PT  - JOURNAL ARTICLE
AU  - Dimitri Marques Abramov
AU  - Saint Clair Gomes
AU  - Carlos Alberto Mourão Júnior
AU  - Adailton Pontes
AU  - Carla Quero Cunha Rodrigues
AU  - Monique Castro Pontes
AU  - Juliana Vieira
AU  - Paulo Ricardo Galhanone
AU  - Leonardo C. deAzevedo
AU  - Vladimir V. Lazarev
TI  - Estimating DSM accuracy for Attention Deficit Hyperactivity Disorder Based on Neurophysiological, Psychological, and Behavioral Correlates
AID  - 10.1101/126433
DP  - 2017 Jan 01
TA  - bioRxiv
PG  - 126433
4099  - http://biorxiv.org/content/early/2017/04/11/126433.short
4100  - http://biorxiv.org/content/early/2017/04/11/126433.full
AB  - Objective To find objective evidence of accuracy of the Diagnostic and Statistical Manual of Mental Disorders (DSM) in the diagnosis of Attention Deficit Hyperactivity Disorder (ADHD), by comparing classifications of subjects based on behavioral (Attentional Network Test - ANT), psychological (Wechsler Intelligence Scale for Children - WISC-III), and neurophysiological (ANT-related potentials) data.Methods Twenty typically developing (TD) boys and 19 boys diagnosed with ADHD according to DSM-IV-TR, aged 10-13 years, were examined using the ANT with simultaneous recording of the respective event-related potentials (ERPs). They also performed the Block Design, Digit Spam, Vocabulary and Arithmetic subtests of the WISC-III. A total of 815 variables of interest (VOI) obtained from the ANT, WISC-III scores, and ERP parameters were grouped by hierarchical clustering and integrated in 2 to 6 resultant vectors (RVs), from clusters at hierarchical levels 1 to 5. These RVs were used for the reclassification of subjects using the k-means method.Results Regarding DSM-IV-TR diagnostics, the RVs from behavioral and psychological data and ERPs from the mid-frontal, mid-parietal plus right frontal, right central, and right temporal channels showed accuracy rates from 0.64 to 0.82 using the k-means reclassification. Among reclassifications with higher agreement (0.82), six subjects were reclassified (4 from the TD group, and 2 from the ADHD group). Assuming the reclassification of these six subjects, the estimated agreement between DSM and biological data was 84.61% with kappa index of 0.69.Conclusion Results suggest biological validity and efficiency of DSM as a tool for ADHD diagnostics.