Valérie Poulin and Johanne Desrosiers
DOI: 10.2340/16501977-0115

Objective: This study aimed: (i) to determine the level of agreement between responses of people with stroke and their proxies on a participation questionnaire, and (ii) to explore patient and proxy characteristics related to disagreement between their responses.
Design: Cross-sectional study.
Subjects: A total of 40 community-dwelling people with stroke and their proxies (total n = 80).
Methods: Participants were evaluated separately in face-to-face interviews using the Assessment of Life Habits questionnaire (LIFE-H 3. 1), which documents participation in daily activities and social roles. Sociodemographic and clinical variables, such as severity of motor impairment and cognitive functions, were also collected.
Results: Moderate to excellent agreement was found for the LIFE-H total score (intraclass correlation coefficient (ICC): 0. 82), daily activities (ICC: 0. 87) and social roles (ICC: 0. 73) sub-scores. Proxies perceived more disruptions in participation than patients in several life domains (p = 0. 035–0. 001), but the differences between their scores were generally not clinically meaningful. Severity of motor disabilities was the best predictor of disagreement between patients and proxies.
Conclusion: These acceptable levels of agreement support the use of proxy responses to estimate the participation of people unable to respond themselves. However, some characteristics contributing to disagreement between respondents, such as severity of motor disabilities, should be considered when interpreting proxy information.