U-shaped relationship between frailty and non-HDL-cholesterol in the elderly: a cross-sectional study

BACKGROUND: The modulation of lipid metabolism has been explored as a potential treatment for frailty, yet the association between non-high-density lipoprotein-cholesterol (non-HDL-C) and frailty remains unclear. METHODS: This study utilized data from five cycles of the National Health and Nutrition Examination Survey (NHANES) and two cycles of the China Health and Retirement Longitudinal Study (CHARLS) to investigate this relationship. A 40-item frailty index scale, encompassing various dimensions of somatic functioning, psychological evaluation, and illness, was developed and individually evaluated for each participant. The variables underwent screening through Least Absolute Shrinkage and Selection Operator (LASSO) regression, univariate logistic regression, and Light Gradient Boosting Machine (LightGBM), with models developed through multivariate logistic regression and the LightGBM algorithm. Subsequently, subgroup analyses and interaction tests were conducted to substantiate correlations. RESULTS: The U-shaped nonlinear association between non-HDL-C and frailty in older adults was validated using the LightGBM algorithm. Non-HDL cholesterol levels in the range of 117.54-194.64 mg/dL were less likely to be frailty, while the likelihood of developing frailty was higher at 47.99-63.87 or 274.01-259.65 mg/dL. Subgroup analyses and interaction tests confirm these results. CONCLUSION: It is plausible that an intricate nonlinear association between non-HDL-C and frailty in the elderly exists, though further rigorously designed studies are imperative to validate this relationship.