Scholarlink

Objective <p>This study investigates the relationship between the Kidney Disease Index (KDI) and cognitive function, evaluating its potential as a predictive marker for cognitive impairment in older adults. We also compare the performance of KDI with traditional kidney function markers, including estimated glomerular filtration rate (eGFR) and urine albumin-to-creatinine ratio (UACR), in predicting cognitive decline.</p> Methods <p>Data from the National Health and Nutrition Examination Survey (NHANES) involving participants aged 60 years or older were analyzed. Multivariable regression models were employed to assess the relationship between KDI, eGFR, UACR, and cognitive impairment. Restricted cubic splines (RCS) were used to explore the dose-response relationship between KDI levels and cognitive impairment, and interaction analyses were performed to assess how KDI impacts cognitive function across various demographic groups. Machine learning techniques, including LASSO, XGBoost, and random forests, were applied for feature selection and model validation to predict cognitive impairment. The performance of the predictive models was evaluated using receiver operating characteristic (ROC) curves and decision curve analysis (DCA). Calibration was assessed through calibration curves to ensure accurate prediction probabilities, while SHAP values were used to interpret the relative importance of KDI and other variables in predicting cognitive impairment.</p> Results <p>A higher KDI level may be significantly associated with lower cognitive scores and an increased risk of cognitive impairment. For each 0.1 unit increase in KDI, cognitive scores decrease by 0.17 points, and the probability of cognitive impairment increases by 15%. Machine learning confirmed that the AUC of KDI (&gt; 0.7) is higher than that of eGFR and UACR, indicating that KDI may outperform eGFR and UACR in predicting cognitive impairment. The SHAP model further validated the predictive value of KDI.</p> Conclusion <p>KDI may be a reliable and effective indicator for predicting cognitive function and the risk of cognitive impairment in the elderly. Research suggests that higher KDI levels may serve as a risk factor for cognitive decline, underscoring the importance of managing KDI for cognitive health, particularly in populations with chronic kidney disease (CKD). Future studies should focus on longitudinal research and clinical interventions to further validate the clinical utility of KDI and develop treatment strategies based on KDI.</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

Exploring the potential relationship between kidney disease index and cognitive dysfunction: a machine learning approach with NHANES data

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年8月26日

查看全文
  • Tianyi Wang,
  • Shuang Tang,
  • Ruwen Zheng,
  • Lin Wang

摘要

Objective

This study investigates the relationship between the Kidney Disease Index (KDI) and cognitive function, evaluating its potential as a predictive marker for cognitive impairment in older adults. We also compare the performance of KDI with traditional kidney function markers, including estimated glomerular filtration rate (eGFR) and urine albumin-to-creatinine ratio (UACR), in predicting cognitive decline.

Methods

Data from the National Health and Nutrition Examination Survey (NHANES) involving participants aged 60 years or older were analyzed. Multivariable regression models were employed to assess the relationship between KDI, eGFR, UACR, and cognitive impairment. Restricted cubic splines (RCS) were used to explore the dose-response relationship between KDI levels and cognitive impairment, and interaction analyses were performed to assess how KDI impacts cognitive function across various demographic groups. Machine learning techniques, including LASSO, XGBoost, and random forests, were applied for feature selection and model validation to predict cognitive impairment. The performance of the predictive models was evaluated using receiver operating characteristic (ROC) curves and decision curve analysis (DCA). Calibration was assessed through calibration curves to ensure accurate prediction probabilities, while SHAP values were used to interpret the relative importance of KDI and other variables in predicting cognitive impairment.

Results

A higher KDI level may be significantly associated with lower cognitive scores and an increased risk of cognitive impairment. For each 0.1 unit increase in KDI, cognitive scores decrease by 0.17 points, and the probability of cognitive impairment increases by 15%. Machine learning confirmed that the AUC of KDI (> 0.7) is higher than that of eGFR and UACR, indicating that KDI may outperform eGFR and UACR in predicting cognitive impairment. The SHAP model further validated the predictive value of KDI.

Conclusion

KDI may be a reliable and effective indicator for predicting cognitive function and the risk of cognitive impairment in the elderly. Research suggests that higher KDI levels may serve as a risk factor for cognitive decline, underscoring the importance of managing KDI for cognitive health, particularly in populations with chronic kidney disease (CKD). Future studies should focus on longitudinal research and clinical interventions to further validate the clinical utility of KDI and develop treatment strategies based on KDI.