Exploring the Biopsychosocial Impact of Chronic Low Back Pain in Workers Through Artificial Intelligence-Driven
Paolo Giaccone1,2, Elisabetta de Rinaldis3, Federico D'Antoni1,2
1Fondazione Policlinico Universitario Campus Bio-Medico Rome Italy.
Study Design:
Cross-sectional retrospective cohort study.
Objectives:
Chronic low back pain (cLBP) is a major cause of disability worldwide, significantly affecting return to work (RTW). This study aimed to assess the biopsychosocial factors influencing occupational functioning in workers with cLBP using artificial intelligence (AI) data-driven patient phenotyping.
Methods:
Data of workers affected by cLBP were collected through a comprehensive assessment of demographic, clinical, and occupational factors. Hierarchical clustering was employed to identify distinct phenotypes based on patient-reported outcome measures (PROMs), including the Oswestry Disability Index (ODI), Visual Analog Scale (VAS), Work Ability Index (WAI), Nordic score, and Patient Health Questionnaire-2 (PHQ-2). Independent t tests and Mann-Whitney U tests were used for phenotype profiling, distinguishing between continuous and categorical responses, respectively, to assess the most discriminative queries and highlight the most significantly different features (p < 0.05).
Results:
A total of 304 patients were included in the analysis. The AI-driven phenotyping approach identified two distinct clusters, representing 51% (Cluster 1) and 49% (Cluster 2) of the dataset. Compared to Cluster 1, Cluster 2 exhibited significantly higher absenteeism (17.00 vs. 5.22 days, p < 0.05), lower WAI (33.34 ± 6.84 vs. 38.96 ± 4.31, p < 0.05), worse pain-related outcomes in terms of higher VAS (5.98 ± 2.06 vs. 4.48 ± 2.48, p < 0.05) and ODI (33.52 ± 16.56 vs. 20.08 ± 13.59, p < 0.05), more frequent occupational exposure to manual handling of loads (84% vs. 16%) and higher psychological distress assessed through PHQ-2 (70% vs. 30%).
Conclusion:
Our study identified the most relevant PROMs differentiating between cLBP clusters of patients, emphasizing different levels of absenteeism and pain-related outcomes.These findings contributed to unravel the data-driven AI potential in suggesting personalized interventions targeting specific biopsychosocial profiles, which may improve clinical outcomes and occupational functioning in workers with cLBP, ultimately enhancing their overall well-being.
