To investigate the association between night-time sleep duration and new-onset frailty among middle-aged and elderly individuals in China, and to examine the mediating effect of depression levels on this relationship, thereby providing evidence for the prevention of frailty in this population.

Using longitudinal data from the China Health and Retirement Longitudinal Study (CHARLS) surveys in 2011, 2013, and 2015, we employed logistic regression models to analyze the relationship between night-time sleep duration, depression levels, and frailty among middle-aged and older adults. Restricted cubic splines were utilized to examine the non-linear association between night-time sleep duration and frailty. A causal mediation model was employed to examine the mediating effect of depression severity between night-time sleep duration and new-onset frailty.

Among 7 582 participants, the frailty prevalence was 21.2%. For each additional hour of night-time sleep duration, depression scores decreased by 0.36 points, with an OR (95% CI) for frailty incidence of 0.92 (0.89-0.96). Conversely, for each 1-point increase in depression severity, the OR (95% CI) for frailty incidence was 1.08 (1.07-1.10). A significant non-linear relationship existed between night-time sleep duration and new-onset frailty (P=0.004). Depression mediated the association between night-time sleep duration and new-onset frailty, with an OR (95% CI) of 0.972 (0.966-0.979), accounting for 35.2% of the mediated effect.

Shortened night-time sleep duration increases the likelihood of new-onset frailty among middle-aged and elderly Chinese adults, with depressive symptoms partially mediating this association.