Of the 18 174 children enrolled in the study, 14 350 children were determined to be term gestation of at least 37 weeks’, 773 were born at 35–36 6/7 weeks gestation, and 422 were born at 32–34 6/7 weeks gestation. One hundred and twenty-two were very preterm (28–31 6/7 weeks gestation) and 35 were born extremely preterm (<28 weeks gestation). Given the small sample size, and the focus of this study, children who were born at less than 32 weeks’ were excluded from analysis. Gestational age was unable to be determined for 2 472 children as their parents did not complete the birth history section; these children were excluded from the analysis. Demographic information of the three groups of interest is shown in Table 1 with p-values from Bonferroni correction analysis. Notably, compared with term infants, there were more male children and more African American children in both of the preterm groups, which is similar to published data that show males and African American infants are more likely to be born preterm [22, 23]. There was less Hispanic children and less children from suburban locations in the 35-36 6/7 weeks’ sample group compared to the term children group, but it was not a significant difference. More children were sampled from the southern region of the UA in all groups. In all three groups, the most common parental education level was less than four years of college, but there was no significant difference between the three groups.

When comparing the academic scores in reading, mathematics, and science from the direct child assessments, children born 35–36 6/7 week had no significant difference from term children in reading, mathematics, or science (Fig. 1). Children born 32–34 6/7 weeks’ had significantly lower scores compared to term children for reading in kindergarten, for science in fifth grade, and for mathematics in kindergarten, second grade, and fourth grade. When comparing individuals over a longitudinal analysis over the six years of testing, gestational age was found to be a significant effector on the model, with a p value of <0.0001 (Supplementary Table 1).

Mean test scores for 32–34 6/7 weeks gestation children and 35–36 6/7 weeks gestation children compared to children at least 37 weeks gestation from kindergarten through fifth grade in A Reading, B Mathematics, and C Science. Asterisk indicates statistical significance of p < 0.05. Analyses were adjusted for gender, race, household income level, school type, and home setting. K stands for Kindergarten.

Teacher assessments were completed for each child in the study to describe if they were below, equal to, or above the average performance for grade level. When comparing children born 32–34 6/7 weeks’ to term children, 32–34 6/7 weeks’ children were more likely to perform significantly below grade average in reading, mathematics, and science in kindergarten with adjusted odds ratios of 1.40 (95th percent confidence interval CI 1.03, 1.89), 1.49 (CI 1.02, 2.17), and 1.89 (CI 1.20, 2.99), respectively (Fig. 2). Significance was lost after kindergarten in each subject. For 35-36 6/7 weeks’ children compared to term children, different trends emerged. In reading, these children appeared to improve as they progressed in grade level; their odds of performing below grade level dropped from 1.19 (CI 0.89, 1.60) to 0.92 (CI 0.71, 1.19) from kindergarten to fifth grade, respectively. In mathematics, 35-36 6/7 weeks’ children were at higher odds of being below grade level performance as grade level increased, from 1.09 (CI 0.84, 1.42) to 1.23 (CI 0.87, 1.73). Science performance for 35–36 6/7 weeks’ children did not exhibit a clear trend, with an adjusted odds ratio for being below grade level that ranged from 1.05 (CI 0.83, 1.32) to 1.29 (CI 0.91, 1.81) between grades.

Odds ratio for student to have below average performance as rated by the teacher from kindergarten through fifth grade in subjects: reading for A 32–34 6/7 weeks gestation children and B 35–36 6/7 weeks gestation children; mathematics for C 32–34 6/7 weeks gestation children and D 35–36 6/7 weeks gestation children; and science for E 32–34 6/7 weeks gestation children and F 35–36 6/7 weeks gestation children. Higher number indicates worse performance. Analyses were adjusted for gender, race, household income level, school type, and home setting. K stands for Kindergarten.

Individualized education plans (IEP) were on file for a significantly larger percentage of 35–36 6/7 weeks’ children compared to term children in kindergarten and first grade, and for every grade from kindergarten to fifth grade in 32–34 6/7 weeks’ children (Fig. 3A). For example, at first grade, 16% of 32–34 6/7 weeks’, 12.4% of 35–36 6/7 weeks’, and 9.5% of children at least 37 weeks gestation had an IEP on file, while at fifth grade these percentages had increased to 22.4%, 15.5%, and 14.2%, respectively. When comparing children born 32–34 6/7 weeks’ to term children, the adjusted odds ratio for an IEP in a 32–34 6/7 weeks’ and 35–36 6/7 weeks’ child was higher when compared to a term child (Fig. 4A), with the highest odds at kindergarten for both 32–34 6/7 weeks’ and 35–36 6/7 weeks’ children at 2.43 (CI 1.48, 4.0) and 1.44 (CI 0.96, 2.16), respectively.

A Unadjusted percentage of children with individualized education plan (IEP) on file with the school and B Unadjusted percentage of children with a learning disability on file with the school. Learning disability included a diagnosis of speech impairment, intellectual disability, visual impairment, hearing impairment, autism, or developmental delay. Asterisk indicates statistical significance of p < 0.05 compared to at least 37 weeks’ children. K stands for Kindergarten.

Odds ratio for A 32–34 6/7 weeks gestation and B 35–36 6/7 weeks gestation children to have an individualized education plan (IEP) on file with the school. Odds ratio for C 32–34 6/7 weeks gestation and D 35–36 6/7 weeks gestation children to have a learning disability on file with the school. Higher number indicates higher odds. Learning disability included a diagnosis of speech impairment, intellectual disability, visual impairment, hearing impairment, autism, or developmental delay. Analyses were adjusted for gender, race, household income level, school type, and home setting. K stands for Kindergarten.

Learning disabilities were also found in a statistically significantly higher proportion of children born 32–34 6/7 weeks’ and 35–36 6/7 weeks’ compared to term children in grades kindergarten through fifth grade, except for 35–36 6/7 weeks’ children in fifth grade (Fig. 3B). For example, in kindergarten, 32.9% of 32–34 6/7 weeks’ children, 26.5% of 35–36 6/7 weeks’ children, and 18.3% of term children had a learning disability diagnosis, while at fourth grade these percentages had shifted to 22.6%, 18.7%, and 14.6%, respectively. Learning disabilities included speech impairments, intellectual disability, visual impairments, hearing impairments, autism, and developmental delay. The adjusted odds ratio for a learning disability was elevated throughout kindergarten to fifth grade, with a 2.41 higher odds (CI 1.61, 3.60) in kindergarten (Fig. 4B). 32–34 6/7 weeks’ children also had the highest odds of a learning disability in kindergarten compared to term children, at 2.05 time higher odds (CI 1.56, 2.69). 32–34 6/7 weeks’ children continued to have elevated odds through fifth grade.