USE OF ATTENTION DEMANDING TASKS TO ESTIMATE FALL RISK AND DAILY WALKING ACTIVITY IN SUBACUTE STROKE
Introduction: Community-dwelling stroke survivors are extremely inactive, leading to a possible cascade of secondary health problems such as obesity and depression and risk for second stroke and mortality. Falls are also a common and serious adverse event, especially prevalent in the first few months after hospital discharge. Customary clinical gait assessments do not adequately predict falls or inactivity after hospital discharge. Objective: To examine whether community-relevant mobility tasks (obstacle crossing and dual-task walking) assessed at hospital discharge can predict fall risk and daily walking activity, respectively, in the first 3 months post discharge. Participants: 47 participants with a mean age of 59.8 (SD=11.7) years and median of 14 (IQR 7-21) days post stroke who were being discharged home from acute care or acute inpatient rehabilitation. Methods: In addition to traditional clinical assessments, participants completed 2 single-task and 2 dual-task walking trials, followed by 4 obstacle-crossing trials (height: 10% of leg length) at their self-selected walking speed, all over an instrumented walkway. Following discharge, participants completed a prospective falls diary and at 3 months post discharge wore an activity monitor to assess daily walking activity. Binary logistic regressions were used to examine the relationship between pre-discharge obstacle-crossing performance and fall status at 3-months post discharge and pre-discharge dual-task gait speed at discharge and walking activity at 3 months post discharge. Results: Participants who failed the obstacle-crossing task at hospital discharge had a significantly higher odds of falling (OR=10.00, 95% CI 2.45, 40.78) in the first 3 months post discharge than those who passed the obstacle-crossing task. A 1-unit change in dual-task gait (DTG) speed resulted in significantly higher odds (OR=23.25, CI 2.46, 219.38) of being a community ambulator at 3 months post hospital discharge. Therefore, for a 0.10 m/s increase in DTG speed, we would expect a 37% increase in the odds of being a community ambulator. Conclusions: Obstacle-crossing appears to be a useful discharge risk-assessment tool for identifying future fallers. The clinical significance of dual-task walking in predicting ambulatory activity after discharge is unclear due to imprecision of the effect estimate, but the present findings suggest that further investigation is warranted.
