Independent and Joint Associations of Diet Quality and Rest–Activity Rhythm Phenotypes with Frailty Among Rural Older Adults

Objectives: This study aimed to identify rest–activity rhythm (RAR) phenotypes and examine the independent and joint associations of RAR phenotypes and diet quality with frailty among rural older adults. Methods: This cross-sectional study included 1203 rural Chinese adults aged 65 to 85 years. Accelerometer-derived inter-daily stability, intra-daily variability, and relative amplitude were used to identify RAR phenotypes through latent profile analysis. Diet quality was assessed using the China Prime Diet Quality Score, and frailty was assessed using the Fried Frailty Phenotype. RAR phenotypes and median-dichotomized diet quality were combined into joint groups. Robust linear and ordered logistic regression examined associations with frailty score and severity, respectively. Subgroup and interaction analyses were conducted by sex and age group. Results: Five RAR phenotypes were identified. Compared with the robust-rhythm phenotype, the severely disrupted phenotype was associated with a 0.516-point higher frailty score and 148.6% higher odds of greater frailty severity, while the fragmented low-amplitude phenotype was associated with 49.3% higher odds. Each 10-point increase in diet quality score was associated with a 0.127-point lower frailty score and 29.3% lower odds of greater frailty severity. Joint analyses showed greater frailty in both typical-rhythm and fragmented low-amplitude phenotypes combined with lower diet quality (β = 0.229, OR = 2.020; β = 0.250, OR = 2.179, respectively), with the highest frailty score observed in the severely disrupted and lower-diet-quality group (β = 0.600; OR = 2.774). Severely disrupted combinations were associated with greater frailty whether diet quality was higher (β = 0.512; OR = 3.092) or lower (β = 0.600; OR = 2.774). No evidence of overall interaction by sex or age group was observed. Conclusions: Better diet quality and more robust RAR patterns were independently associated with lower frailty. Considering both factors may provide a broader perspective on behavioral factors related to frailty in older adults.