Janan Abbas

Department of Physical Therapy, Zefat Academic College, Israel

Abstract Title: Are health science undergraduates associated with low back pain? A 2-year prospective study with machine learning analysis

Biography: Dr. Janan Abbas Dr. Janan Abbas is a senior lecturer at the Physical Therapy Department at Zefat Academic College in Israel, whose research focuses on spine-related disorders. Her scope of expertise includes spinal stenosis and lumbar spine disorders such as low back pain. She has also published on predictive modelling for degenerative lumbar spinal stenosis/ or LBP among undergraduates, applying special techniques such as a machine learning approach.

Research Interest: Background. Low back pain (LBP) is the most common and challenging disorder in health care. Growing evidence suggests that student lifestyles, particularly those in health science programs, lead to a higher prevalence of LBP than their counterpart in the general population. Machine learning (ML) is continually gaining importance in medical predictive analytics. It could also enhance the ability to detect patterns of clinical characteristics in LBP; however, longitudinal studies are limited. Aims. This study aims to investigate the factors contributing to LBP among health science students using the ML approach. Methods. A 2-year prospective study of 197 first-year health-science students (e.g., physiotherapy and nursing) was recruited in May 2022 (baseline period) at Zefat Academic College in Israel. A self-administered Standardized Nordic Questionnaire was used. Data about the presence of 1-month LBP were also recorded. A supervised random forest algorithm of ML was utilized to analyze data and prioritize the importance of variables related to LBP (baseline and follow-up). Results. 197 students (88.7%) responded and completed the questionnaire (baseline) with mean age and body mass index values of 23 (years) ± 3.8 and 23 (kg/m2) ±3.5, respectively. 46% of the students have a 1-month LBP, and 74% experienced LBP at some point in their lives. The decision tree revealed that a history of LBP, disability and physical activity, were significantly associated with LBP in the baseline. Surprisingly, history of LBP remains the significant variable associated with LBP after 2 years of follow-up. Conclusion: This study indicates that the history of pain is the primary variable that significantly associates with LBP among health science undergraduates. This result could also strengthen the reliability and consistency of the ML analysis for predicting LBP. Keywords. LBP, machine learning, students, physical activity.