Trajectories of a Composite Lipid, Glucose, and Refined Obesity Index Predict Cardiovascular Disease

Figures

• 

  Figure 1. Flowchart of study population inclusion and exclusion screening based on CHARLS data. CHARLS, China Health and Retirement Longitudinal Study; CVD, cardiovascular disease; LDL‐C, low‐density lipoprotein cholesterol; CHG, cholesterol‐high‐density lipoprotein‐glucose; NON‐CVD, noncardiovascular disease.

• 

  Figure 2. Determination of the optimal number of clusters for CHG‐ROD trajectories. The optimal number of clusters (K) for the longitudinal trajectories of the CHG‐ROD was determined using the elbow method (WSS). Two clusters were identified via the K‐means method (based on Euclidean distance) for CHG‐BMI, CHG‐WC, CHG‐WHtR, CHG‐VAI, and CHG‐CVAI. CHG, cholesterol‐high‐density lipoprotein‐glucose; CHG‐ROD, the composite index of CHG and refined obesity derivatives; WSS, Within‐Cluster Sum of Square; WC, waist circumference; WHtR, waist‐to‐height ratio; BMI, body mass index; VAI, visceral adiposity index; CVAI, CVD‐associated adiposity index.

• 

  Figure 3. K‐means trajectory patterns: CHG‐ROD scatter plots 2012 versus 2015. A scatter plot comparing standardized values of CHG‐BMI, CHG‐WC, CHG‐WHtR, CHG‐VAI, and CHG‐CVAI between 2012 (x‐axis) and 2015 (y‐axis). Each cluster is marked by a unique color and shape to distinguish groups and highlight temporal variations, with two‐cluster patterns identified for CHG‐BMI, CHG‐WC, CHG‐WHtR, CHG‐VAI, and CHG‐CVAI. Clustering analysis categorized these composite indices into two trajectories: Pattern 1 (well‐controlled) and Pattern 2 (poorly controlled). CHG, cholesterol‐high‐density lipoprotein‐glucose; CHG‐ROD, the composite index of CHG and refined obesity derivatives; WC, waist circumference; WHtR, waist‐to‐height ratio; BMI, body mass index; VAI, visceral adiposity index; CVAI, CVD‐associated adiposity index.

• 

  Figure 4. K‐means trajectory patterns: CHG‐ROD line plots 2012–2015. Line graphs display mean values with error bars for each cluster (2012–2015), illustrating dynamic changes in metabolic‐adiposity markers over the study period. These graphs highlight two distinct profiles: consistently well‐controlled and poorly controlled. CHG‐BMI, CHG‐WC, CHG‐WHtR, and CHG‐VAI maintained stable trajectories (persistently low or high levels), whereas CHG‐CVAI showed a slight upward trend in both clusters from 2012 to 2015, differing from the other four indices. CHG, cholesterol‐high‐density lipoprotein‐glucose; CHG‐ROD, the composite index of CHG and refined obesity derivatives; WC, waist circumference; WHtR, waist‐to‐height ratio; BMI, body mass index; VAI: visceral adiposity index; CVAI, CVD‐associated adiposity index.

• 

  Figure 5. Dose–response of baseline/cumulative CHG‐ROD to CVD risk using restricted cubic spline. (A) Dose–response relationship of the baseline levels of CHG‐ROD with incident CVD; (B) dose–response relationships of cumulative exposures of CHG‐ROD with incident CVD. CHG, cholesterol‐high‐density lipoprotein‐glucose; CHG‐ROD, the composite index of CHG and refined obesity derivatives; WC, waist circumference; WHtR, waist‐to‐height ratio; BMI, body mass index; VAI, visceral adiposity index; and CVAI, CVD‐associated adiposity index.

• 

  Figure 6. Component contribution rates in cumulative CHG‐ROD via WQS regression. Cumulative CHG‐ROD composite indices (cuCHG‐BMI, cuCHG‐WC, cuCHG‐WHtR, cuCHG‐VAI, and cuCHG‐CVAI). CHG, cholesterol‐high‐density lipoprotein‐glucose; CHG‐ROD, the composite index of CHG and refined obesity derivatives; CVD, cardiovascular disease; WQS, weighted quantile sum; cu, cumulative; FPG, fasting blood glucose; TC, total cholesterol; HDL, high‐density lipoprotein cholesterol; LDL‐C, low‐density lipoprotein cholesterol; WC, waist circumference; WHtR, waist‐to‐height ratio; BMI, body mass index; VAI, visceral adiposity index; and CVAI, CVD‐associated adiposity index.