Supplements

A logistic regression-based risk classification model can predict insomnia symptoms in community residents with satisfactory performance.

A machine learning model using logistic regression achieved AUC values of 0.812 in internal validation and 0.795 in external validation for classifying insomnia symptom risk. The top five contributing factors were sleep time, sleep latency, bedtime, health status, and age.

Last updated: Jul 14, 2026•0 RCTs•📖 Read as article →

Evidence Score

Evidence Score34/100
Human RCT☆☆☆☆☆
Meta-analysis☆☆☆☆☆
Mechanism★★★★★
Safety★★★★☆
Confidencelow

Study Evidence

Study 1. Machine learning-based insomnia symptom risk classification model for residents of Hebei province.

observational

Wang Y, Qi C, Cao X, Meng F, Zhao Y, Bai P, Song J, Wang S, Liu B, Song B · Public health (2026)

Participants: N/A
Duration: Cross-sectional (surveys in 2021 and 2024)
Intervention: Machine learning-based risk classification model using logistic regression algorithm
Outcome: Insomnia symptom risk classification performance (AUC), and feature importance via SHAP values
Effect Size: AUC = 0.812 (internal validation), AUC = 0.795 (external validation)
Population: Community-based residents of Hebei Province, China (16,848 eligible participants from 30 counties)

Result:

PubMed: 42413427DOI: 10.1016/j.puhe.2026.106409

Mechanism Graph

Collect cross-sectional survey data on sleep-related variables and insomnia symptoms (ISI)
↓
Train six machine learning algorithms on 2021 dataset (training set n=6863)
↓
Select logistic regression as optimal model based on AUC performance
↓
Apply SHAP analysis to identify top influential features: sleep time, sleep latency, bedtime, health status, age

Limitations

  • ⚠Cross-sectional design prevents causal inference
  • ⚠Model was developed and validated only within Hebei Province, limiting generalizability to other populations

Frequently Asked Questions

What was the best performing machine learning model in this study?▼

The logistic regression model showed the best classification performance with AUC values of 0.812 (internal) and 0.795 (external).

Which factors were most important for predicting insomnia symptoms?▼

The top five factors were sleep time, sleep latency, bedtime, health status, and age, as identified by SHAP analysis.

How many participants were included in this study?▼

The study enrolled 16,848 eligible participants from 30 counties in Hebei Province, China.

What tool was used to assess insomnia symptoms?▼

Insomnia symptoms were assessed using the Insomnia Severity Index (ISI).

Products

Affiliate links coming soon. We only recommend products that match the doses and forms used in the cited research.

References

  1. 1.Wang Y, Qi C, Cao X, Meng F, Zhao Y, Bai P, Song J, Wang S, Liu B, Song B. "Machine learning-based insomnia symptom risk classification model for residents of Hebei province.." Public health, 2026. PMID: 42413427 DOI: 10.1016/j.puhe.2026.106409
Disclaimer: This content is for educational purposes only and is not medical advice. Evidence scores reflect the quality and quantity of available research, not clinical recommendations. Always consult a healthcare professional before starting any supplement or intervention.