V. Souza Santos1, S. J Kamper2, B. Aparecida3, M. Nascimento Leite1, B. T Saragiotto1,4, L. O P Costa1, T. P Yamato1,4
1Universidade Cidade de São Paulo, Master’s and Doctoral Program in Physical Therapy, São Paulo, Brazil, 2The University of Sydney, School of Health Sciences, Faculty of Medicine and Health, Sydney, Australia, 3Universidade Cruzeiro do Sul, Interdisciplinary Master and Doctoral in Health Sciences, São Paulo, Brazil, 4The University of Sydney, Institute for Musculoskeletal Health, School of Public Health, Faculty of Medicine and Health, Sydney, Australia
Background: Musculoskeletal pain is common in children and adolescents with prevalence range of 4-40%. Psychosomatic symptoms seem to influence both the onset and the development of pain. However, there are just few validated questionnaires for the children and adolescents population.
Purpose: To translate, cross-culturally adapt the Psychosomatic Questionnaire for Children and Adolescents into Brazilian-Portuguese and test interpretability and its measurement properties.
Methods: We conducted a clinical measurement study. This study was approved by Human Ethics Committee of Universidade Cidade de São Paulo (UNICID) (CAAE: 18752219.0000.0064). We recruited 107 Brazilian children and adolescents with musculoskeletal pain from public and private schools in Sao Paulo-Brazil. The questionnaire was completed twice with a 7-day interval. We tested the interpretability using celling and floor effects, internal consistency using Cronbach’s Alpha (adequate between 0.70 – 0.95) and reliability using Intraclass Correlation Coefficient2,1 (interpreted as <0.40 – low reliability; 0.40-0.75 – moderate reliability; 0.75-0.90 – substantial reliability; e >0.90 – excellent reliability). The agreement (measurement error) was measure through Standard Error of Measurement (SEM) that is calculated by formula SEM= SD√1-ICC and through Smallest Detectable Change (SDC) that was measured with the formula SDC= 1.96√2.SEM. Construct validity was measure through Pearson correlation (r) between final scores of psychosomatic questionnaire and the Spence Children’s Anxiety Scale (SCAS-Brazil). Our hypothesis was both questionnaires would have a positive, moderate to strong correlation; r ≥ 0.30.
Results: Fifty-eight percent of the participants were girls and the mean age was 11 years (SD 2.13). The questionnaire showed adequate interpretability due to the absence of ceiling and floor effects. Internal consistency measured by the Cronbach’s Alpha was 0.69. Reliability measured by the Intraclass Correlation Coefficient2,1 was 0.88 (95% CI 0.82 to 0.92). Smallest Detectable Change was 1.63 points out of 18 points. Agreement (measurement error) was 0.99 (5.50 % related final score). Finally, we observed a moderate correlation with the Spence Children’s Anxiety Scale (SCAS-Brazil), in line with our a-priori hypothesis.
Conclusion(s): The Brazilian-Portuguese version of the Psychosomatic Questionnaire for Children and Adolescents has acceptable measurement properties and it is a good option for assessing psychosomatic symptoms in clinical practice and research. Future works must investigate this questionnaire in another health conditions to understand if this questionnaire will have the same measurement properties.
Implications: The Brazilian-Portuguese Version of the Psychosomatic Questionnaire for Children and Adolescents can be used in new research and in the clinical practice involving children and adolescents that speak Portuguese.
Funding, acknowledgements: We would like to thank São Paulo Research Foundation (FAPESP-Brazil) for all financial support to conduct this study.
Keywords: Measurement properties, Psychosomatic symptoms, Musculoskeletal pain
Topic: Paediatrics
Did this work require ethics approval? Yes
Institution: Universidade Cidade de São Paulo (UNICID)
Committee: Human Ethics Committee of Universidade Cidade de São Paulo (UNICID)
Ethics number: CAAE: 18752219.0000.0064
All authors, affiliations and abstracts have been published as submitted.
