A hierarchical machine learning model for predicting self-harm and suicidal behavior in hospitalized patients with schizophrenia using clinical history and nursing observations - Summary - MDSpire

A hierarchical machine learning model for predicting self-harm and suicidal behavior in hospitalized patients with schizophrenia using clinical history and nursing observations

  • By

  • Liang, Chen

  • Meng, Xianfeng

  • Duan, Ying

  • Yang, Wei

  • Zhu, Gang

  • Wang, Jinhuan

  • Sun, Ying

  • Wang, Mingtao

  • Liu, Miao

  • Sun, Chenhui

  • Hu, Kunyuan

  • Shao, Wei

  • Ren, Jintao

  • Shao, Xiaojun

  • Zhang, Yang

  • April 24, 2026

  • 0 min

Share

Objective:

To develop and evaluate a two-layered machine learning framework for identifying schizophrenia inpatients at high risk of self-harm or suicidal acts.

Key Findings:
  • Previous self-harm (OR = 4.323), hopelessness/depression (OR = 3.090), younger age (OR = 0.938), and higher educational level (OR = 1.357) were independent predictors of self-harm/suicidal behavior.
  • Dynamic indicators such as negative self-evaluation (OR = 2.303) and self-reported depression (OR = 1.812) were significant predictors.
  • The optimized static LR model achieved an AUC of 0.7564, while the dynamic SVM model reached an AUC of 0.8531.
  • The fused model improved performance with an AUC of 0.9048, sensitivity of 0.8542, specificity of 0.7789, and accuracy of 0.8042.
Interpretation:

The hierarchical machine learning model effectively identifies high-risk schizophrenia inpatients for self-harm and suicidal behavior, enhancing clinical detection and intervention strategies.

Limitations:
  • Retrospective design may introduce bias.
  • Generalizability may be limited to the specific population studied.
Conclusion:

A hierarchical machine learning approach integrating clinical history and nursing assessments can effectively identify high-risk patients, supporting timely preventive strategies in psychiatric settings.

Original Source(s)

A hierarchical machine learning model for predicting self-harm and suicidal behavior in hospitalized patients with schizophrenia using clinical history and nursing observations

  • by Liang, Chen , Meng, Xianfeng , Duan, Ying , Yang, Wei , Zhu, Gang , Wang, Jinhuan , Sun, Ying , Wang, Mingtao , Liu, Miao , Sun, Chenhui , Hu, Kunyuan , Shao, Wei , Ren, Jintao , Shao, Xiaojun , Zhang, Yang

  • April 24, 2026

Related Content

Romanian male patients with the dual diagnosis of schizophrenia and alcohol use disorder: a prospective study of clinical, social, and treatment-related factors affecting quality of life

Could Digital Twins Forecast Forensic Risk?

Researchers outline a proposed three-level framework spanning patient monitoring, ward operations, and forensic system planning.

A systematic review of obesity prevalence and its contributing factors among individuals diagnosed with schizophrenia