Scholarlink

Background <p>Identifying mild cognitive impairment (MCI) in its early stages is vital for averting dementia and fostering healthy aging. However, most screening tools depend on neuropsychological assessments that are expensive, time-consuming, and hard to implement in rural or resource-limited areas. With the increasing use of digital health technologies in community care, there is a pressing need for data-driven models that can be turned into simple, digital tools for large-scale MCI risk screening. This study aimed to identify factors linked to MCI among individuals aged 60 and above in rural China, develop and validate a data-driven, digitally implementable prediction model, and provide evidence to support early intervention strategies.</p> Methods <p>Data were collected from 3,375 participants aged ≥ 60 in the 2020 China Health and Retirement Longitudinal Study (CHARLS). Participants were allocated randomly to a training group (70%) and a validation group (30%). Least absolute shrinkage and selection operator (LASSO) regression was used for variable selection, followed by multivariable logistic regression to develop the predictive model. Model performance was evaluated using receiver operating characteristic (ROC) analysis, calibration plots, and decision curve analysis (DCA).</p> Results <p>3,375 individuals aged 60 and older were included in this analysis, based on data from the 2020 China Health and Retirement Longitudinal Study (CHARLS); 723 individuals (21.4%) were identified as having MCI. Seven independent predictors were retained in the final model: education, sleep duration, depressive symptoms, inactivity, drinking, hobbies, and retirement. The nomogram demonstrated good discrimination (AUC = 0.734 in the training set; 0.735 in the validation set) and strong calibration, indicating stable and reliable predictive performance. The model allows for personalized MCI risk assessment using accessible variables, supporting early screening among rural older adults.</p> Conclusions <p>This research developed and validated a data-driven nomogram with the potential for digital implementation to facilitate early MCI screening in rural China. This model aligns with national and global strategies for healthy aging and may support the implementation of scalable, technology-assisted cognitive health management in primary care and public health systems.</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

Predicting mild cognitive impairment among rural older adults in China: development and validation of a risk prediction model

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月26日

查看全文
  • Jingzheng Yan,
  • Yaning Wang,
  • Zhaotai Wang,
  • Yingjuan Cao

摘要

Background

Identifying mild cognitive impairment (MCI) in its early stages is vital for averting dementia and fostering healthy aging. However, most screening tools depend on neuropsychological assessments that are expensive, time-consuming, and hard to implement in rural or resource-limited areas. With the increasing use of digital health technologies in community care, there is a pressing need for data-driven models that can be turned into simple, digital tools for large-scale MCI risk screening. This study aimed to identify factors linked to MCI among individuals aged 60 and above in rural China, develop and validate a data-driven, digitally implementable prediction model, and provide evidence to support early intervention strategies.

Methods

Data were collected from 3,375 participants aged ≥ 60 in the 2020 China Health and Retirement Longitudinal Study (CHARLS). Participants were allocated randomly to a training group (70%) and a validation group (30%). Least absolute shrinkage and selection operator (LASSO) regression was used for variable selection, followed by multivariable logistic regression to develop the predictive model. Model performance was evaluated using receiver operating characteristic (ROC) analysis, calibration plots, and decision curve analysis (DCA).

Results

3,375 individuals aged 60 and older were included in this analysis, based on data from the 2020 China Health and Retirement Longitudinal Study (CHARLS); 723 individuals (21.4%) were identified as having MCI. Seven independent predictors were retained in the final model: education, sleep duration, depressive symptoms, inactivity, drinking, hobbies, and retirement. The nomogram demonstrated good discrimination (AUC = 0.734 in the training set; 0.735 in the validation set) and strong calibration, indicating stable and reliable predictive performance. The model allows for personalized MCI risk assessment using accessible variables, supporting early screening among rural older adults.

Conclusions

This research developed and validated a data-driven nomogram with the potential for digital implementation to facilitate early MCI screening in rural China. This model aligns with national and global strategies for healthy aging and may support the implementation of scalable, technology-assisted cognitive health management in primary care and public health systems.