Abstract
IMPORTANCE: Atopic dermatitis (AD) may impair school performance through sleep disturbance, comorbidities, or psychosocial effects; however, population-based longitudinal evidence is limited. OBJECTIVE: To investigate the association of AD with academic performance and whether this association is modified or confounded by disease phenotype and socioeconomic background. DESIGN, SETTING, AND PARTICIPANTS: Parallel, population-based cohort studies in Denmark and England, including children with or without AD. In Denmark, children born between 1986 and 2000 from nationwide linked registries were followed up through 2018. In England, children born between 1991 and 1992 from the Avon Longitudinal Study of Parents and Children, linked to the National Pupil Database, were followed up through 2009. Data analysis was conducted from October 2022 to April 2024. EXPOSURES: AD, defined as hospital diagnosis before age 13 years (Denmark) or 2 or more maternal reports of AD by age 11.5 years (England). Danish analyses were repeated in exposure-discordant full siblings to account for family factors. MAIN OUTCOMES AND MEASURES: Academic performance on final compulsory national examinations at approximately age 16 years: nonpassing grade (binary outcome; Poisson regression prevalence ratios) and mean academic performance score (continuous outcome; linear regression mean differences). RESULTS: A total of 782 837 children were included between both studies. In Denmark (n = 776 214; 10 259 children with AD; 386 408 [49.8%] female), the prevalence of nonpassing grades (12.0% vs 11.2%; fully adjusted prevalence ratio [aPR], 1.06 [95% CI, 1.01-1.12]) and mean academic performance scores (fully adjusted mean difference, -0.06 points [95% CI, -0.11 to -0.01]) were similar between children with vs without AD. Active AD was associated with a higher prevalence of nonpassing grades than no AD (aPR, 1.19 [95% CI, 1.03-1.37]). Sibling analyses yielded similar results. In England (n = 6623; 2967 children with AD; 3332 [50.3%] female), nonpassing grades were slightly less common (37.7% vs 47.4%; aPR, 0.88 [95% CI, 0.83-0.93]) and mean performance scores were higher (10.48 points [95% CI, 6.49-13.86]) in children with vs without AD. Phenotype analyses showed that better performance in children with moderate-declining and moderate-frequent AD primarily explained these results, whereas those with mild-intermittent or severe-frequent AD performed similarly to the unaffected or rare comparator. Overall, no consistent variation by disease phenotype or socioeconomic background was found across academic performance outcomes and settings. CONCLUSIONS AND RELEVANCE: In these parallel cohort studies, triangulation across 2 large cohorts and multiple analyses provided reassurance that AD is likely not associated with meaningfully diminished academic performance among adolescents taking national compulsory examinations.