Association between daily salt intake and cardiometabolic indicators among employees of a research institute in Beijing, China: a cross-sectional study

Objectives To examine the associations between daily salt intake (DSI) and a comprehensive panel of cardiometabolic indicators among employees undergoing annual occupational health examinations. Design A cross-sectional, single-centre observational study. Setting Annual occupational health examinations at an aerospace research institute in Beijing, China. Participants Between May 2025 and July 2025, 2642 participants were enrolled. Each participant provided a fasting morning spot urine sample and written informed consent. Methods Spearman correlation, multivariable linear regression and restricted cubic spline models were used to examine the associations between DSI and cardiometabolic indicators. Model robustness was assessed through stratified analysis and multivariable logistic regression analysis. Results After adjusting for confounding factors, DSI showed no significant associations with systolic blood pressure (SBP), diastolic blood pressure (DBP), high-density lipoprotein cholesterol (HDL-C) or low-density lipoprotein cholesterol (LDL-C) but exhibited significant negative associations with uric acid (UA), serum creatinine (Cr), triglycerides, fasting plasma glucose and total cholesterol (all p<0.05). Only DBP showed a nonlinear pattern, with an estimated inflection point at a DSI of 8.66 g. In stratified analysis, DSI was inversely associated with UA and Cr across all subgroups while showing no significant association with SBP, DBP, HDL-C or LDL-C. In multivariable logistic regression, higher DSI was associated with lower odds of dyslipidaemia (OR=0.936), impaired fasting glucose (OR=0.832), hyperuricaemia (OR=0.847), higher serum creatinine (OR=0.874) but not hypertension. Conclusions In this occupational cohort, DSI showed either nonsignificant or inverse associations with cardiometabolic indicators after adjustment for confounders. Stratified and multivariable logistic regression analyses showed generally consistent patterns of association.