Association between the triglyceride-high-density lipoprotein-glucose-body index and new-onset cardiometabolic multimorbidity across blood pressure strata: a prospective cohort study

Background Cardiometabolic multimorbidity (CMM), the coexistence of two or more cardiometabolic diseases, imposes a growing burden on middle-aged and older adults in China. Insulin resistance (IR) is a crucial modifiable contributor to this clustering. The triglyceride-glucose (TyG) index, a recognized IR surrogate, overlooks the protective role of high-density lipoprotein cholesterol (HDL-C) and the contribution of body mass index (BMI). The novel triglyceride-high-density lipoprotein-glucose-body index (TyHGB) integrates all four domains; however, its association with new-onset CMM, and whether this varies across blood pressure (BP) strata, remain unclear. Methods Data were obtained from the China Health and Retirement Longitudinal Study (CHARLS, 2011–2020). We included 7,830 participants free of CMM at baseline (fewer than two of heart disease, stroke, or diabetes) with classifiable BP status (normal, high-normal, or hypertension). Cox models with three incremental adjustment sets evaluated the TyHGB–CMM association; missing covariates were handled using multiple imputation by chained equations (m = 20) with Rubin's rules. Multiplicative and additive (relative-excess-risk scale) interactions with BP status were tested. Restricted cubic splines explored non-linearity; incremental predictive value (change in 9-year time-dependent area under the curve, ΔAUC) was assessed within strata. Results During approximately 9 years of follow-up, 803 (10.3%) incident CMM events were recorded. In the fully adjusted model, TyHGB was associated with higher CMM risk (hazard ratio [HR] = 1.135 per unit, 95% CI: 1.115–1.155; per-SD HR = 1.334, 1.282–1.388; quartile 4 versus 1 HR = 4.268, 3.271–5.570). The association was consistent across BP strata (P for interaction = 0.567), showed significant non-linearity (P for non-linearity < 0.001), and was predominantly expressed through diabetes-containing event combinations. The additive interaction between elevated TyHGB and hypertension did not persist after accounting for baseline cardiometabolic disease composition. Within-stratum predictive gains were modest (ΔAUC: 0.012–0.025). Conclusions The TyHGB index was independently associated with new-onset CMM, with consistent strength across BP strata, expressed predominantly through diabetes-inclusive events. The co-occurrence of elevated TyHGB and hypertension identifies a jointly adverse phenotype with potential value for combined screening. These observational findings in middle-aged and older Chinese adults warrant external validation. Trial registration Not applicable.