Minimal Important Changes in Accelerometer-Measured Physical Activity and Sedentary Time in Older Patients With Subacute Stroke: A Multicenter Study

Objective: To estimate the minimal important change (MIC) in the amount of time spent at different physical activity intensities (sedentary behavior [SB], light-intensity physical activity [LPA], and moderate-to-vigorous intensity physical activity [MVPA]) among older patients with subacute stroke undergoing inpatient rehabilitation, in the overall cohort and in subgroups stratified by walking ability. Methods: This prospective longitudinal observational study enrolled 299 subacute stroke patients from 12 rehabilitation units. SB, LPA, and MVPA were measured using an accelerometer. The average daily time spent in each intensity and Functional Independence Measure (FIM) and FIM-Motor subscale (FIM-M) scores were assessed at admission and 30 days later. Anchor-based improvement was defined as a ≥22-point increase in FIM or ≥17-point increase in FIM-M. Patients were stratified by Functional Ambulation Categories (FAC) as assisted walking (FAC ≤ 2) or unassisted walking (FAC ≥ 3). MICs were estimated using adjusted predictive modeling when the Pearson correlation between physical activity changes and anchor outcomes was ≥0.3. Results: A sufficient correlation was observed only in patients with FAC ≤ 2; thus, MICs were estimated only in this subgroup. MICs based on FIM were −42.1 minutes for SB, 44.4 minutes for LPA, and 2.9 minutes for MVPA. MICs based on FIM-M were −40.7 minutes for SB, 41.5 minutes for LPA, and 2.6 minutes for MVPA. Conclusion: MICs for time spent at different physical activity intensities were identified only in older patients with subacute stroke who require walking assistance. These findings provide clinically meaningful reference values for interpreting changes in physical activity during rehabilitation.