Trends and Mechanisms Linking Pediatric Attention-Deficit/Hyperactivity Disorder and Atopic Disorders

Literature retrieval was conducted in the Web of Science Core Collection (WoSCC) and Scopus databases, covering the period from 2005 to 2025. Keyword selection was based on broad field definitions and relevant terminology, including medical terms, Medical Subject Headings (MeSH) vocabulary, and synonyms related to AD and ADHD. The search strategy comprised the following topic sets: (“Attention-deficit/hyperactivity disorder” OR “ADHD”) AND (“atopic disorders” OR “asthma” OR “atopic eczema” OR “allergic rhinitis”) AND (“child*”), using the TS field in WoSCC and the TITLE-ABS-KEY field in Scopus. The literature screening process (Figure 1) was performed independently by two researchers; disagreements were resolved with the involvement of a third researcher who verified the final results. To ensure the scientific quality and consistency of included records, only English-language original research articles and review articles were retained for analysis. Under this search scheme, 498 records from WoSCC and 697 records from Scopus were collected and downloaded. After removal of duplicates, 838 eligible publications were included for bibliometric analyses and molecular-mechanism exploration of the association between AD and ADHD.

To systematically investigate research trends and interaction mechanisms between AD and ADHD, we employed a series of bibliometric and bioinformatic tools. Bibliometric analyses were performed using CiteSpace (6.3.R3), VOSviewer (1.6.20), and the bibliometrix R package (v5.2.1). Quantitative indicators of publication output and collaboration patterns were calculated using bibliometrix. VOSviewer was used for in-depth analyses of co-authorship networks and keyword co-occurrence. CiteSpace was applied to map and cluster countries, institutions, journals, and keywords, and to generate a dual-map overlay of citation trajectories. Parameter settings were as follows: the time span was set to 2001–2025 with a 1-year time slice; institutions and keywords were specified as node types. The path selection method was set to “pathfinder”, while all other parameters were kept at default values. The K value was set to 25, and keyword clustering was performed using the log-likelihood ratio (LLR) algorithm.

For molecular mechanism exploration, GeneCards (https://www.genecards.org/) was used to screen potential gene targets associated with AD and ADHD, and corresponding target screening was also performed for the AD subtypes asthma, atopic eczema, and allergic rhinitis. Based on overlaps among disease-associated genes, Cytoscape (v3.10.0) with StringApp (v2.2.0) was further used to interrogate molecular interactions and construct protein–protein interaction (PPI) networks. These networks were visualized and analyzed in Cytoscape to identify hub genes and key modules. Enrichment analyses of shared gene targets were conducted to elucidate their roles in biological processes, molecular functions, and disease pathways. Multiple R packages—clusterProfiler, ggplot2, ComplexHeatmap, enrichplot, and DOSE—were used to perform Disease Ontology (DO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Ontology (GO) enrichment analyses.