The Dynamic Gait Index (DGI) is a useful test to evaluate balance and gait. AIMS: The objectives of this study were to culturally adjust the DGI to the Portuguese language and to assess its reliability. METHODS: The method proposed by Guillemin et al. (1993) was used for a cultural adaptation of this tool. A prospective study was performed with 46 patients that were assessed in the cultural adaptation phase. The items that not understood by 20% or more patients were reworded and reapplied. The final Portuguese version of DGI was applied to 35 elderly in order to check intra and interobserver reliability. The Spearman rank coefficient was used to correlate intra and interobserver scores and the Wilcoxon test was applied to compare these scores. Internal consistency was analyzed by the Cronbach alpha coefficient. RESULTS: There were statistically significant correlations among the scores for intra and interobserver assessments for all items (p<0.001), which were classified as good and very strong correlations (ranging from r=0.655 to r=0.951). The DGI demonstrated high internal consistency in intra and interobserver assessments (varying from <FONT FACE=Symbol>µ</FONT > or = 0.820 to <FONT FACE=Symbol>µ</FONT > or = 0.894). CONCLUSION: The DGI was culturally adjusted to Brazilian Portuguese and proved to be a reliable tool.
equilibrium; rehabilitation outcome assessment; reproducibility of results; dizziness