Scholarlink

Background <p>Non-fatty chronic liver disease (CLD) is a major global health burden. Individuals with metabolic syndrome (MetS) have a substantially increased risk of developing non-fatty CLD. This study aimed to develop and validate risk prediction models for non-fatty CLD among middle-aged and older adults with MetS.</p> Methods <p>Data were obtained from the China Health and Retirement Longitudinal Study (CHARLS). The dataset was randomly divided into a training set (70%) and a testing set (30%). Least absolute shrinkage and selection operator (LASSO) logistic regression with ten-fold cross-validation was applied for predictor selection. Two risk prediction models and nomograms were constructed to predict the 9-year risk of incident non-fatty CLD: a non-laboratory model and a laboratory-based model. Model performance was evaluated using receiver operating characteristic curves and the area under the curve (AUC), calibration curves, decision curve analysis (DCA), and clinical impact curves (CIC).</p> Results <p>Among 4,748 participants, 253 developed non-fatty CLD during the approximately 9-year follow-up period, yielding a cumulative incidence of 5.33%. LASSO regression identified six non-laboratory predictors (male sex, dyslipidemia, heart disease, kidney disease, digestive system disease, and lack of piped water) in non-laboratory model and two additional laboratory indicators (low platelet count and high triglyceride level) in laboratory model. The non-laboratory model achieved AUCs of 0.762 (95% CI: 0.720–0.803) and 0.744 (95% CI: 0.680–0.808) in the training and testing sets, respectively, while the laboratory model showed improved discrimination with AUCs of 0.792 (95% CI: 0.754–0.830) and 0.769 (95% CI: 0.712–0.825). Calibration curves demonstrated good agreement between predicted and observed risks. DCA and CIC indicated that both models provided meaningful clinical net benefit across a wide range of risk thresholds.</p> Conclusion <p>Both the non-laboratory and laboratory-based risk prediction models for 9-year risk of non-fatty CLD in individuals with MetS demonstrated good predictive performance. However, given the reliance on self-reported outcome data without objective confirmation, these findings should be considered exploratory and hypothesis-generating. Pending external validation, these models may eventually serve as practical tools for early risk stratification in this high-risk population.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Development and validation of a risk model for non-fatty chronic liver disease in middle-aged and older adults with metabolic syndrome

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月15日

查看全文
  • Xiumei Zhou,
  • Xiaoxin Wu,
  • Lingxiao Jin,
  • Xuhong Zhao,
  • Qiaozhen Zhang,
  • Junfeng Yin

摘要

Background

Non-fatty chronic liver disease (CLD) is a major global health burden. Individuals with metabolic syndrome (MetS) have a substantially increased risk of developing non-fatty CLD. This study aimed to develop and validate risk prediction models for non-fatty CLD among middle-aged and older adults with MetS.

Methods

Data were obtained from the China Health and Retirement Longitudinal Study (CHARLS). The dataset was randomly divided into a training set (70%) and a testing set (30%). Least absolute shrinkage and selection operator (LASSO) logistic regression with ten-fold cross-validation was applied for predictor selection. Two risk prediction models and nomograms were constructed to predict the 9-year risk of incident non-fatty CLD: a non-laboratory model and a laboratory-based model. Model performance was evaluated using receiver operating characteristic curves and the area under the curve (AUC), calibration curves, decision curve analysis (DCA), and clinical impact curves (CIC).

Results

Among 4,748 participants, 253 developed non-fatty CLD during the approximately 9-year follow-up period, yielding a cumulative incidence of 5.33%. LASSO regression identified six non-laboratory predictors (male sex, dyslipidemia, heart disease, kidney disease, digestive system disease, and lack of piped water) in non-laboratory model and two additional laboratory indicators (low platelet count and high triglyceride level) in laboratory model. The non-laboratory model achieved AUCs of 0.762 (95% CI: 0.720–0.803) and 0.744 (95% CI: 0.680–0.808) in the training and testing sets, respectively, while the laboratory model showed improved discrimination with AUCs of 0.792 (95% CI: 0.754–0.830) and 0.769 (95% CI: 0.712–0.825). Calibration curves demonstrated good agreement between predicted and observed risks. DCA and CIC indicated that both models provided meaningful clinical net benefit across a wide range of risk thresholds.

Conclusion

Both the non-laboratory and laboratory-based risk prediction models for 9-year risk of non-fatty CLD in individuals with MetS demonstrated good predictive performance. However, given the reliance on self-reported outcome data without objective confirmation, these findings should be considered exploratory and hypothesis-generating. Pending external validation, these models may eventually serve as practical tools for early risk stratification in this high-risk population.