Lower sleep quality predicts one-year low back pain incidence in healthy adults
A supervised machine learning model identified lower sleep quality as one of the most important predictors of one-year low back pain incidence among 156 healthy participants. The model achieved an accuracy of 0.815 and an ROC AUC of 0.839 for out-of-sample predictions.
Evidence Score
Study Evidence
Study 1. Exploring the role of pain-related fear and lifting biomechanics in predicting low back pain incidence using supervised machine learning.
observationalBangerter C, Faude O, Dörig M, Meier ML, Hasler CC, Schmid S ยท Scientific reports (2026)
Result:
Mechanism Graph
Limitations
- โ Observational design cannot establish causality between sleep quality and LBP incidence
- โ Sample size of 156 is relatively small for machine learning models, potentially limiting generalizability
Frequently Asked Questions
How was sleep quality measured in this study?โผ
Sleep quality was assessed at baseline using a self-report questionnaire, though the specific instrument is not detailed in the abstract.
What other factors predicted low back pain besides sleep quality?โผ
Higher BMI, greater lifting-specific pain-related fear, and reduced lumbar spine range of motion were also among the most important predictors.
Can improving sleep quality reduce the risk of low back pain?โผ
This study only shows an association, not causation. Further interventional research is needed to determine if improving sleep quality lowers LBP risk.
What machine learning model was used?โผ
The researchers used an Explainable Boosting Machine (EBM) to predict one-year LBP incidence from baseline variables.
Products
Affiliate links coming soon. We only recommend products that match the doses and forms used in the cited research.
References
- 1.Bangerter C, Faude O, Dörig M, Meier ML, Hasler CC, Schmid S. "Exploring the role of pain-related fear and lifting biomechanics in predicting low back pain incidence using supervised machine learning.." Scientific reports, 2026. PMID: 42481619 DOI: 10.1038/s41598-026-62770-2