Low Back Pain Among Health Sciences Undergraduates: Results Obtained from a Machine-Learning Analysis.

Area of Science:

• Public Health
• Biomedical Informatics
• Epidemiology

Background:

• Low back pain (LBP) is a prevalent and challenging health issue, with incidence increasing with age.
• Sedentary behavior among students is a potential contributing factor to LBP risk.
• Machine learning (ML) offers advanced analytical capabilities for predicting and preventing medical conditions like LBP.

Purpose of the Study:

• To identify factors associated with low back pain (LBP) among health sciences students.
• To utilize machine learning for analyzing complex health data patterns.
• To enable targeted prevention strategies for LBP in student populations.

Main Methods:

• A modified Standardized Nordic Questionnaire was administered to 222 freshman health sciences students.
• A supervised random forest algorithm was employed to analyze data and rank variable importance.
• A decision tree visualized the ML model's predictive power, identifying high-risk patterns.

Main Results:

• 46% of participating students reported experiencing LBP in the past month.
• Key factors significantly associated with LBP included a history of pain (score=1), disability (score=0.34), and physical activity levels (score=0.21).
• Prolonged sitting (over 3 hours daily) was reported by 60% of students.

Conclusions:

• Nearly half of health science students experience LBP, highlighting a significant public health concern.
• Machine learning analysis identified a history of pain as the most critical factor linked to LBP.
• Further research and interventions are warranted to address LBP in student populations.