Associations Between Accelerometer-Based Free-Living Walking and Self-Reported Walking Capability Among Community-Dwelling Older People.

The authors examined whether accelerometer-based free-living walking differs between those reporting walking modifications or perceiving walking difficulty versus those with no difficulty. Community-dwelling 75-, 80-, or 85-year-old people (N = 479) wore accelerometers continuously for 3-7 days, and reported whether they perceived no difficulties, used walking modifications, or perceived difficulties walking 2 km. Daily walking minutes, walking bouts, walking bout intensity and duration, and activity fragmentation were calculated from accelerometer recordings, and cut points for increased risk for perceiving walking difficulties were calculated using receiver operating characteristic analysis. The authors' analyses showed that accumulating ≤83.1 daily walking minutes and walking bouts duration ≤47.8 s increased the likelihood of reporting walking modifications and difficulties. Accumulating walking bouts ≤99.4 per day, having walking bouts ≤0.119 g intensity, and ≥0.257 active to sedentary transition probability fragmented activity pattern were associated only with perceiving walking difficulties. The findings suggest that older people's accelerometer-based free-living walking reflects their self-reported walking capability.