Reliability of the Dynamic Gait Index (Finnish version) in individuals with neurological disorders

Purpose: To produce a Finnish version of the Dynamic Gait Index (DGI) and establish its reliability in people with neurological disorders affecting balance. Method: A consecutive sample of 33 participants in inter-rater and 30 participants in intra-rater sample were tested. The DGI was administered in two testing sessions 1-2 h apart. Reliability was assessed with the intra-class correlation coefficient (ICC), Bland-Altman 95% limits of agreement, coefficient of reproducibility (CR), and standard error of the mean (SEM). Results: The instructions from the original DGI were preserved during the translation process with no need for cultural adaptations. The relative reliability of the total scores proved to be high in inter-rater (ICC = 0.90) and intra-rater (ICC = 0.91) testing. The relative reliability between single items varied from poor to high in inter-rater (r = 0.57-0.96) and intra-rater (r = 0.20-0.97) testing. Absolute reliability of the total score was high, and in single items, it varied similarly to relative reliability. The SEM was 1.2 points in inter-rater and 0.8 points in intra-rater testing. Conclusions: The Finnish version of DGI was produced and found to be a reliable tool for clinical testing of balance in neurological patients with balance disorder.