RESUMO
OBJECTIVE: To explore the relationship between walking measurements (i.e., walking speed, walking performance and walking confidence), and participation in ambulatory people with chronic stroke. MATERIALS AND METHODS: Participation was measured using the mobility domain of Brazilian version of the Stroke Impact Scale 3.0. Walking measures were walking speed, assessed by 10-m Walk Test, walking performance, assessed by ABILOCO, walking confidence, assessed by mGES. Pearson correlation coefficients were used to explore the relationships between the walking measures and social participation, and step-wise multiple linear regression analysis was used to identify which walking measures would explain participation after stroke. RESULTS: Ninety-five chronic stroke individuals (38 men), with a mean age of 67 (SD 13) years were assessed. Significant positive correlations, of high magnitude, were found between participation and all walking measures (r ≥ .53; p < .001). Regarding the regression analysis, walking confidence alone explained 44% (F = 72.4; p < .001) of the variance in participation. When perceived locomotion ability was included in the model, the explained variance increased to 48% (F = 42.8; p < .001). CONCLUSION: All walking measures were correlated with social participation after stroke, but only perceived locomotion ability and walking confidence explained the variance in participation. Clinicians should be encouraged to evaluate real-life performance and personal factors that may limit community participation after stroke.