Abstract
The objective of this study was to identify the minimum wear time criteria to produce reliable estimates of 24-hour movement behaviors and compositions in a study of older women. We used data collected in 2015-2017 in a subsample (n = 1282) of The Study of Women's Health Across the Nation (SWAN). Data from hip-worn accelerometers and wrist-worn actigraphy devices were harmonized into sleep-wake activity cycles. We identified minimum sleep-wake interval duration for 24-hour-equivalent estimates (reference [23, 25] hour range) and minimum number of sleep-wake cycles for 7-day-equivalent estimates using tests of equivalence with a confidence approach for duration and percent of the sleep-wake interval for each behavior. For intervals in the 23- to 25-hour range, participants spent a mean 7.8 hours (32.8%) in sleep, 7.4 hours (31.2%) in sedentary behavior, 4.9 hours (20.7%) in low light intensity physical activity (PA), 2.6 hours (10.8%) in high light intensity PA, and 1.1 hours (4.5%) in moderate/vigorous intensity PA. The minimum wear time criteria to produce reliable estimates was at least 1 day with a minimum of 6 hours of data. Our results are likely driven by high adherence to device wear protocols and are likely applicable to other studies of midlife and older adults.