Análise da confiabilidade e validade da versão brasileira do Little Developmental Coordination Disorder Questionnaire (LDCDQ-BR) para crianças de 3 e 4 anos de idade

Abstract

Good motor coordination is essential for children to be able to perform daily activities at home, school, and social settings. However, some children, even having sufficient opportunities to acquire these skills, have marked impairment in the development of motor coordination, as expected for the age group, without any relationship with neurological and/or physical disorders, thus characterizing the Developmental Coordination Disorder (DCD). The early identification of DCD makes it possible to carry out orientations or interventions focused on promoting the development of motor skills important for the child's development and improving participation in different contexts. Given the relevance of detecting DCD in younger children and the lack of adequate instruments, occupational therapists from Israel developed the Little Developmental Coordination Disorder Questionnaire (Little DCDQ), a parental questionnaire focused on children aged 3 and 4 years. This questionnaire was translated and adapted to Brazilian Portuguese (LDCDQ-BR) and the present study aimed to investigate the validity, reliability and define preliminary cut-off points for the questionnaire. Eight hundred and four envelopes were sent to parents, containing the following instruments: LDCDQ-BR, Strength and Difficulties Questionnaire (SDQ), a demographic questionnaire, and the Brazil Economic Classification Criterion. Three hundred twelve (38.8%) of the questionnaires sent returned correctly, seventy-seven parents answered the questionnaire twice for the teste retest reliability, and among the whole sample, 119 children were assessed with the Movement Assessment Battery for Children - 2nd Edition (MABC-2) for concurrent validity. Participated in the study 158 (48.8%) boys and 154 girls aged 3 (n = 154) and 4 (n = 158), being 155 from public schools. There was no significant difference in total scores related to age, sex, gender, and frequency of extra motor activity. In the subcategories, only regarding gender, there was a difference in the control during the movement (p = 0.007), in which boys performed better than girls. Regarding type of school, public school children had lower total scores (p = 0.001) and in the all subcategories of the LDCDQ-BR (Mann-Whitney test). Rasch analysis showed adequate mean values of Infit (1.04, t=0.43) and Outfit (0.98, t=0.00), indicating good functioning of the items with only one item identified as erratic, suggesting unidimensionality of the questionnaire. The reliability of the items (0.97) was excellent, and average (0.72) for children, with separation in two levels of ability, since the questionnaire proved to be easy for the participants, as 9.9% of the children obtained maximum score. The principal component analysis suggested the possibility of a second dimension, but reanalysis with two sub-categories resulted in a reduction in the reliability of people's measurements, as well as reanalysis considering the three sub-categories of the questionnaire. The internal consistency of the LDCDQ-BR was excellent (0.91), and excluding any item has little effect on this value, which supports the use of the total score, based on the 15 items. Test-retest reliability (CCI) of the total score was good (0.77) and the items ranged from 0.44 to 0.78. The LDCDQ-BR score had a weak but significant correlation with the MABC-2 percentiles (r = 0.301, p <0.01). ROC curve, estimated the cut-off point of 64 points (J = 0.352), AUC=0.712 (p <0.05), sensitivity of 68% and specificity of 67%. The LDCDQ-BR has acceptable measurement properties, it is important to investigate its usefulness in the early identification of motor coordination difficulties. Future studies with a sample involving a greater number of children with motor difficulties are necessary to check discrimination and obtain more accurate cut-off points.