Association between frailty and subjective and objective sleep indicators in Japanese community-dwelling older adults

Abstract

Studies have linked frailty to sleep duration and/or quality using questionnaire-based subjective sleep assessments. This study clarified the relationship between frailty status and subjective and objective sleep indicators among community-dwelling older adults in a rural Japanese area. This cross-sectional cohort study analyzed the data of older adult participants in the FESTA Study, assessing subjective and objective sleep indicators using the Pittsburgh Sleep Quality Index (PSQI) and an actigraph, respectively. Frailty status was determined using the Japanese version of the Cardiovascular Health Study (J-CHS) and the Kihon Checklist (KCL). Its relationship was examined through multivariate logistic regression analysis. The data of 537 older adults (median age = 76 years; 177 men and 360 women) were analyzed. Aside from age, depression, and dietary variety score, the PSQI score and the number of awakening episodes after sleep onset were significantly associated with non-robustness when the J-CHS was utilized (OR 1.086 95%CI 1.017-1.159 P = .014 and OR .964 95%CI .934-.994 P = .019, respectively). When the KCL was utilized, non-robustness was significantly associated with the PSQI score (OR 1.100 95%CI 1.028-1.178 P = .006), along with age, gender, number of comorbidities, and depression. Among the seven items of the PSQI, non-robustness was significantly related to daytime dysfunction due to sleepiness. The associations between non-robustness and objective sleep indicators varied by the frailty assessment method, while non-robustness was significantly associated with subjective sleep quality, regardless of the diagnostic tool used for frailty status and age. Therefore, subjective sleep quality may be more reliable for preventing and/or managing frailty in older adults.

Keywords: Accelerometer; Actigraph; Frailty; Pittsburgh sleep quality index; Robustness; Sleep indicator.