Predicting depression among rural and urban disabled elderly in China using a random forest classifier

Yu Xin,Xiaohui Ren

doi:10.1186/s12888-022-03742-4

Abstract

With global aging, the number of elderly with physical disabilities is also increasing. Compared with the ordinary elderly, the elderly who lose their independence are more likely to have the symptoms of depression. Reducing depression may help to alleviate the disability process of those who find themselves in the disabled stages. Therefore, the purpose of this study is to explore the predictive effects of demographic characteristics, health behavior, health status, family relations, social relations, and subjective attitude on depression in rural and urban disabled elderly to improve early depression symptom recognition.A total of 1460 older adults aged 60 and disabled were selected from China Family Panel Studies (CFPS). Depression was assessed according to The Center for Epidemiologic Studies Depression Scale (CES-D). This paper used the random forest classifier to predict the depression of the disabled elderly from six aspects: demographic characteristics, health status, health behavior, family relationship, and social relationship. The prediction model was established based on 70% of the training set and 30% of the test set. The depression rate of rural disabled elderly was 57.67%, and that of urban disabled elderly was 44.59%. The mean values of the 10-k cross-validated results were 0.71 in rural areas and 0.70 in urban areas. AUC:0.71, specificity: 65.3%, sensitivity: 80.6% for rural disabled elderly with depression; AUC:0.78, specificity: 78.1%, sensitivity: 64.2% for urban disabled elderly with depression, respectively. There are apparent differences in the top ten predictors between rural and urban disabled elderly. The common predictors were self-rated health, changing in perceived health, disease or accidence experience within the past 2 weeks, life satisfaction, trusting people, BMI, and having trust in the future. Non-common predictors were chronic diseases, neighborly relations, total medical expenses within 1 year, community emotion, sleep duration, and family per capita income. Using random forest data to predict the depression of the disabled elderly may lead to early detection of depression.

Highlights

Aging is a challenge for all countries, and the crisis is even more significant in low- and middle-income countries [1]
Data collection The data were derived from the China Family Panel Studies (CFPS)
We identified hyper-parameters used commonly in Random Forest Classifier (RFC) [20]: (1) n-estimators; (2) max depth; (3) min_samples_split; and (4) min_samples_leaf

Summary

Introduction

Aging is a challenge for all countries, and the crisis is even more significant in low- and middle-income countries [1]. Increasing incidences of physical disability are associated with aging, and that significant heterogeneity exists among older populations [2]. Xin and Ren BMC Psychiatry (2022) 22:118 early and late stages of disability. Reducing depression may help alleviate the disability process of those who find themselves in the above stages [5]. Compared with the ordinary elderly, the disabled elderly are more likely to have the symptoms of depression [6, 7]. Özlem and Ünsal found that the incidence of depression was 57.8% (n = 201) in disabled people [8]. The depression of the disabled elderly needs social attention

Objectives

Methods

Results

Discussion

Conclusion