Azim Syeda Ishra, Mallitt Kylie-Ann, Wong Tsz Ying, Lin Ping-I, Lam-Cassettari Christa, Masi Anne, Diaz Antonio Mendoza, Montgomery Alicia, McHugh Catherine, Whitehouse Andrew, Dadds Mark R, Eapen Valsamma
Discipline of Psychiatry and Mental Health, School of Clinical Medicine, UNSW Medicine & Health, University of New South Wales, Sydney, Australia.
School of Public Health, University of Sydney, Sydney, Australia.
J Autism Dev Disord. 2025 Aug 13. doi: 10.1007/s10803-025-06983-w.
Approximately 8% of all children experience developmental and mental health conditions. Similarities in characteristics across neurodevelopmental conditions-such as difficulties in communication and language, social interaction, motor coordination, attention, activity regulation, behavior, mood, and sleep-make it challenging to attribute these characteristics exclusively to specific diagnoses and assessments. The purpose of this study was to identify symptomatic domains across neurodevelopmental conditions in children and to explore dimension reduction for transdiagnostic assessment. Cluster and factor analyses were conducted on two datasets to determine the co-occurrence of conditions in children. Deidentified data were sourced from two data repositories, the Australian Autism Biobank and the Child Behavior Research Clinic. Four clusters were identified in the behavior clinic dataset based on emotional, behavioral, and social characteristics. The autism biobank dataset revealed three distinct clusters categorized by cognitive skills, adaptive skills, and autistic traits. Factor analysis was subsequently used to explore which characteristics could be grouped into domains. The study aimed to characterize and capture the complexities of co-occurring conditions within neurodevelopmental conditions by identifying distinct domains derived from both cluster and factor analysis. By integrating findings from two clinical datasets with children having both behavioral and neurodevelopmental conditions, we highlighted how clusters of comorbidity and behavioral profiles align with factors. Thus, the findings emphasizes the importances of adopting a transdiagnostic approach to diagnosis, rather than viewing neurodevelopmental conditions in isolation.
约8%的儿童存在发育和心理健康问题。神经发育障碍在特征上具有相似性，比如在沟通与语言、社交互动、运动协调、注意力、活动调节、行为、情绪和睡眠方面存在困难，这使得将这些特征完全归因于特定诊断和评估具有挑战性。本研究的目的是确定儿童神经发育障碍中的症状领域，并探索用于跨诊断评估的降维方法。对两个数据集进行了聚类分析和因子分析，以确定儿童中多种疾病的共现情况。去识别化的数据来自两个数据存储库，即澳大利亚自闭症生物样本库和儿童行为研究诊所。基于情绪、行为和社会特征，在行为诊所数据集中识别出了四个聚类。自闭症生物样本库数据集揭示了按认知技能、适应技能和自闭症特征分类的三个不同聚类。随后使用因子分析来探索哪些特征可以归为同一领域。该研究旨在通过识别聚类分析和因子分析得出的不同领域，来描述和捕捉神经发育障碍中共存疾病的复杂性。通过整合来自两个临床数据集（针对患有行为和神经发育障碍的儿童）的结果，我们强调了共病聚类和行为概况与因子是如何相互匹配的。因此，研究结果强调了采用跨诊断方法进行诊断的重要性，而不是孤立地看待神经发育障碍。
