents born in Denmark, with neither of their parents born in Denmark.

O-BP, offspring of parents from the background population; O-fT1D, offspring of fathers with type 1 diabetes; O-mT1D, offspring of mothers with type 1 diabetes.

In the first sensitivity analysis assessing the potential significance of the increased probability of missing a test score among O-mT1D and O-fT1D, the main analysis was estimated (Model 3) after recoding all missing test scores to 1 (assuming those missing the test scores would have performed worst of all). The mean adjusted test scores differences were −1.73 [95% CI −2.64 to −0.82, p < 0.001] for O-mT1D versus O-BP and −0.82 [95% CI −1.54 to −0.10, p = 0.03] for O-fT1D versus O-BP. The difference between O-mT1D and O-fT1D was insignificant (p = 0.12).

In the second sensitivity analysis using inverse probability weighting for nonresponse, the mean adjusted (Model 3) test score differences were −1.60 [95% CI −2.49 to −0.71, p < 0.001] for O-mT1D versus O-BP and −0.71 [95% CI −1.42 to −0.00, p = 0.049] for O-fT1D versus O-BP. The difference between O-mT1D and O-fT1D was not statistically significant (p = 0.13).

Test scores in math or reading

In adjusted analyses that included covariates in Model 3, the overall mean differences in math test scores were lower for both O-mT1D and O-fT1D compared with O-BP (mean difference −2.61 [95% CI −3.71 to −1.50, p < 0.001] and −1.19 [95% CI −2.08 to −0.31, p = 0.01], respectively). Similarly, analyses of math test scores stratified by grades showed that both O-mT1D and O-fT1D had lower mean test scores in third grade compared with O-BP (mean difference −2.38 [95% CI −3.82 to −0.94, p = 0.001 and −1.17 [95% CI −2.27 to −0.07, p = 0.04], respectively). However, in sixth grade, only O-mT1D were found to have a lower mean test score than O-BP (mean difference −2.77 [95% CI −4.17 to −1.40, p < 0.001]) (Fig 1).

Fig 1

Multiple linear regression analyses comparing test scores in math or reading in offspring of mothers (O-mT1D) and fathers (O-fT1D) with type 1 diabetes compared with O-BP, mean test score difference (95% CI).

Notes: Differences are adjusted for grade-, topic-, and year-specific fixed effects, offspring sex, parity, number of siblings, offspring with type 1 diabetes, maternal smoking during pregnancy, parental highest educational level, income, immigrant or descendant status, age, and parents living together (corresponding to Model 3). p-Value from Wald test (F-test) of equality of the regression coefficients to O-mT1D and O-fT1D is reported. O-BP, offspring of parents from the background population; O-fT1D, offspring of fathers with type 1 diabetes; O-mT1D, offspring of mothers with type 1 diabetes.

O-mT1D were found to have a lower overall reading test score (mean difference −1.10 [95% CI −2.05 to −0.14, p = 0.02]) and lower reading test scores in fourth grade than O-BP (mean difference −2.19 [95% CI −3.58 to −0.80, p = 0.002]) (Fig 1).

Neither math test scores nor reading test scores differed for O-mT1D versus O-fT1D except for the overall math test score, which were lower for O-mT1D (p = 0.0496) (Fig 1).

Discussion

Principal findings

In this retrospective register-based cohort study, we found that both O-mT1D and O-fT1D had lower test scores than O-BP. However, there were no differences in test scores for O-mT1D and O-fT1D. O-mT1D had a slightly increased risk of missing a test score compared with O-fT1D and a lower overall math test score.

Comparisons to other studies

Several previous studies have examined the association between exposure to maternal diabetes during pregnancy and cognitive ability in offspring [1-10]. However, to our knowledge, only 4 existing studies have distinguished between type 1 diabetes, type 2 diabetes, and gestational diabetes [1-4], 3 of which are small clinical follow-up studies [1,2,4]. Similar, we are not aware of previous studies comparing O-mT1D with O-fT1D, hence taking advantage of the ability to adjust for potential genetic factors. The 4 previous studies of O-mT1D addressed cognitive function in different ways; some estimated intelligence quotients using standardized tests [1,2,4], while one study used mandatory school exit exams as a proxy for cognitive function [3]. One study found O-mT1D to have impaired cognitive function compared with O-BP [4], while 3 studies found similar cognitive function [1-3]. Two of the studies were included in a recent systematic review and meta-analysis, concluding that exposure to maternal type 1 diabetes during pregnancy is associated with a lower intelligence quotient in the offspring (pooled weighted mean difference −4.62 (95% CI −6.75 to −2.50) [1,4,5]. Among these 4 studies, the study by Knorr and colleagues had the design that was most similar to our study. They assessed mandatory school exit exam grades, which are shown to be highly correlated with the test scores used in our study [14,16]. Knorr and colleagues included O-mT1D from a prospectively sampled clinical cohort (n = 707) born in Denmark from 1992 to 1999 and found that O-mT1D had an insignificant, slightly lower adjusted mean grade point average than O-BP (adjusted mean difference –0.07 [95% CI –0.23 to 0.09]) [3]. Our study included test scores from multiple grades and a cohort of O-mT1D that was 3-fold larger than in the study by Knorr and colleagues, which possibly explains why our results reached statistical significance contrary to the study by Knorr and colleagues Moreover, the outcome of the current study was a low stake, computerized, and nonteacher-assessed test. The differences in test scores between O-mT1D and O-BP in our study, represented as Cohen’s d, were not larger than 0.1 (numerically), and a Cohen’s d less than 0.20 is generally considered a small effect size [22].

Strengths and limitations

The major strength of this study is the population design. The register-based study design enabled us to include all offspring attending public schools over a 7-year period and allowed us to correct for covariates shown to be strongly associated with intelligence and school performance [14,20]. Finally, other studies evaluating the potential adverse impact of intrauterine hyperglycemia on offspring cognitive ability in O-mT1D have used O-BP as a reference group. However, to our knowledge, no previous studies have used O-fT1D as an unexposed reference group. We included O-fT1D as an unexposed reference group, thereby adjusting for any potentially shared genes for diabetes and impaired school performance. Also, having a parent suffering from a serious chronic disease like diabetes may in itself cause familial stress and be detrimental to school performance [23,24].

The present study has several limitations. First, the data did not include information on HbA1c during pregnancy because this information is not available in the Danish national registries. However, other Danish clinical studies on pregnant women with type 1 diabetes indicate that HbA1c is measured at least 5 times during pregnancy, and from 1992 to 1999, the HbA1c treatment target was not reached at any time during pregnancy according to the recommended HbA1c levels presently used in Denmark [3,25,26]. Also, during the years in which the included adolescents in our study were born, very few pregnant women with type 1 diabetes were using an insulin pump, and less than 50% were treated with human insulin [27,28]. Whether the glycemic level differed between mothers and fathers with type 1 diabetes during the time period covered by the present study can unfortunately not be explored using the information available in the Danish national registers.

Second, O-mT1D were more likely to have a missing test score than O-fT1D and O-BP. Nevertheless, sensitivity analyses on missing test scores did not change the conclusion.

Third, the current study only included children attending public schools because school tests are not mandatory in private schools. However, adjusted analyses revealed no significant difference between children attending public and those attending or private schools among the 3 groups.

Fourth, residual confounding cannot be excluded in studies based on a cohort design. Ethnicity or race was not considered in our study as this information is not available in Danish registers. However, immigrant or descendent status was included in our multivariable model, which to some extent counts as a proxy of ethnicity or race. Furthermore, obstetrics and perinatal complications are considered mediators on the casual pathway of maternal diabetes status and school performance; hence, adjusted analyses including these covariates are questionable because unknown confounding by factors affecting both the obstetrics and perinatal complications and school performance may introduce bias [29].

Finally, extrapolating our results to other populations should be done with caution because Denmark has a high-quality healthcare system that is free of charge, and all pregnant women with type 1 diabetes are referred for treatment at 4 highly specialized obstetric departments. Therefore, the difference in school performance may be even larger in countries where access to healthcare is seriously influenced by socioeconomic factors. Also, glycemic control in pregnant women with type 1 diabetes has improved in Denmark since the included cohort was born [30].

Parents or offspring with type 2 diabetes or mothers with gestational diabetes were not excluded from the analyses. However, we expect the potential effect of parents with either type 2 diabetes or gestational diabetes in O-BP or O-fT1D to be deluded by the large number of healthy parents in these groups, mainly because very few parents will have developed type 2 diabetes before birth due to their young age. Conceptually, we also think that the relevant comparison group is the “average child” in the background population, and by excluding children where the parents had other diagnoses than type 1 diabetes, the comparison group would be made healthier.

Conclusions and further implications

Among Danish schoolchildren attending public school, O-mT1D achieved lower mean test scores than O-BP after adjusting for parental educational level and other confounding factors. However, mean test scores were indistinguishable from those in O-fT1D, indicating that the lower school performance in O-mT1D is associated with having a parent with a chronic disease rather than associated with intrauterine hyperglycemia. These results are reassuring for women with type 1 diabetes, as one of their main concerns is whether dysregulation of diabetes during their pregnancy may cause impaired cognitive development in their coming children. This study is, to our knowledge, the first to present evidence of an alternative explanation for the previously observed adverse effect of maternal type 1 diabetes during pregnancy on offspring cognitive development.

No direct information on HbA1c levels during pregnancy was available in our study; hence, we were not able to determine the association between HbA1c levels during pregnancy and school performance in O-mT1D, but this association is a potential subject for future studies in the field.

Flow of study design.

(TIF)

Click here for additional data file.

Prospective analysis plan: School performance in Danish children exposed to maternal type 1 diabetes in utero: A nationwide retrospective cohort study.

(DOCX)

Click here for additional data file.

STROBE guideline.

STROBE, Strengthening the Reporting of Observational Studies in Epidemiology.

(DOCX)

Click here for additional data file.

Multiple linear regression analyses comparing test scores in offspring of mothers (O-mT1D) and fathers (O-fT1D) with type 1 diabetes compared with offspring in the background population (O-BP), mean test score differencSSe (95% CI)—Model 3 with sequential adjustment for potential mediators.

(DOCX)

Click here for additional data file.

3 Nov 2021

Dear Dr Spangmose,

Thank you for submitting your manuscript entitled "School performance in Danish children exposed to maternal type 1 diabetes in utero: a nationwide cohort study" for consideration by PLOS Medicine.

Your manuscript has now been evaluated by the PLOS Medicine editorial staff and I am writing to let you know that we would like to send your submission out for external peer review.

However, before we can send your manuscript to reviewers, we need you to complete your submission by providing the metadata that is required for full assessment. To this end, please login to Editorial Manager where you will find the paper in the 'Submissions Needing Revisions' folder on your homepage. Please click 'Revise Submission' from the Action Links and complete all additional questions in the submission questionnaire.

Please re-submit your manuscript within two working days, i.e. by .

Login to Editorial Manager here: https://www.editorialmanager.com/pmedicine

Once your full submission is complete, your paper will undergo a series of checks in preparation for peer review. Once your manuscript has passed all checks it will be sent out for review.

Feel free to email us at plosmedicine@plos.org if you have any queries relating to your submission.

Kind regards,

Beryne Odeny

Senior Editor

PLOS Medicine

20 Jan 2022

Dear Dr. Spangmose,

Thank you very much for submitting your manuscript "School performance in Danish children exposed to maternal type 1 diabetes in utero: a nationwide cohort study" (PMEDICINE-D-21-04566R1) for consideration at PLOS Medicine.

Your paper was evaluated by a senior editor and discussed among all the editors here. It was also discussed with an academic editor with relevant expertise, and sent to independent reviewers, including a statistical reviewer. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

[LINK]

In light of these reviews, I am afraid that we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to consider a revised version that addresses the reviewers' and editors' comments. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.

In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments, the changes you have made in the manuscript, and include either an excerpt of the revised text or the location (eg: page and line number) where each change can be found. Please submit a clean version of the paper as the main article file; a version with changes marked should be uploaded as a marked up manuscript.

In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements: http://journals.plos.org/plosmedicine/s/figures. While revising your submission, please upload your figure files to the PACE digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at PLOSMedicine@plos.org.

We expect to receive your revised manuscript by Feb 10 2022 11:59PM. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here: http://journals.plos.org/plosmedicine/s/competing-interests.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

We look forward to receiving your revised manuscript.

Sincerely,

Beryne Odeny,

PLOS Medicine

plosmedicine.org

-----------------------------------------------------------

Requests from the editors:

1) Please revise your title according to PLOS Medicine's style. Please consider “School performance in Danish children exposed to maternal type 1 diabetes in utero: a nationwide retrospective cohort study”

2) The Data Availability Statement (DAS) requires revision. For each data source used in your study, if the data are not freely available, please describe briefly the ethical, legal, or contractual restriction that prevents you from sharing it. Please also include an appropriate contact (web or email address) for inquiries (this cannot be a study author).

3) Please move the ethical approval statement from the end of the main text to the Methods section.

4) Abstract:

a) Please combine the Methods and Findings sections into one section, “Methods and findings”

b) Please ensure that all numbers presented in the abstract are present and identical to numbers presented in the main manuscript text.

c) Please quantify the main results (with 95% CIs and p values).

d) In the last sentence of the Abstract Methods and Findings section, please describe the main limitation(s) of the study's methodology.

e) Conclusions: Please address the study implications without overreaching what can be concluded from the data and emphasize what is new considering the current evidence.

f) Please remove the funding statement from the end of the abstract

5) At this stage, we ask that you include a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract. Please see our author guidelines for more information: https://journals.plos.org/plosmedicine/s/revising-your-manuscript#loc-author-summary

6) Please avoid assertions of primacy ("… to provide among/ some of the first evidence...."). Instead, state “to our knowledge” or similar.

7) Please conclude the Introduction with a clear description of the study hypothesis.

8) Did your study have a prospective protocol or analysis plan? Please state this (either way) early in the Methods section.

a) If a prospective analysis plan (from your funding proposal, IRB or other ethics committee submission, study protocol, or other planning document written before analyzing the data) was used in designing the study, please include the relevant prospectively written document with your revised manuscript as a Supporting Information file to be published alongside your study, and cite it in the Methods section. A legend for this file should be included at the end of your manuscript.

b) If no such document exists, please make sure that the Methods section transparently describes when analyses were planned, and when/why any data-driven changes to analyses took place.

c) In either case, changes in the analysis-- including those made in response to peer review comments-- should be identified as such in the Methods section of the paper, with rationale.

9) Please state whether the health data were de-identified (anonymized) prior to access.

10) Please ensure that the study is reported according to the STROBE guideline, and include the completed STROBE checklist as Supporting Information. Please add the following statement, or similar, to the Methods: "This study is reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (S1 Checklist)."

The STROBE guideline can be found here: http://www.equator-network.org/reporting-guidelines/strobe/

11) When completing the STROBE checklist, please use section and paragraph numbers, rather than page numbers.

12) Please avoid assertions of primacy (“This is the first study to...."). Instead, state “to our knowledge” or similar.

13) Was ethnicity or race considered in this study?

14) In your statistical analyses, please account for clustering of data at the school level. Generalized Estimating Equations (GEE) or hierarchical/ multilevel models, among others, may be useful in this case.

15) We note the potential for unobservable confounding in this observational study. Have you considered using robust methods such as propensity score matching to address this?

16) Please discuss whether missing scores were missing (at random/ completely at random) and the justification for your approach to handling missing data.

17) In the results text and tables, please quantify the main results with both 95% CIs and p values (where applicable).

18) Throughout the text, please remove language that implies causality, such as “more likely explained by” in the discussion and conclusion sections. Refer to associations instead.

19) References:

a) Please select the PLOS Medicine reference style in your citation manager. In-text reference call outs should be presented as follows noting the absence of spaces within the square brackets: "... countries [1,2]."

b) In the reference section, please ensure six names appear before et al.

20) Please remove the “Data sharing”, “Declaration of Interests”, “transparency”, and “Funding” statements at the end of the main text. This information is captured in the metadata obtained in the submission form.

Comments from the reviewers:

Reviewer #1: "School performance in Danish children exposed to maternal type 1 diabetes in utero: a nationwide cohort study" studies both the possible adverse effects of mothers with Type 1 diabetes (O-mT1D, n=2,144) and fathers with Type 1 diabetes (O-fT1D, n=3,474) on their offspring, against a background population (O-BP, n=616,455). This was performed as a population-based retrospective cohort study with data from 2010 to 2016 using nationally-standardized test scores for reading and math, at various public school grade levels. Adjusted (pooled) analyses found statistically-significant (adverse) mean test score difference of -1.62 for o-mT1D and −0.86 for O-fT1D, against the control group. Various sensitivity analyses involving different sets of covariates were attempted (Table 2, Supplement Table 1), with generally consistent outcomes for the T1D vs. control groups.

The scale of the study and the examination of fathers with Type 1 diabetes are particular strengths, with the latter relatively novel to the area, and relevant to answering hypotheses about potential mechanisms for the observed adverse effects relating to non-maternal factors. Some issues might however be considered:

1. Inclusion/exclusion and group criteria might be considered to be summarized in a flowchart if possible.

2. Related to the above, the treatment of other diabetes types (Type 2, gestational, etc.) in both parents and offspring might be clarified, prefably also in the flowchart if included. Were they excluded/covariates?

3. In Line 99, it is stated that "Offspring in whom both the mother and the father were diagnosed with diabetes (n=30, Line 192) were excluded". It might be clarified if this refers to T1D only for both, and the number of such cases.

4. In Line 144, it is stated that test scores were analyzed in a pooled analysis, with each test score (assumed for different topics/grades) as a unique observation. It might be discussed as to whether pooling across topics/time is appropriate, for example since reading scores would seem to be represented twice as much as math scores, simply by (arbitrary) virtue of having twice as many tests taken. More details on how the pooling was setup might be provided, possibly in supplementary material. Nonetheless, individual tests appear to generally support the main findings (Figure 1).

5. While some prior work has been cited as justification for the choice of covariates (Line 152), certain common covariates (e.g. age at each test, ethnicity, individual school performance) oft considered in studies involving student test scores, appear not to have been included. It might be briefly clarified as to whether such effects are assumed to have been covered by other included covariates.

6. In Line 207, it is stated that "The mean of test scores in both math and reading were 54.2 (SD 24.8) in O-mT1D, 54.4 (SD 24.8) in O-fT1D, and 56.4 (SD 24.7) in O-BP". It is assumed that the reported figures are the means for (math and reading) together, and not that the individual means (and SD) were identical; as such, the phrasing might be considered to be reworded (including whether the weighing for math and reading were identical), and results for individual topics might be considered for discussion.

Minor issues:

7. In Line 88, "inclusion of only offspring born in Denmark in whom there was..." might be considered to be reworded to avoid the phrase "only offspring", which might suggest children without siblings, if that is not the intended meaning.

Reviewer #2: School performance in Danish children exposed to maternal type 1 diabetes in utero: a nationwide cohort study

Overall:

An interesting twist to examination of the effect of in utero exposure to type 1 diabetes is described by using a comparison group of fathers with type 1 diabetes rather than just those offspring without exposure to type 1 diabetes in utero. The linking of databases in Denmark provides a unique opportunity to examine the outcomes of interest.

Children with both parents having type 1 diabetes were excluded, n=30.

From what I can tell rates of diabetes were increasing in the female population of reproductive age during the gestation of the offspring studied, mostly type 1 but there could have been an increase in type 2 and gestational diabetes. My question is, were these offspring, from women with type 2 or gestational diabetes during the pregnancy, removed from the comparison groups?

The introduction brings up the point that there are few human studies that examine the type 1, type 2 and gestational diabetes separately and these are only small studies. I would have thought the authors may have been able to distinguish between the type of in utero diabetes exposure in this population using ICD codes. These fetus' are exposed to hyper- and hypo-glycemia also. I feel that these offspring do not belong in the comparison groups. However, given the sample size this may be a moot point, but at least worth mentioning in the limitations.

Specific questions:

Lines 116-117: I am not sure of the importance of the likelihood of attending a private school as an outcome. I see the use of the likelihood of missing a test in the public school but am not clear about the former. Reporting of the results lines181 to 183 clarifies this. Please make it clear in the methods the reasoning for this "outcome", although I believe it is more a matter of checking that the population under study is representative.

Lines 190-192: It would be useful to the reader to repeat the number of offspring.

"2144 offspring of 1597 unique mothers and 3474 offspring of 2419 fathers with type 1 diabetes, and 616 455 offspring of 401 406 mothers and 402 785 fathers from the background population were included in the cohort."

Of note - I don't believe they were all adolescents unless I am misunderstanding the population. Mean age at first test was 10.1, but as many had more than one test, I am not sure what age is representative of the population. It might be useful to put age at first test for each group in table 1 under the "Test" as age at test is included in the models.

I am surprised by inclusion of both maternal and paternal education and income in model 3, I am presuming collinearity was checked. Sorry, I am confused, in re-reading the notes I see that parental highest education level and income were included in the model. Is that highest income or total, is it a category of percentile rank or a continuous variable? Please clarify. Maybe it would be preferable to only show in the table the variables included in the model.

The differences between math and reading results are fascinating and I feel that something is lost by putting both in the same model, there also appear to be grade differences. Model 3 is also used for these analyses, were other models considered?

Reviewer #3:

Thank you for the opportunity to review this paper.

Utilising national population data, the authors primarily examined the association between maternal typeI diabetes on offspring school performance, but also, in respect to fathers with type1 diabetes and the background population. This is an excellent study and the paper is well written and succinct. Inclusion of fathers with diabetes is particularly important and adds significantly to our understanding of developmental origins of human capital.

Minor points for consideration:

1. The inclusion of both parity and number of siblings may over-adjust the model - have the authors examined collinearity effects and is there strong enough justification to include both variables?

2. As anticipated, sex does have an association with school performance and is appropriate to adjust for. However, there are data suggestion a sexual dimorphism in the effect of maternal glucose on fatal growth, particularly for males. Given the sufficient power of the sample size, it would useful to sex-stratify and examine the association of maternal and paternal diabetes on the math outcome in sex-specific models to test if diabetes exposure effects male school performance more.

3. Figure 1 is useful to include but the view quality of the figure needs to be improved

A potential confounder is how well maternal glucose was managed during pregnancy - it is unfortunate that no data could be presented on this, but the authors have discussed this within the limitation section.

Any attachments provided with reviews can be seen via the following link:

[LINK]

28 Feb 2022

Submitted filename: Response to reviewers.docx

Click here for additional data file.

16 Mar 2022

Dear Dr. Spangmose,

Thank you very much for re-submitting your manuscript "School performance in Danish children exposed to maternal type 1 diabetes in utero: a nationwide retrospective cohort study" (PMEDICINE-D-21-04566R2) for review by PLOS Medicine.

I have discussed the paper with my colleagues and the academic editor and it was also seen again by one reviewer. I am pleased to say that provided the remaining editorial and production issues are dealt with we are planning to accept the paper for publication in the journal.

Please address the remaining issues within 1 week to enable publication of your manuscript in April. The remaining issues that need to be addressed are listed at the end of this email. Any accompanying reviewer attachments can be seen via the link below. Please take these into account before resubmitting your manuscript:

[LINK]

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

In revising the manuscript for further consideration here, please ensure you address the specific points made by each reviewer and the editors. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments and the changes you have made in the manuscript. Please submit a clean version of the paper as the main article file. A version with changes marked must also be uploaded as a marked up manuscript file.

Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. If you haven't already, we ask that you provide a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract.

We expect to receive your revised manuscript within 1 week. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

We ask every co-author listed on the manuscript to fill in a contributing author statement. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript.

Please note, when your manuscript is accepted, an uncorrected proof of your manuscript will be published online ahead of the final version, unless you've already opted out via the online submission form. If, for any reason, you do not want an earlier version of your manuscript published online or are unsure if you have already indicated as such, please let the journal staff know immediately at plosmedicine@plos.org.

If you have any questions in the meantime, please contact me or the journal staff on plosmedicine@plos.org.

We look forward to receiving the revised manuscript by Mar 23 2022 11:59PM.

Sincerely,

Beryne Odeny,

PLOS Medicine

plosmedicine.org

-----------------------------------

1) Abstract conclusion: please add a sentence addressing the study implications for policy and practice.

2) Thank you for providing your STROBE checklist. Please replace the page numbers with paragraph numbers per section (e.g. "Methods, paragraph 1"), since the page numbers of the final published paper may be different from the page numbers in the current manuscript.

3) In table 2, please define what the income percentile ranks mean e.g. Q1 is which income range etc

4) To help us extend the reach of your research, please provide any Twitter handle(s) that would be appropriate to tag, including your own, your coauthors’, your institution, funder, or lab.

-------------------------

Requests from Editors:

Comments from Reviewers:

Reviewer #1: We thank the authors for addressing our previous comments. A few further points might be considered:

1. The clarifications for other diabetes types (involving Table R1 in the response) might be included in supplementary material, and/or briefly mentioned in the text, if possible.

2. The test score pooling setup might be described in greater detail, if possible (and offered).

3. Limitations relating to available covariates might be briefly discussed.

4. Results as presented in Table 2 appear changed from the previous revision, despite apparently involving the same population. This might be clarified.

Any attachments provided with reviews can be seen via the following link:

[LINK]

25 Mar 2022

Submitted filename: Response to reviewers vol. 2.docx

Click here for additional data file.

29 Mar 2022

Dear Dr Spangmose,

On behalf of my colleagues and the Academic Editor, Dr. Lars Åke Persson, I am pleased to inform you that we have agreed to publish your manuscript "School performance in Danish children exposed to maternal type 1 diabetes in utero: a nationwide retrospective cohort study" (PMEDICINE-D-21-04566R3) in PLOS Medicine.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. Please be aware that it may take several days for you to receive this email; during this time no action is required by you. Once you have received these formatting requests, please note that your manuscript will not be scheduled for publication until you have made the required changes.

In the meantime, please log into Editorial Manager at http://www.editorialmanager.com/pmedicine/, click the "Update My Information" link at the top of the page, and update your user information to ensure an efficient production process.

PRESS

We frequently collaborate with press offices. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximise its impact. If the press office is planning to promote your findings, we would be grateful if they could coordinate with medicinepress@plos.org. If you have not yet opted out of the early version process, we ask that you notify us immediately of any press plans so that we may do so on your behalf.

We also ask that you take this opportunity to read our Embargo Policy regarding the discussion, promotion and media coverage of work that is yet to be published by PLOS. As your manuscript is not yet published, it is bound by the conditions of our Embargo Policy. Please be aware that this policy is in place both to ensure that any press coverage of your article is fully substantiated and to provide a direct link between such coverage and the published work. For full details of our Embargo Policy, please visit http://www.plos.org/about/media-inquiries/embargo-policy/.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Thank you again for submitting to PLOS Medicine. We look forward to publishing your paper.

Sincerely,

Beryne Odeny

PLOS Medicine