Relative Age and ADHD Diagnosis in U.S. Elementary Schools: Evidence from a National Longitudinal Study
Saved in:
| Title: | Relative Age and ADHD Diagnosis in U.S. Elementary Schools: Evidence from a National Longitudinal Study |
|---|---|
| Language: | English |
| Authors: | Eric Hengyu Hu (ORCID |
| Source: | Society for Research on Educational Effectiveness. 2025. |
| Availability: | Society for Research on Educational Effectiveness. 2040 Sheridan Road, Evanston, IL 60208. Tel: 202-495-0920; e-mail: contact@sree.org; Web site: https://www.sree.org/ |
| Peer Reviewed: | Y |
| Publication Date: | 2025 |
| Document Type: | Reports - Research |
| Education Level: | Elementary Education Early Childhood Education Kindergarten Primary Education |
| Descriptors: | Children, Longitudinal Studies, Surveys, Elementary School Students, Kindergarten, Attention Deficit Hyperactivity Disorder, Educational Diagnosis, Age Differences, Grade Level Differences, Student Behavior, Predictor Variables, Public Schools, Student Characteristics |
| Assessment and Survey Identifiers: | Early Childhood Longitudinal Survey |
| Abstract: | Background: Attention deficit hyperactivity disorder (ADHD) is a highly prevalent disorder, with an estimated 7 million (11.4%) U.S. children aged 3 to 17 years have ever been diagnosed (Danielson et al., 2024). Characterized by persistent inattention, hyperactivity, and impulsivity, ADHD is associated with significant academic, social, and psychological challenges (Caye et al., 2020; Holland & Sayal, 2019; Whitely et al., 2019). While early diagnosis and intervention are important, diagnostic rates vary widely, raising concerns about both underdiagnosis and overdiagnosis. A notable factor influencing diagnosis is a child's relative age in their school-grade cohort--determined by school entry cutoff dates (Caye et al., 2020). Children who are younger relative to classmates may show age-typical behaviors, such as high activity or inattention, that resemble ADHD symptoms (Elder, 2010; Layton et al., 2018). Studies across countries have consistently found that relatively younger children are more likely to be diagnosed with ADHD than those who are older, suggesting developmental immaturity may be mistaken for disorder-related behaviors (Whitely et al., 2019). Teacher referrals--often the first step in diagnosis--may unintentionally compare younger children to older peers, leading to potential misdiagnosis. Given the educational and psychological implications of an ADHD label, understanding the extent of the relative age effect in U.S. school settings is essential. Study's Purpose: To address these concerns, this study examines the relationship between children's relative age and ADHD diagnosis in U.S. elementary schools using a nationally representative longitudinal dataset. Specifically, we assess whether relative age at kindergarten entry predicts the likelihood and timing of an ADHD diagnosis and whether this relationship is moderated by key sociodemographic factors. Methods: The study draws on data from the Early Childhood Longitudinal Study--Kindergarten Cohort of 2010-11 (ECLS-K:2011), which followed a nationally representative group of children who began kindergarten in fall 2010 or spring 2011. From the original 18,170 participants, we excluded those missing birthdate information (n = 10), living in states without a defined kindergarten entry cutoff (n = 3,350), lacking ADHD diagnosis data (n = 2,330), or attending private schools (n = 2,190). The final analytic sample included 11,410 children. ADHD Diagnosis (Outcome): ADHD diagnoses were reported by parents during spring surveys in each elementary grade. We created two outcome variables: (1) newly diagnosed ADHD in each grade and (2) cumulative ADHD diagnosis by fifth grade. Relative Age (Predictor): We calculated each child's age at kindergarten entry based on their birthdate and state-specific cutoff date (which varied by state). Relative age was calculated as the difference in months between a child's age and the benchmark of 60 months (5 years old). A higher score reflected a younger relative age. In addition to this continuous measure, we created a binary indicator: children with a score > 0 were classified as "Relatively Younger" (n = 520), and those [less than or equal to] 0 as "Older or Exact Age 5" (n = 10,890). Covariates: We controlled for a comprehensive set of covariates measured at kindergarten, including sociodemographic characteristics, family background, early academic performance, behavioral and executive functioning, and school context factors (see Supplemental Tables S3-S5). Analyses: We used logistic regression to examine whether relative age predicted ADHD diagnosis in both grade-specific and cumulative models. We estimated three models: (a) unadjusted models with only relative age as a predictor, (b) adjusted models including all covariates, and (c) moderation models including interactions between relative age and child characteristics (race/ethnicity, gender, SES, home language). To analyze timing of diagnosis, we employed discrete-time logistic regression using the same three model specifications. All models accounted for clustering of students within schools, and missing covariate data were handled using multiple imputation. There were no missing values for the key outcome or predictor variables. We conducted several "robustness checks": 1. Replaced continuous relative age with the binary version. 2. Used ADHD medication use (parent-reported) as an alternative outcome. 3. Excluded children born in the cutoff month (relative age = 0), and in an extended analysis, also excluded children born one month before and after the cutoff (relative age = -1, 0, or 1). The supplemental material includes: descriptive statistics for the overall sample (Table S1); the sample distribution by state kindergarten entry age policy (Table S2); detailed estimation results for additional covariates corresponding to Tables 1-3 (Tables S3-S5); and results from the robustness checks (Tables S6 and S7 using the dummy-coded predictor; Table S8 using medication usage as the outcome; Tables S9 and S10 excluding children born in or near the kindergarten cutoff month). Results: Table 1 displays odds ratios for relative age predicting newly diagnosed ADHD in each grade. Across both unadjusted and adjusted models, relative age was not significantly associated with diagnosis at any grade. Table 2 presents odds ratios for cumulative ADHD diagnosis by fifth grade, and again, no significant associations were found in any model. Table 3 provides results from discrete-time logistic regressions, showing no significant association between relative age and the timing of ADHD diagnosis. The moderation models also revealed no significant interactions with sociodemographic factors. Conclusion: Our findings suggest limited evidence supporting the relative age effect on ADHD diagnoses among elementary school-aged children in the United States. Across multiple analyses--including grade-specific diagnoses, cumulative diagnoses, and timing of diagnoses--we found no significant overall association between children's younger relative age and increased likelihood of an ADHD diagnosis. |
| Abstractor: | As Provided |
| Entry Date: | 2026 |
| Access URL: | https://www.sree.org/2025-conference |
| Accession Number: | ED677796 |
| Database: | ERIC |
| Abstract: | Background: Attention deficit hyperactivity disorder (ADHD) is a highly prevalent disorder, with an estimated 7 million (11.4%) U.S. children aged 3 to 17 years have ever been diagnosed (Danielson et al., 2024). Characterized by persistent inattention, hyperactivity, and impulsivity, ADHD is associated with significant academic, social, and psychological challenges (Caye et al., 2020; Holland & Sayal, 2019; Whitely et al., 2019). While early diagnosis and intervention are important, diagnostic rates vary widely, raising concerns about both underdiagnosis and overdiagnosis. A notable factor influencing diagnosis is a child's relative age in their school-grade cohort--determined by school entry cutoff dates (Caye et al., 2020). Children who are younger relative to classmates may show age-typical behaviors, such as high activity or inattention, that resemble ADHD symptoms (Elder, 2010; Layton et al., 2018). Studies across countries have consistently found that relatively younger children are more likely to be diagnosed with ADHD than those who are older, suggesting developmental immaturity may be mistaken for disorder-related behaviors (Whitely et al., 2019). Teacher referrals--often the first step in diagnosis--may unintentionally compare younger children to older peers, leading to potential misdiagnosis. Given the educational and psychological implications of an ADHD label, understanding the extent of the relative age effect in U.S. school settings is essential. Study's Purpose: To address these concerns, this study examines the relationship between children's relative age and ADHD diagnosis in U.S. elementary schools using a nationally representative longitudinal dataset. Specifically, we assess whether relative age at kindergarten entry predicts the likelihood and timing of an ADHD diagnosis and whether this relationship is moderated by key sociodemographic factors. Methods: The study draws on data from the Early Childhood Longitudinal Study--Kindergarten Cohort of 2010-11 (ECLS-K:2011), which followed a nationally representative group of children who began kindergarten in fall 2010 or spring 2011. From the original 18,170 participants, we excluded those missing birthdate information (n = 10), living in states without a defined kindergarten entry cutoff (n = 3,350), lacking ADHD diagnosis data (n = 2,330), or attending private schools (n = 2,190). The final analytic sample included 11,410 children. ADHD Diagnosis (Outcome): ADHD diagnoses were reported by parents during spring surveys in each elementary grade. We created two outcome variables: (1) newly diagnosed ADHD in each grade and (2) cumulative ADHD diagnosis by fifth grade. Relative Age (Predictor): We calculated each child's age at kindergarten entry based on their birthdate and state-specific cutoff date (which varied by state). Relative age was calculated as the difference in months between a child's age and the benchmark of 60 months (5 years old). A higher score reflected a younger relative age. In addition to this continuous measure, we created a binary indicator: children with a score > 0 were classified as "Relatively Younger" (n = 520), and those [less than or equal to] 0 as "Older or Exact Age 5" (n = 10,890). Covariates: We controlled for a comprehensive set of covariates measured at kindergarten, including sociodemographic characteristics, family background, early academic performance, behavioral and executive functioning, and school context factors (see Supplemental Tables S3-S5). Analyses: We used logistic regression to examine whether relative age predicted ADHD diagnosis in both grade-specific and cumulative models. We estimated three models: (a) unadjusted models with only relative age as a predictor, (b) adjusted models including all covariates, and (c) moderation models including interactions between relative age and child characteristics (race/ethnicity, gender, SES, home language). To analyze timing of diagnosis, we employed discrete-time logistic regression using the same three model specifications. All models accounted for clustering of students within schools, and missing covariate data were handled using multiple imputation. There were no missing values for the key outcome or predictor variables. We conducted several "robustness checks": 1. Replaced continuous relative age with the binary version. 2. Used ADHD medication use (parent-reported) as an alternative outcome. 3. Excluded children born in the cutoff month (relative age = 0), and in an extended analysis, also excluded children born one month before and after the cutoff (relative age = -1, 0, or 1). The supplemental material includes: descriptive statistics for the overall sample (Table S1); the sample distribution by state kindergarten entry age policy (Table S2); detailed estimation results for additional covariates corresponding to Tables 1-3 (Tables S3-S5); and results from the robustness checks (Tables S6 and S7 using the dummy-coded predictor; Table S8 using medication usage as the outcome; Tables S9 and S10 excluding children born in or near the kindergarten cutoff month). Results: Table 1 displays odds ratios for relative age predicting newly diagnosed ADHD in each grade. Across both unadjusted and adjusted models, relative age was not significantly associated with diagnosis at any grade. Table 2 presents odds ratios for cumulative ADHD diagnosis by fifth grade, and again, no significant associations were found in any model. Table 3 provides results from discrete-time logistic regressions, showing no significant association between relative age and the timing of ADHD diagnosis. The moderation models also revealed no significant interactions with sociodemographic factors. Conclusion: Our findings suggest limited evidence supporting the relative age effect on ADHD diagnoses among elementary school-aged children in the United States. Across multiple analyses--including grade-specific diagnoses, cumulative diagnoses, and timing of diagnoses--we found no significant overall association between children's younger relative age and increased likelihood of an ADHD diagnosis. |
|---|
