Fat-soluble vitamin deficiency in children with neurodevelopmental disorders: a cross-sectional study

Neurodevelopmental disorders (NDDs) affect 15-20% of children worldwide, and emerging evidence suggests nutritional status may influence neurodevelopmental outcomes. However, disorder-specific patterns of fat-soluble vitamin deficiency remain incompletely characterized in large clinical populations.

We conducted a retrospective cross-sectional study of 14,911 children aged 1-18 years who underwent serum vitamin testing between 2022 and 2025 at a tertiary pediatric center in China. Participants included healthy controls (n = 6391) and children with attention-deficit/hyperactivity disorder (ADHD; n = 4392), tic disorders (n = 2020), autism spectrum disorder (ASD; n = 1312), and comorbid neurodevelopmental disorders (n = 796). Serum concentrations of vitamin A, 25(OH)D, and vitamin E were measured using standardized assays. Group differences were assessed using age-, sex-, and season-adjusted analyses, including Analysis of covariance (ANCOVA) and multivariable logistic regression.

After adjustment, children with ADHD and tic disorders exhibited modestly lower serum 25(OH)D concentrations compared with controls (adjusted Cohen's d = 0.273 for ADHD, small effect), whereas ASD profiles were largely comparable to controls. Vitamin A deficiency affected approximately one quarter of participants across all groups, with a modest increase in ADHD (odds ratio [OR] = 1.26, 95% confidence interval [CI]: 1.14-1.40; small effect size), indicating widespread background deficiency with a modest additive risk in ADHD. Although absolute vitamin E deficiency rates were low, children with ADHD showed higher relative odds of deficiency (OR = 2.91). Multiple concurrent deficiencies (≥ 2 vitamins) were more common in ADHD (OR = 2.81) and moderately increased in tic disorders (OR = 1.66), while triple deficiencies remained rare. Children with comorbid conditions showed intermediate patterns. Clustering analyses identified heterogeneous vitamin profiles, with low-vitamin patterns disproportionately represented in ADHD and tic-disorder groups.

In this large pediatric cohort, fat-soluble vitamin status differed across NDD subtypes. Children with ADHD exhibited higher burdens of vitamin D deficiency and multiple concurrent deficiencies, while ASD profiles were comparatively preserved. These findings highlight nutritional heterogeneity among pediatric NDD populations and underscore the need for prospective studies to clarify temporal relationships and clinical relevance.