RESUMO
Background : While the research design of articles published in medical journals and in some physical therapy journals has already been evaluated, this has not been investigated in Brazilian physical therapy journals. Objective : To describe the research design used in all articles published in Brazilian scientific journals that are freely available, have high Qualis rankings, and are relevant to physical therapy over a 7-year period. Method : We extracted the bibliometric data, research design, research type (human or animal), and clinical area for all articles published. The articles were grouped into their level of evidence, and descriptive analyses were performed. We calculated the frequency, proportions of articles, and 95% confidence interval of these proportions with each research design in each journal. We cross-tabulated the clinical areas with research designs (expressed as number and percentages). Results : A total of 1,458 articles from four Brazilian journals were found: Revista Brasileira de Fisioterapia, Revista Fisioterapia em Movimento, Revista Fisioterapia e Pesquisa, and Revista Acta Fisiátrica. The majority of articles were classified as level II of evidence (60%), followed by level III (29%) and level I (10%). The most prevalent research designs were cross-sectional studies (38%), single-case or case-series studies, and narrative reviews. Most articles reported human research and were in the musculoskeletal, neurologic, and cardiothoracic areas. Conclusions : Most of the research published in Brazilian physical therapy journals used levels II and III of evidence. Increasing the publication rate of systematic reviews and randomized controlled trials would provide more high-quality evidence to guide evidence-based physical therapy practice. .