What is the clinical evidence from this study?

Study design: Cross-Sectional. Population: Maintenance hemodialysis (n=362). Intervention: Random Forest model vs. Other machine learning models. Primary outcome: AUC for predicting social isolation.

February 18, 2026Frontiers in Psychiatry2 citationsOpen Access

Predicting social isolation in maintenance hemodialysis patients using machine learning methods: a cross-sectional study

YLYing Li WZWen Zhao BWBoyang Wang

Key Result

The Random Forest machine learning model effectively predicted the risk of social isolation in maintenance hemodialysis patients with an AUC of 0.95.

Study Design

Type

Cross-Sectional (n=362)

Multicenter

Structured PICO

Can machine learning models accurately predict social isolation in maintenance hemodialysis patients?

Population

362 patients with maintenance hemodialysis (MHD) from a tertiary hospital in Shanghai

Intervention

Machine learning-based risk prediction models (Random Forest, Decision Tree, K-Nearest Neighbor, Logistic Regression, Resilient Network, Extreme Gradient Boosting, and Support Vector Machine)

Outcome

Prediction of social isolation

A Random Forest machine learning model can accurately predict the risk of social isolation in maintenance hemodialysis patients, identifying heart failure and psychosocial factors as key predictors.

Limitations

Absence of genomic data constrains mechanistic interpretation
Single-center cross-sectional study with a moderate sample size limits generalizability
Risk of underfitting and unstable performance evaluation due to lack of k-fold nested cross-validation
Coding approach for predictor variables may not adequately capture the true underlying data structure
Convenience sampling method may have introduced selection bias
Single-center study
Needs validation in larger multicenter cohorts

Abstract

Background This study aims to develop and validate a machine learning-based risk prediction model for social isolation in maintenance hemodialysis (MHD) patients, and at the same time determine the key risk factors. Method 362 patients with MHD were recruited from a tertiary hospital in Shanghai and randomly divided into the training group and the detection group. We implemented and compared seven machine learning algorithms: Random Forest (RF), Decision Tree (DT), K-Nearest Neighbor (KNN), Logistic Regression (LR), Resilient Network (EN), Extreme Gradient Boosting (XGB), and Support Vector Machine (SVM). Result In our MHD cohort, the incidence of social isolation was 45.856%. The comparative analysis shows that RF is the best prediction model (AUC = 0.95). Feature importance analysis identified significant predictors: Place of residence (1.277), Heart failure (HF) (0.559), Anxiety (0.306), Monthly household income (0.269), Age (0.255) Sleep condition (0.138). Conclusion The prediction model based on RF has a good effect in identifying the social isolation risk of MHD patients. These findings enable clinicians to stratify high-risk populations and implement timely and targeted intervention measures, effectively reducing the risk of adverse consequences. Future multicenter studies should validate these results in larger cohorts.

Ask AI

Helpful

Bookmark

View Full Paper