Introduction Most children/adolescents with disability live in low and middle-income countries and, worldwide, they are more likely to have mental health problems and achieve worse academic performance compared to those with typical development. Objective To assess whether Brazilian children/adolescents with four types of disabilities are more likely to have psychiatric disorders and educational deficits than children/adolescents with typical development. Method A multicenter cross-sectional study involving a school-based probabilistic sample of second to sixth graders (N = 1,674) from public schools in four Brazilian regions. The four types of disabilities (intellectual, visual, hearing, and motor) were assessed using the Ten Questions Questionnaire. Psychiatric disorders were measured with the Schedule for Affective Disorders/Schizophrenia for School-Age Children (K-SADS-PL), and academic performance was evaluated using the Teste de Desempenho Acadêmico – TDE (the academic performance test). Results A logistic regression model with cluster-robust errors identified the following statistically significant associations with three of the four types of disability (the exception was hearing). Intellectual disability was associated with anxiety (p < 0.01), depression (p < 0.01), attention deficit hyperactivity disorder (ADHD) (p < 0.001), school failure (p < 0.01), and poor academic performance (p < 0.01). Visual disability was associated with depression (p < 0.01). Motor disability was marginally associated with ADHD (p = 0.08). Conclusions Presence of disabilities (intellectual, visual, and motor) in children/adolescents was associated with psychiatric disorders, school failure, and academic performance. It is therefore important to identify presence of disabilities and plan and deliver specific interventions and specialized educational care for the needs presented by these children/adolescents. This is particularly important in low and middle-income countries, where these disabilities are frequent among children/adolescents.

IntroductionAttention Deficit Hyperactivity Disorder (ADHD) is a Neurodevelopmental Disorder characterized by persistent pattern of inattention and hyperactivity / impulsivity. There is considerable difficulty in diagnosing ADHD, mainly to discriminate what could be symptoms arising from ADHD or typical age behaviors. The decision tree model is a statistical algorithm, a predictive model built with comparisons of values for a given objective that can be compared with other constant values, placing these variables in a database at hierarchical levels.ObjectivesThis study aims to apply the decision tree model in directing the screening of ADHD complaints to analyze which cognitive and behavioral parameters would be better associations with ADHD accurate diagnosisMethodsWe used a database of research protocol with 202 children assessed with complaints of ADHD and a control group with 185 participants. Decision tree analyzed parameters selected from the cognitive instruments, such voluntary attention, Continuous Performance Test indexes, WCST indexes, Wechsler Intelligence indexes and behavioral scales from CBCL/6-1 and TRF/6-18.ResultsThe highlighted results points to WCST index like: “Perseverative answers” and “Perseverative errors” and “learning to learn” joint to “CPT omissions” and behavioral scales as “CBCL ADHD”, and “CBCL Problems of Attention” produces accuracy of diagnosis discrimination from 84.7% to 60% in the precision of the decision tree.ConclusionsThe decision tree and machine learning approaches can be effective in directing the screening of typical ADHD complaints.DisclosureNo significant relationships.