Neurodevelopmental disorders in pediatric patients with atopic dermatitis: a systematic review and meta-analysis

Background Neurodevelopmental disorders (NDs) are increasingly recognized among children with atopic dermatitis (AD). This systematic review and meta-analysis aimed to summarize the prevalence of NDs in pediatric patients with AD and quantify the associations between AD and specific neurodevelopmental outcomes. Methods We searched Embase, PubMed, Cochrane Library, Web of Science, and PsycINFO from their inception to 1 December 2025. Observational studies reporting adjusted estimates and/or prevalence for NDs in pediatric patients with AD were included. Exploratory subgroup analyses were undertaken to ascertain sources of heterogeneity. Funnel plots and Egger's test were used to assess publication bias. Results From 2,514 initial records, 39 studies reporting ND outcomes in pediatric patients with AD were included. The pooled prevalence rates for NDs were as follows: attention-deficit/hyperactivity disorder (ADHD, 6.01%), autism spectrum disorder (ASD, 0.81%), learning disability (LD, 11.34%), conduct disorder (CD, 2.76%), and peer problems (5.98%). AD was associated with higher odds of ADHD (OR = 1.34, 95% CI, 1.21–1.50), ASD (OR = 2.11, 95% CI, 1.07–4.16), LD (OR = 1.71, 95% CI, 1.41–2.06), intellectual disability (OR = 1.39, 95% CI, 1.33–1.45), and CD (OR = 1.35, 95% CI, 1.16–1.58), but not peer problems (OR = 1.10, 95% CI, 0.96–1.26). HR-based estimates for ADHD (HR = 1.54, 95% CI, 0.97–2.44) and ASD (HR = 2.09, 95% CI, 0.54–8.19) were directionally similar but less precise. Substantial heterogeneity was present across several analyses. Conclusion Pediatric AD was associated with higher odds of several neurodevelopmental outcomes compared with children without AD. However, substantial heterogeneity, potential detection bias, and the observational nature of the available evidence limit causal interpretation. Further well-designed longitudinal studies using standardized exposure and outcome definitions, adequate control of confounding, and clearer temporal characterization are needed to clarify the consistency and directionality of these associations. Systematic Review Registration https://www.crd.york.ac.uk/PROSPERO/view/CRD42024545405 , PROSPERO CRD42024545405.