Association between living arrangements in early old age and subsequent trajectories of protein intake among older adults in China

BACKGROUND: Although most studies have emphasised the influence of living arrangements on nutritional intake, they have typically been limited to assessing associations within the same timeframe. How living arrangements in early old age influence the longitudinal development of protein intake with advancing age remains insufficiently understood. METHODS: This study used data from the China Health and Nutrition Survey (CHNS) 1997-2011 and included 1,162 participants aged 65-79 years. Group-based trajectory modelling (GBTM) was applied to identify distinct trajectories of protein intake. Latent Markov modelling (LMM) was then used to estimate the sequence of living arrangement states for each participant, from which living arrangement patterns in early old age were derived. Multinomial logistic regression and inverse probability weighting (IPW) were employed to examine the association between early-old-age living arrangements and subsequent protein intake trajectories. RESULTS: GBTM identified three trajectories of daily protein intake: a low trajectory, a medium trajectory, and a high trajectory. LMM revealed three living arrangement patterns in early old age: independent living, spousal co-residence, and multigenerational co-residence. After applying IPW, compared with individuals living alone, spousal co-residence was associated with 1.95 (Coefficient = 0.67, 95% CI: 0.24-1.12) times the relative odds, and multigenerational households were associated with 1.62 (Coefficient = 0.48, 95% CI: 0.07-0.93) times the relative odds, of following the high trajectory relative to the low trajectory. In contrast, no significant associations were observed between these co-residence patterns and the medium trajectory. CONCLUSION: This study identifies significant correlations between initial living arrangements and subsequent protein intake trajectories, highlighting that spousal co-residence is consistently linked to higher and more stable protein intake patterns.