JMIR Publications

ABSTRACT

Background:

Identifying cognitive impairment early enough could support timely intervention that may hinder or delay the trajectory of cognitive impairment, thus increasing the chances for successful cognitive aging.

Objective:

We aimed to build a prediction model based on machine learning for cognitive impairment among Chinese elderly community-dwelling elders with normal cognition.

Methods:

A prospective cohort of 6,718 older people from the Chinese Longitudinal Healthy Longevity Survey (CLHLS) register, followed between 2008 and 2011, was used to develop and validate the prediction model. Participants were included if they were 60 years old or more, community-dwelling elders, and had cognitive MMSE ≥ 18. They were excluded if they had a severe diagnosed disease (cancer, dementia) or were living in institutions. Cognitive impairment was identified using the Chinese version of the Mini-Mental State Examination. Several machine learning algorithms (Random Forest, XGBoost, Naïve Bayes and logistic regression) were used to assess 3-year risk of developing cognitive impairment. Optimal cutoffs and adjusted parameters were explored in validation data and further evaluated the model in test data. A nomogram was established to vividly present the prediction model.

Results:

The mean age of the participants was 80.4 ± 10.3 years, and 50.8% were female. During a 3-year follow-up, 991 (14.8%) participants were identified to have cognitive impairment. Out of 45 features, four features were finally selected to develop a model: age, instrumental activities of daily living, marital status and baseline cognitive function. The concordance index of the model constructed by logistic regression was 0.814 (95%CI 0.781 - 0.846). Older people with normal cognitive functioning with a nomogram score of less than 170 were considered to have low 3-year risk of cognitive impairment, and those with a score of 170 or greater were considered to have a high 3-year risk.

Conclusions:

This simple and feasible cognitive impairment prediction model could identify elderly community-dwelling elders at the greatest 3-year risk for cognitive impairment, which could be helpful for community nurses in early identification of dementia.