JMIR Publications

ABSTRACT

Background:

Older outpatients with chronic diseases are susceptible to frailty and functional disabilities. Recently, various wearable devices have been developed to generate health logs in daily life; however, data from patients with chronic diseases remain scarce. Previous studies have shown that subjective sleep efficiency assessed by self-questionnaires are associated with grip strength, but whether objective sleep efficiency detected by wearable device is associated with physical function and body composition remained unclear.

Objective:

To clarify the association between sleep duration, sleep efficiency, grip strength, frailty, and sarcopenia in outpatients with chronic diseases and the influence of diabetes, in the Smart Watch Innovation for Geriatrics and Gerontology (SWING-Japan) project.

Methods:

In this cross-sectional study, the wrist-worn device Silmee 22 was used in an acute care geriatric hospital between April 2023 and December 2024. Data from patients who wore the device for 7¬–90 nights were used for the analyses, and individual average sleep duration and efficiency (total sleep duration/time spent in bed) were calculated. Low grip strength was defined as <28 kg for men and <18 kg for women. Sarcopenia was defined based on the Asiancriteria and frailty based on the Japanese version of Cardiovascular Health Study and Kihon Checklist criteria. Sleep indices were compared between individuals with and without these conditions. Next, logistic regression analysis was performed with lower grip strength as the dependent variable and sleep efficiency as the independent variable, adjusted for age, sex, body mass index, glycohemoglobin, cognitive function (Mini-Mental State Examination), Geriatric Depression Scale-15 score, and daily step count measured using the ankle-worn device, WalkX. Subgroup analyses were performed and stratified according to the status of diabetes mellitus.

Results:

Data from 165 patients (median age, 77 years; duration of device use, 65 days; diabetes prevalence, 55%) were used for analyses. Sleep efficiency, but not sleep duration, was significantly lower in patients with low grip strength than in those without (58.0% [52.7–64.0%] vs 62.7% [55.8–78.7%], P=0.006). Sleep efficiency was not associated with frailty, low muscle mass, or slow walking speed. After adjustment with the above variables, low sleep efficiency was associated with the risk of lower grip strength (OR =1.04 (95%CI 1.01–1.08, P=0.02) by 1%-point decrease in sleep efficiency). When stratifying patients by diabetic status, the association was stronger in patients with diabetes (OR =1.05 (95%CI 1.002–1.093, P=0.039, p for interaction =0.09).

Conclusions:

In older patients with chronic diseases, objective low sleep efficiency evaluated with wrist-worn wearable devices was associated with low muscle strength, with a stronger association in patients with diabetes mellitus. Sleep efficiency may potentially be a new target to prevent the deterioration of muscle strength, and a wrist-worn device could help assess individual sleep status in this population.