Related Experiment Video
Updated: Aug 6, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Development and Validation of a Predictive Model for Mild Cognitive Impairment in Older Adults with Multimorbidity
Lili Shao1, Ruyi Zhang2, Yinqing Huang2
1Department of Geriatric Medicine, The Affiliated Kangning Hospital of Wenzhou Medical University, Zhejiang Provincial Clinical Research Center for Mental Disorder, Wenzhou, Zhejiang, 325000, People's Republic of China.
Objective:
To identify factors associated with mild cognitive impairment (MCI) in older adults with multimorbidity and to develop and validate a predictive model for early screening.
Methods:
This cross-sectional study consecutively enrolled 238 older adult inpatients with multimorbidity at the Affiliated Kangning Hospital of Wenzhou Medical University, China, between April 2022 and February 2025. Participants were assessed using a self-designed general information questionnaire and the Montreal Cognitive Assessment Basic Scale (MoCA-B). MCI was diagnosed according to the Chinese Expert Consensus. Associated factors were identified using logistic regression analysis. A nomogram prediction model was constructed based on these factors. The model's performance was evaluated using the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA). An external validation cohort (n=68) was used to further test the model.
Results:
MCI prevalence was 26.81%. Risk factors: age ≥80 years (OR=3.23), hearing impairment (OR=4.04), and emotional disorders (OR=3.25). Protective factors: higher education (OR=0.23), more frequent physical exercise (OR=0.15), and more frequent social activities (OR=0.26) (all P<0.05). The AUC was 0.862 (training) and 0.832 (external validation). Calibration curves showed good agreement, and DCA indicated net clinical benefit across threshold probabilities of (10-65%).
Conclusion:
Age, hearing impairment, emotional disorders, education, physical exercise, and social activities are significantly associated with MCI in older adults with multimorbidity. The nomogram demonstrates good predictive accuracy and clinical utility, aiding early identification and targeted prevention.
Insights
Mild cognitive impairment (MCI) in older adults with multiple health conditions is linked to age, hearing loss, and emotional issues. Higher education, exercise, and social engagement may protect against MCI.
Area of Science:
- Gerontology
- Neurology
- Public Health
Background:
- Multimorbidity is common in older adults, increasing the risk of cognitive decline.
- Early identification of mild cognitive impairment (MCI) is crucial for timely intervention.
- Existing screening tools may not fully capture the complexities of MCI in multimorbid populations.
Purpose of the Study:
- To identify factors associated with MCI in older adults experiencing multimorbidity.
- To develop and validate a predictive model for the early screening of MCI in this demographic.
Main Methods:
- A cross-sectional study of 238 older adult inpatients with multimorbidity.
- Assessment using a general information questionnaire and the Montreal Cognitive Assessment Basic Scale (MoCA-B).
- Logistic regression for factor identification and nomogram construction; ROC, calibration, and DCA for model validation, including an external cohort.
Main Results:
- MCI prevalence was 26.81%.
- Risk factors included advanced age (≥80 years), hearing impairment, and emotional disorders.
- Protective factors were higher education, frequent physical exercise, and social activities.
Conclusions:
- Key factors associated with MCI in multimorbid older adults identified: age, hearing, emotional state, education, physical activity, and social engagement.
- A nomogram model demonstrated good predictive accuracy and clinical utility for early MCI identification.
- The findings support targeted prevention strategies for MCI in this vulnerable population.
