Supplements

Restless legs syndrome severity and neurological disability are the strongest predictors of poor sleep quality in patients with multiple sclerosis

In a cross-sectional study of 173 patients with MS, 48.0% were poor sleepers (PSQI > 5) with a mean PSQI of 6.06. XGBoost machine learning identified restless legs syndrome severity (IRLS) as the strongest predictor of PSQI, followed by EDSS (disability) and depression, while cardiometabolic and inflammatory factors contributed inconsistently.

Last updated: Aug 28, 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 Prediction of Sleep Quality in Patients with Multiple Sclerosis.

observational

Cucu LE, Baciu LC, Onicescu OM, Ignat BE, Săcărescu A, Oancea A, Grosu C, Chirica C, Popescu G, Maștaleru A, Bîlcu RV, Mustață A, Roca M, Leon MM · Medical sciences (Basel, Switzerland) (2026)

Participants: N/A
Duration: N/A (cross-sectional)
Intervention: No intervention; observational study using machine learning to predict sleep quality from 28 clinical, cardiometabolic, and inflammatory predictors
Outcome: Sleep quality assessed by PSQI global score
Effect Size: N/A (test R2 = 0.451 for XGBoost)
Population: Adult patients with multiple sclerosis (mean age 39.66 years, 69.9% female, 90.2% RRMS)

Result:

PubMed: 42646617DOI: 10.3390/medsci14040483

Mechanism Graph

Restless legs syndrome severity (IRLS) is the strongest predictor of PSQI
↓
Neurological disability (EDSS) and depression also contribute significantly
↓
Cardiometabolic and inflammatory factors show inconsistent and sometimes opposite effects
↓
Disease-related and symptomatic factors outweigh cardiometabolic and inflammatory contributions

Limitations

  • ⚠Cross-sectional design prevents causal inference
  • ⚠Modest sample size (n=173) and single-center study
  • ⚠Exploratory machine learning approach requires confirmation in larger independent cohorts

Frequently Asked Questions

What is the most important predictor of poor sleep in MS patients?▼

Restless legs syndrome severity (IRLS) was the strongest predictor of PSQI across all machine learning models.

How common is poor sleep in this MS cohort?▼

48.0% of the 173 patients were classified as poor sleepers (PSQI > 5), with a mean PSQI of 6.06.

Do cardiometabolic and inflammatory factors predict sleep quality in MS?▼

They contributed inconsistently and some showed effects opposite to physiological expectation, suggesting they are less important than disease-related and symptomatic factors.

What machine learning model performed best?▼

XGBoost achieved the best predictive performance with a test R2 of 0.451.

Products

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

References

  1. 1.Cucu LE, Baciu LC, Onicescu OM, Ignat BE, Săcărescu A, Oancea A, Grosu C, Chirica C, Popescu G, Maștaleru A, Bîlcu RV, Mustață A, Roca M, Leon MM. "Machine Learning-Based Prediction of Sleep Quality in Patients with Multiple Sclerosis.." Medical sciences (Basel, Switzerland), 2026. PMID: 42646617 DOI: 10.3390/medsci14040483
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.