To investigate the relationship between different uric acid-to-high-density lipoprotein cholesterol ratios (UHR) and new-onset diabetes in middle-aged and older individuals, providing a reference basis for diabetes prevention and treatment in this population.

Using data from the China Health and Retirement Longitudinal Study (CHARLS), we selected middle-aged and older adults who participated in five waves of the survey from 2011 to 2020. Restricted cubic spline (RCS) analysis was employed to examine the potential nonlinear associations of baseline UHR and cumulative UHR with diabetes risk. The UHR trajectory was categorized into three classes using the k-means clustering algorithm: low-level increasing class (Class 1), moderate-level increasing class (Class 2), and high-level decreasing class (Class 3). Cox proportional hazards regression models were then used to analyze the association between different UHR parameters and new-onset diabetes. Further subgroup analyses were conducted by sex, age, and household registration type, with interaction terms included to test for interactions. Logistic regression sensitivity analyses were performed to validate the robustness of the findings.

This study included 4 638 participants, with 342 new cases of diabetes (7.37%)occurring during the 5-year follow-up period. RCS analysis revealed that higher baseline UHR and cumulative UHR levels were associated with increased diabetes risk in middle-aged and older adults. After adjusting for confounders, diabetes risk significantly increased in Class 2 and Class 3 groups compared to Class 1, with HRs of 1.42 (95% CI: 1.10-1.83) and 1.79 (95% CI: 1.27-2.53),respectively. Furthermore, compared with the lowest tertile group (T1) of baseline UHR and cumulative UHR, the adjusted HRs for the highest tertile group (T3) were 1.64 (95% CI: 1.21-2.23) and 1.60 (95% CI: 1.18-2.18), respectively.

Elevated UHR levels increase the risk of developing diabetes. Continuous monitoring of UHR and early intervention can help reduce the incidence of diabetes in middle-aged and older adults.