JMIR Publications

ABSTRACT

Background:

Accurate measurement of daily physical activity (PA) is important as it links to and/or modifies health outcomes in older adults and persons living with complex health conditions, such as cerebrovascular (CVD) or neurodegenerative disease (NDD). Wrist-worn sensors are widely used to estimate activity intensity in daily life but there is concern such data is unreliable in these cohorts and often confounded by transient activities of daily living or disease-related behaviors that impact arm movement. By contrast, walking is a critical behavior that composes much of daily PA and can be precisely measured using trunk or lower limb sensors.

Objective:

The goals of this work were to 1) determine the accuracy of wrist-based estimates of PA intensity during known walking periods in older adults and persons living with CVD or NDD, and 2) explore factors that influence the variability of wrist-derived intensity estimates.

Methods:

Thirty-five older adults (n=23 with CVD or NDD) wore an accelerometer on the dominant wrist and ankle for 7-to-10-days of continuous monitoring in two related studies. Stepping was detected using the ankle accelerometer. Analyses were restricted to gait bouts ≥ 60 seconds long and with a cadence ≥ 80 steps per min (LONG walks). Wrist accelerometer data were analyzed within LONG walks using 15-second epochs and published intensity thresholds to classify epochs as sedentary, light, or moderate to vigorous intensity physical activity (MVPA). Participants were stratified into quartiles based on percent of walking epochs classified as sedentary and tested for differences in behavioral or demographic traits between the top and bottom quartiles. A case series was performed to illustrate factors and behaviors that can affect intensity estimates.

Results:

Collectively, participants had 107.7 (SD 55.8) LONG walks with a median cadence of 107.3 (SD 10.8) steps per minute, representing 28.2 (SD 13.5) % of total walking time. The proportion of measured intensity (% sedentary and active) differed significantly from the expected distribution of only active epochs for all participants (all P<.001). Across participants, intensity classification was 22.9 (SD 15.8) % sedentary, 27.7 (SD 14.6) % light, and 49.3 (SD 25.5) % MVPA. Participants in the highest quartile of sedentary behavior during LONG walks were significantly older (t16=4.24, P<.001) and had more variable wrist movement (accelerometer vector magnitude coefficient of variation) (t16=2.13, P=.049) compared to the lowest quartile.

Conclusions:

The current best practice wrist accelerometer method is prone to misclassifying activity intensity estimates during walking in older adults and persons living with complex health conditions. A multi-device approach may be warranted to advance methods for accurately assessing PA in these groups.