Research Square · Nov 18, 2026
Joint profiles of remnant cholesterol and the uric acid-to-HDL cholesterol ratio in transitions across cardiovascular-diabetes-kidney multimorbidity and death
By Wang Y, Wang H, Li Y et al.
Background The accumulation of cardiovascular, diabetes, and kidney conditions is a dynamic public-health process. Remnant cholesterol (RC) and the uric acid-to-high-density lipoprotein cholesterol ratio (UHR) are readily available metabolic markers, yet little is known about how their combined levels relate to progression through successive multimorbidity states. Methods We included 10,560 CHARLS participants with valid baseline RC, UHR, age, observed disease state, and linked mortality status. Four joint RC/UHR profiles were examined using a covariate-adjusted discrete-time Markov model to estimate destination-specific relative risk ratios across the multimorbidity pathway. Benjamini-Hochberg adjustment was applied to the 27 prespecified profile tests. Continuous exposure, RC-quality, and cut-point analyses assessed robustness. A separate HRS analysis was placed in the Supplement as supportive external context because its biomarkers and disease-state definitions were not directly equivalent to CHARLS. Results Across 36,210 observed interval pairs from 10,343 participants, 1,269 deaths were flagged by 2020. Compared with the low RC/low UHR profile, the high RC/high UHR profile was associated with transition from no disease domain to one domain (RRR 1.245, 95% CI 1.101–1.407; BH q = 0.006) and from one domain to multimorbidity (RRR 1.437, 95% CI 1.184–1.745; BH q = 0.006). The estimate for one domain to death was higher but did not remain below 0.05 after multiplicity adjustment (RRR 1.518, 95% CI 1.079–2.134; nominal P = 0.017; BH q = 0.114). Excluding 190 participants with negative calculated RC values produced similar estimates for the two main transitions. Median-cutpoint analyses showed the same directions but were not significant after BH adjustment (q = 0.140 and q = 0.207). The continuous RC-by-UHR interaction was not clearly associated with progression. Conclusion In this longitudinal cohort, the joint high RC/high UHR profile identified participants more likely to enter a first cardiovascular, diabetes, or kidney disease domain and then progress to multimorbidity. The study demonstrates the value of a stage-specific multi-state framework for revealing heterogeneity that is hidden by a single multimorbidity endpoint. The findings are intended to guide further prevention and replication research, not to define causal effects or clinical thresholds.