Changes in frailty and incident circadian syndrome in middle-aged and older adults: a nationwide prospective cohort study

Abstract

Circadian syndrome (CircS) is more predictive than metabolic syndrome (MetS). However, few studies have examined its risk factor. This study aims to investigate whether the static state or dynamic change of frailty is correlated with incident CircS in Chinese adults. Data were obtained from the China Health and Retirement Longitudinal Study (CHARLS). Baseline frailty was evaluated by frailty index (FI) and frailty status: frail (FI ≥ 0.25), pre-frail (0.10 < FI < 0.25), and robust (FI ≤ 0.10). Changes in frailty were assessed through changes in frailty status, total FI, delta FI and K-means clustering classifications. CircS was defined as the presence of at least 4 of 7 components (MetS components plus depression and short sleep duration). Restricted cubic splines (RCS) and logistic regression models were utilized to assess the association of baseline frailty or its alterations with CircS incidence after adjusting for potential confounders. A total of 5,042 and 3,527 participants were included in baseline and dynamic analyses, respectively. Compared with stable robust participants, those who transitioned from robust to pre-frail/frail status exhibited increased CircS risk (OR: 1.38, 95%CI: 1.01–1.86, P=0.040). Conversely, pre-frail participants recovering to robust status exhibited reduced risk compared with stable frail participants (OR: 0.59, 95%CI: 0.40–0.85, P=0.006). Higher total FI (highest tertile: OR: 1.43, 95%CI: 1.14–1.80, P=0.002) and sustained high FI (Class 3: OR: 1.47, 95%CI: 1.07–2.00, P=0.016) were associated with elevated CircS risk. Individuals with concurrent high total FI and large increase in FI exhibited even greater risk (Class 3: OR: 1.43, 95%CI: 1.10–1.85, P=0.006). RCS analysis identified a non-linear association between delta FI and CircS incidence. Among middle-aged and older adults, frailty changes are associated with incident CircS risk, whereas baseline frailty is not predictive. Monitoring frailty trajectories may facilitate early identification of individuals at CircS risk.