As people age, declines in physiological functions can lead to changes in sleep and dietary patterns, potentially influencing cognitive function. However, longitudinal evidence examining how food consumption mediates the relationship between sleep parameters and cognitive function in Chinese older adults remains limited. Examining these longitudinal trajectories, beyond single-timepoint measurements, is crucial for understanding their dynamic nature and impact on cognitive function. Therefore, this study aimed to assess the longitudinal trajectories of sleep and food consumption in older adults, and to examine the mediating role of food consumption patterns in the sleep-cognition relationship. The sample population included 8724 older adults who derived from waves 4 to 7 (2005 to 2014) in the Chinese Longitudinal Healthy Longevity Survey (CLHLS). Self-reported sleep duration, sleep quality, and food consumption were evaluated in face-to-face interviews. The main outcomes were global cognitive composite z scores. Group-based trajectory modeling (GBTM) identified distinct long-term change patterns of sleep duration, sleep quality and food consumption, and generalized estimating equations (GEE) evaluated their prospective associations with global cognitive scores. Additionally, structural equation modeling (SEM) was performed to assess the mediating role of food consumption in the relationships between sleep duration, sleep quality, and cognitive function. Food consumption mediated the associations between sleep duration, sleep quality, and cognitive function in older adults. After adjusting for confounders, decreasing low fruit and vegetable consumption mediated a negative impact on cognition, with both shortened (0.024, 95%CI: 0.003 to 0.048) and excessive sleep duration (0.076, 95%CI: 0.040 to 0.117) linked to decreasing low fruit and vegetable consumption. A rapid decline in meat, fish, egg, and bean consumption also mediated the negative effect of shorter sleep duration on cognition (mediation effects: -0.009, 95%CI: -0.014 to -0.005), whereas increased milk consumption acted as a complete mediator, positively influencing cognition. No mediating effect was observed for nuts consumption. Furthermore, stable, high-quality sleep promoted increased consumption of fruit and vegetable (0.039, 95%CI: 0.020 to 0.058), meat, fish, egg, and bean (-0.011, 95%CI: -0.133 to -0.088), milk (0.022, 95%CI: 0.001 to 0.043), and nuts (0.043, 95%CI: 0.021 to 0.065), all positively associated with cognitive levels. This study illustrates that food consumption can improve the decline in cognitive levels in older adults that may be undermined by unhealthy sleep habits. Our research findings are of great importance to both clinical practice and public health policy.
Wu et al. (2026) studied this question.