Related Experiment Video
Updated: Oct 10, 2026

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Screening patterns for cognitive dysfunction and influencing factors analysis for referral and training rate:
Jian Dong1, Hongrui Ma2,3, Min Deng1
1Health Management Center, Xuanwu Hospital, Capital Medical University, Beijing, China.
Abstract:
BackgroundImplementing cognition health screening, risk assessment, and early intervention for cognitive impairment-related diseases among appropriate demographics holds significant importance in China.ObjectiveThis study aims to present a cognition health management model and to explore factors influencing referral and cognition training rates for participants enrolled in a health management center.MethodsThis cohort study examined 836 cases of potential cognitive dysfunction at a health management center in China from January to October 2023. Baseline characteristics, three-month follow-up data, and clinical outcomes were recorded. Bivariate logistic regression was employed to identify predictive factors for referral and cognition training rates.ResultsIn this study, the referral and cognition training rates among participants were only 6.58% and 4.07%, respectively. The primary barriers identified were lack of awareness and accurate understanding of cognition health and training, often linked to lower levels of education. Comorbidities of cerebrovascular disease (adjusted OR: 0.277; 95%CI: 0.112-0.683, p = 0.005) and higher levels of fasting blood glucose upon admission (adjusted OR: 0.849; 95%CI: 0.746-0.955, p = 0.013) were associated with referral. A lower MoCA score on admission (adjusted OR: 0.856; 95%CI: 0.742-0.986, p = 0.032) was associated with receiving cognition training.ConclusionsTargeted educational campaigns are essential to address low awareness of cognition health, particularly among those with lower education. Moreover, considering comorbidities and biochemical markers can enhance the prediction and implementation of referral and cognition training rates.
