BACKGROUND: The Manual Ability Classification System (MACS) has been widely used to describe the manual ability of children with cerebral palsy (CP); however its reliability has not been verified in Brazil. OBJECTIVE: To establish the inter- and intra-rater reliability of the Portuguese-Brazil version of the MACS by comparing the classifications given by therapists and parents of children with CP. METHOD: Data were obtained from 90 children with CP between the ages of 4 and 18 years, who were treated at the neurology and rehabilitation clinics of a Brazilian hospital. Therapists (an occupational therapist and a student) classified manual ability (MACS) through direct observation and information provided by parents. Therapists and parents used the Portuguese-Brazil version of the MACS. Intra- and inter-rater reliability was obtained using unweighted Kappa coefficient (k) and intra-class correlation coefficient (ICC). The Chi-square test was used to identify the predominance of disagreements in the classification of parents and therapists. RESULTS: An almost perfect agreement resulted among therapists [K=0.90 (95% CI 0.83-0.97); ICC=0.97 (95%CI 0.96-0.98)], as well as with intra-rater (therapists), with Kappa ranging between 0.83 and 0.95 and ICC between 0.96 and 0.99 for the evaluator with more and less experience in rehabilitation, respectively. The agreement between therapists and parents was fair [K=0.36 (95% CI 0.22-0.50); ICC=0.79 (95% CI 0.70-0.86)]. CONCLUSIONS: The Portuguese version of the MACS is a reliable instrument to be used jointly by parents and therapists.

BACKGROUND: Several studies have demonstrated the importance of using the Gross Motor Function Classification System (GMFCS) to classify gross motor function in children with cerebral palsy, but the reliability of the expanded and revised version has not been examined in Brazil (GMFCS E & R). OBJECTIVE:: To determine the intra- and inter-rater reliability of the Portuguese-Brazil version of the GMFCS E & R applied by therapists and compare to classification provided by parents of children with cerebral palsy. METHOD: Data were obtained from 90 children with cerebral palsy, aged 4 to 18 years old, attending the neurology or rehabilitation service of a Brazilian hospital. Therapists classified the children's motor function using the GMFCS E & R and parents used the Brazilian Portuguese version of the GMFCS Family Report Questionnaire. Intra- and inter-rater reliability was obtained through percentage agreement and Cohen's unweighted Kappa statistics (k). The Chi-square test was used to identify significant differences in the classification of parents and therapists. RESULTS: Almost perfect agreement was reached between the therapists [K=0.90 (95% confidence interval 0.83-0.97)] and intra-raters (therapists) with K=1.00 [95% confidence interval (1.00-1.00)], p<0.001. Agreement between therapists and parents was substantial (k=0.716, confidence interval 0.596-0.836), though parents classify gross motor impairment more severely than therapists (p=0.04). CONCLUSIONS: The Portuguese version of the GMFCS E & R is reliable for use by parents and therapists. Parents tend to classify their children's limitations more severely, because they know their performance in different environments.

CONTEXTUALIZAÇÃO: Em função da complexidade das manifestações clínicas da paralisia cerebral (PC) e das dificuldades na sua classificação baseada apenas nos tipos motores e topografia de distribuição corporal, estudiosos canadenses propuseram o Gross Motor Function Classification System (GMFCS). Apesar de esse sistema de classificação estar sendo bastante utilizado no Brasil, ele ainda não havia sido adapatado transculturalmente para tal. OBJETIVOS: Realizar a adaptação transcultural para o Brasil do Sistema de Classificação da Função Motora Grossa (GMFCS) para a PC e verificar a confiabilidade entre observadores do instrumento adaptado com crianças brasileiras. MÉTODOS: Este estudo consistiu em duas etapas, sendo a primeira relacionada com o processo de adaptação transcultural, e a segunda referente à testagem do instrumento. A adaptação transcultural do instrumento foi feita por meio da tradução, retrotradução, análise semântica, análise de conteúdo, retrotradução da versão final e aprovação dos autores do instrumento. A testagem do instrumento ocorreu junto a 40 crianças com PC, as quais foram avaliadas por dois examinadores para verificar a confiabilidade entre observadores. RESULTADOS: Os resultados demonstram que as etapas de tradução e retrotradução não apresentaram dificuldades, e a equivalência semântica e a conceitual foram obtidas. A confiabilidade entre examinadores demonstrou que as avaliações quase não diferiam e que havia excelente correlação e consistência interna do constructo, com coeficiente de correlação intraclasse (ICC) de 0,945 (com intervalo de 0,861 a 0,979) e a de Cronbach de 0,972. CONCLUSÕES: A versão final do GMFCS mostrou bom potencial de aplicabilidade por graduandos e profissionais da área de neuropediatria.

BACKGROUND: Due to the complexity of clinical symptoms of cerebral palsy and the difficulties in classifying it based upon the motor types and the topography of the body distribution only, Canadian researchers have proposed the Gross Motor Function Classification System (GMFCS). Although this classification system has been largely used in Brazil, it has not been cross culturally adapted yet. OBJECTIVES: To perform the cross adaptation of the Gross Motor Function Classification System for the Cerebral Palsy (GMFCS) into Brazilian-Portuguese and to verify the reliability among observers of the adapted instrument in Brazilian children. METHODS: This study was performed in two stages; the first stage was related to the process of cross-cultural adaptation and the second stage tested the instrument. Translation, back-translation, semantic and content analysis, back-translation of the final version and the approval of the authors were used for the cross-cultural adaptation. The test of the instrument was performed in 40 children with cerebral palsy, who were evaluated by two raters to verify the reliability among the observers. RESULTS: The results showed that the stages of translation and back-translation did not present any difficulties and the semantic and conceptual equivalence was achieved. The reliability among the observers showed that the evaluations do not differ and that there is an excellent correlation and internal consistency of the construct with an ICC of 0.945 (95% CI 0.861 to 0.979) and a Cronbach a of 0.972. CONCLUSIONS: The final version of the GMFCS showed good potential of applicability for undergraduate students and professionals of the neuropediatric area.