Article

Purpose: To develop a telephone version of the Chedoke-McMaster Stroke Assessment
Activity Inventory (CMSA-AI) and estimate the test-retest reliability, interrater
reliability (between participant and proxy), and construct validity of the scores
for individuals with stroke. Methods: Adults with stroke and their caregivers or
proxies were included. Participants were assessed with the CMSA-AI at discharge
from a stroke rehabilitation unit and interviewed using the telephone version
(TCMSA-AI). Two months after discharge, participants were evaluated with the
CMSA-AI and interviewed over the phone using the TCMSA-AI on two occasions 2-3
days apart. Proxies were interviewed with the TCMSA-AI within another 2-3 days.
Results: The mean age of the 53 participants with stroke was 62 years; 59% were
male; 43% had right-side hemiparesis; 42 completed follow-up interviews; and 18
had proxies who also participated. Test-retest reliability showed an intra-class
correlation coefficient of 0.98 (95% CI: 0.96, 0.99) for the total score, 0.96
(95% CI: 0.91, 0.98) for the Gross Motor Function Index, and 0.96 (95% CI: 0.91,
0.98) for the Walking Index, and an interrater reliability (between participant
and proxy) of 0.75 (95% CI: 0.28, 0.90) for total score. Spearman's rho
correlation between CMSA-AI and TCMSA-AI total scores was 0.62 (lower-sided 95%
CI: 0.42) at discharge and 0.90 (lower-sided 95% CI: 0.82) at 2 months after
discharge. Correlations between the change scores of the CMSA-AI and TCMSA-AI
were 0.50 or lower. Conclusion: There is potential for remote evaluation of the
functional mobility of individuals with stroke in research and clinical settings.

Langue : ANGLAIS