Cardiometabolic index and mortality risks: elevated cancer and reduced cardiovascular mortality risk in a large cohort

Abstract

Background: With metabolic disorders on the rise globally, the cardiometabolic index (CMI) has emerged as a crucial predictor of mortality risks linked to cancer, cardiovascular disease, and diabetes. This novel index, which combines lipid metabolism and body composition, is the focus of this study, aimed at exploring its association with all-cause and specific mortality in an all-age adult population.

Methods: A longitudinal cohort study including 5,728 participants aged over 18 from nine cycles between 2001 and 2018 was enrolled and assessed. CMI served as the exposure variable, while outcomes included all-cause mortality and mortality due to cardiovascular disease, cancer, and diabetes. The Cox frailty model and average marginal effects were employed to evaluate the contribution of CMI to all-cause and specific mortality collectively. Restricted cubic spline analyses and stratified analyses were conducted to investigate potential nonlinear effects and interactions.

Results: The decreased participants exhibited considerably higher CMI than the alive's. A positive association was found between CMI and all-cause mortality (HR=1.05, 95% CI=1.01-1.10). Notably, CMI was linked to an increased risk of cancer mortality (HR=1.02) and a reduced risk of cardiovascular disease mortality (HR=0.85). Furthermore, the average marginal effect of CMI on diabetes mortality was the largest (AME=0.499). The RCS curves revealed that participants had the lowest risk of all-cause mortality at a CMI of 0.618. Sensitivity analyses further supported these findings.

Conclusion: This study represents the first comprehensive assessment on the contribution of CMI to mortality across an all-age adult population, providing some insights for the comprehensive assessment of health and disease states.

Keywords: All-cause and specific mortality; Average marginal effects; Cardiometabolic index; Cox Frailty model; Longitudinal cohort study.

Fig. 1

A flowchart for participant selection

Fig. 2

Kaplan–Meier curves of the survival rate of participants with CMI quartiles. CMI was divided into 4 groups from smallest to largest, and survival probabilities for the 4 causes were calculated by stratification. Kaplan-Meier survival curves indicated that there was no difference between all-cause and specific-cause mortality among stratified participants by CMI. P-values for all-cause mortality (A), cancer mortality (B), diabetes mortality (C), and CVD mortality (C), were calculated in a stratified manner. P-value for all-cause mortality (A), cancer mortality (B), diabetes mortality (C), and CVD mortality (D) is 0.3, 0.18, 0.85, and 0.89 respectively

Fig. 3

Dose-response curves of CMI and all-cause mortality and specific-mortality. Restricted cubic spline curves (RCS) were used to explore potential dose-response patterns, and thresholds were calculated for each curve. For all-cause mortality (A), the hazard ratio decreases and then increases with increasing CMI until a plateau. For cancer mortality (B), the hazard ratio rises and then falls with increasing CMI. For CVD and diabetes mortality (C and D), the risk ratios declined with increasing CMI to the thresholds and then stabilized

Fig. 4

Subgroup analysis for all-cause mortality. Gender, age, race, BMI, PIR, and history of hypertension, diabetes, hypercholesterolemia, and cardiovascular disease alcohol use, smoke now were all adjusted except for the covariates themselves. Bold values indicate p-value < 0.05, which is statistically significant