Lipid accumulation product and WHO risk chart-defined cardiovascular high-risk status in community adults: a cross-sectional analysis from the ChinaHEART program.
Lipid accumulation product (LAP), derived from waist circumference and triglycerides, may reflect central lipid overaccumulation, but its relevance for a screening-defined cardiovascular disease (CVD) high-risk label is uncertain. We examined whether LAP could help flag community adults who may need more formal risk assessment rather than serve as a stand-alone prognostic tool.
This cross-sectional analysis initially considered 6,860 adults aged 35-75 years from the ChinaHEART screening program in Luohe. After excluding participants with missing LAP or screening-defined CVD high-risk outcome data, 6,803 participants were included in the base analytic sample. LAP was analyzed continuously and dichotomized using a rounded cutoff of 42 derived from the Youden index in the LAP-only receiver operating characteristic (ROC) analysis. Logistic regression, restricted cubic splines, subgroup analyses, and ROC analyses were conducted under four stepwise models. Exploratory least absolute shrinkage and selection operator (LASSO), XGBoost, and TreeSHAP analyses were added as supplementary pattern-recognition analyses.
ROC analysis identified an optimal LAP cutoff of 42.12 in the LAP-only model; the rounded value of 42 was used for categorical analyses. LAP alone showed modest discrimination for the concurrent screening-defined CVD high-risk label (AUC 0.583, 95% CI 0.566-0.600), which increased to 0.628 (95% CI 0.613-0.643) after adding age and sex and to 0.632 (95% CI 0.619-0.649) in Model 3. Using the rounded cutoff of 42, 42.3% of participants were classified as high LAP. The prevalence of screening-defined CVD high-risk status was higher in the high-LAP group than in the low-LAP group (28 vs. 17%, P < 0.001).
Higher LAP was associated with a greater likelihood of meeting a contemporaneous screening-defined CVD high-risk criterion. These findings support LAP as a pragmatic screening trigger for further formal assessment in community settings, not as evidence of causal risk stratification or prediction of hard cardiovascular events.
This cross-sectional analysis initially considered 6,860 adults aged 35-75 years from the ChinaHEART screening program in Luohe. After excluding participants with missing LAP or screening-defined CVD high-risk outcome data, 6,803 participants were included in the base analytic sample. LAP was analyzed continuously and dichotomized using a rounded cutoff of 42 derived from the Youden index in the LAP-only receiver operating characteristic (ROC) analysis. Logistic regression, restricted cubic splines, subgroup analyses, and ROC analyses were conducted under four stepwise models. Exploratory least absolute shrinkage and selection operator (LASSO), XGBoost, and TreeSHAP analyses were added as supplementary pattern-recognition analyses.
ROC analysis identified an optimal LAP cutoff of 42.12 in the LAP-only model; the rounded value of 42 was used for categorical analyses. LAP alone showed modest discrimination for the concurrent screening-defined CVD high-risk label (AUC 0.583, 95% CI 0.566-0.600), which increased to 0.628 (95% CI 0.613-0.643) after adding age and sex and to 0.632 (95% CI 0.619-0.649) in Model 3. Using the rounded cutoff of 42, 42.3% of participants were classified as high LAP. The prevalence of screening-defined CVD high-risk status was higher in the high-LAP group than in the low-LAP group (28 vs. 17%, P < 0.001).
Higher LAP was associated with a greater likelihood of meeting a contemporaneous screening-defined CVD high-risk criterion. These findings support LAP as a pragmatic screening trigger for further formal assessment in community settings, not as evidence of causal risk stratification or prediction of hard cardiovascular events.