Premium
Aiding diagnosis of attention‐deficit/hyperactivity disorder and its subtypes: discriminant function analysis of event‐related potential data
Author(s) -
Smith Janette L.,
Johnstone Stuart J.,
Barry Robert J.
Publication year - 2003
Publication title -
journal of child psychology and psychiatry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.652
H-Index - 211
eISSN - 1469-7610
pISSN - 0021-9630
DOI - 10.1111/1469-7610.00191
Subject(s) - discriminant function analysis , psychology , attention deficit hyperactivity disorder , event related potential , audiology , attention deficit , developmental psychology , clinical psychology , psychiatry , electroencephalography , medicine , machine learning , computer science
Background: Robust evidence exists that the brain functioning of children and adolescents with attention‐deficit/hyperactivity disorder (AD/HD) differs from that of normal controls. The purpose of this study was to further investigate whether event‐related potential (ERP) measures of brain function could be used to reliably classify normal controls versus children and adolescents with two subtypes of AD/HD. Methods: Behavioural and topographic ERP data from children aged 8–12 years ( n = 75) and adolescents aged 13–18 years ( n = 75) were entered into stepwise discriminant function analyses separating controls and subjects with AD/HD, and also subjects with AD/HD‐Predominantly Inattentive type (AD/HDin) from those with AD/HD‐Combined type (AD/HDcom). Results: For children aged 8–12 years, controls and children with AD/HD could be separated with an overall classification accuracy of 73.3%, and the AD/HD subtypes were classified with 69.4% overall accuracy. For adolescents aged 13–18 years, 58.7% of the control vs. AD/HD subjects were correctly classified, and 62.7% of subjects in the subtypes. Classification function coefficients for each of these discriminations are presented for the purposes of cross‐validation. Conclusions: In children aged 8–12 years, analysis of ERP data may aid a clinician in diagnosing AD/HD, although the clinical utility of ERP analysis is reduced for adolescents aged 13–18 years. Suggestions are made to combine ERP measures with other measures of brain function in order to improve classification accuracy, and also to predict drug response in children diagnosed with AD/HD.