RÉSUMÉ
BACKGROUND: This study was designed to translate, culturally adapt, and validate a Brazilian Portuguese version of the Short Musculoskeletal Function Assessment Questionnaire (SMFA-BR). METHODS: The SMFA was translated from English into Brazilian Portuguese. Translations were synthesized, translated back into English, and then submitted to a committee of clinical, psychometric, and language experts. The questionnaire was then administered to 220 trauma patients at a midsize hospital in southern Brazil. Test-retest reliability was examined at one and seven days. Scale reliability and validity were assessed, and factor structure was analyzed. RESULTS: Patients with only one region of dysfunction reported less dysfunction than did patients with two or three regions of dysfunction. Both the SMFA-BR dysfunction and the SMFA-BR bother scores had significant correlations with all Brazilian Short Form-36 (SF-36) subscales except for bodily pain. Test-retest reliabilities, as determined by intraclass correlation analyses, were 0.99 (95% confidence interval, 0.97, 1.00) at one day (n = 10) and 0.99 (95% confidence interval, 0.96, 1.00) at seven days (n = 17) for the dysfunction index and 0.99 (95% confidence interval, 0.98, 1.00) at one day (n = 10) and 0.97 (95% confidence interval, 0.97, 1.00) at seven days (n = 17) for the bother index. Cronbach alpha reliabilities were 0.95 (95% confidence interval, 0.93, 0.97) and 0.91 (95% confidence interval, 0.89, 0.94) for the dysfunction and bother indices, respectively. Minimal ceiling and floor effects were observed for the bother subscale only. Exploratory factor analysis yielded a three-factor structure: upper-extremity dysfunction, lower-extremity dysfunction, and bother. CONCLUSIONS: A reliable and valid Brazilian Portuguese version of the SMFA was developed. It will facilitate the examination of functional performance within a large patient population as well as allow cross-cultural comparisons.