Depressão infantil: elaboração de um instrumento para avaliação e tomada de decisão em saúde

• Ano de defesa: 2017
• Autor(a) principal: Sá, Lays Andrade de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Ciências Exatas e da Saúde; Programa de Pós-Graduação em Modelos de Decisão e Saúde; UFPB
• Palavras-chave em Português: Depressão; Avaliação psicológica; Validação; Precisão; Teoria de resposta ao Item; Depression; Psychological assessment; Validation; Precision; Item response theory; CIENCIAS DA SAUDE::SAUDE COLETIVA
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/tede/9044

The child depression construct evaluated in the present study was one whose symptoms are associated with Major Depressive Disorder described in DSM-5. The purpose of this dissertation was to elaborate and validate an Inventory for Evaluation of Depression in Children. For this purpose, was applied to the Child Depression Inventory, developed and validated in this study, a sample of 211 children; divided into two groups: one with psychological service demands and another without this demand. The research was exploratory type, with data analysis of quantitative approach, based on the Classical Theory of Tests and the Theory of Response to the Item. Were performed Analyzes which made it possible to verify the psychometric evidence: Factorial Analysis of the main axes (PAF) with oblique rotation and Chronbach alpha indexes. PAF indicated a structure of 58 items with a correlation matrix composed of 7 factors, namely: 1. Self-destructive thoughts; 2. Apathy; 3. Euphoric feelings; 4. Irritability; 5. Feelings of utility; 6. Deep despondency; And 7. Psychomotor restlessness and retardation. These factors presented, in the great majority, a level of internal consistency above 0.70; and for the total set, 0.926. It was used Mann-Whitney U Test and t-test for independent samples in order to compare the two sets of samples. it was notes Significant statistical differences between them In both tests. The TRI analysis was done by the Gradual Response Model for polytomic items. Regarding parameter b (difficulty), of the total of items, only 3 were considered easy; 8 as medium and 16 considered extremely difficult (59.26%). For parameter a, item I11 was the most discriminatory and I29 or less. Factors 1, 4 and 5 were the ones that presented the best performance in providing information about the construct. After a TRI calculation, the instrument resulted in 27 items. Overall, the goal was achieved and the results described were informative. In general, it can be stated that the objectives of this dissertation have been reached.