293: Protection against pain: Examination of a positive pain trajectory in The Raine Study.

Topic: Clinical Science
Subtopic: Non-specific to any diagnosis, Psychology, Artificial Intelligence/Machine Learning, Epidemiology

Background and Aims: Pain trajectory research has focused on negative trajectories despite evidence of trajectories with positive prognoses and associated protective factors.

The Raine Study (www.rainestudy.org.au) involves participants born between 1989 and 1992 in Western Australia (Gen2). While positive pain trajectories exist in the Raine Study Gen 2, these have not been a research focus. Multidimensional protective variables (e.g. marital status, adiposity, sleep) have been collected from Raine Study Gen2 participants at multiple timepoints to 28 years old.

This study aimed to derive positive musculoskeletal pain trajectories and determine associated protective factors in the Raine Study Gen2.

Methods: Participants: The Raine Study Gen 2 (n=1069).

Musculoskeletal pain intensity measured at 17-, 22-and 27-years was entered into group-based trajectory modelling. Of three derived trajectories, one (n=457, 42.8%) reported minimal pain (<1/10) over time.

Protective variables across multiple dimensions (Sociodemographics (e.g. relationship status), education and work (e.g. work colleague support), psychosocial (e.g. pain coping), health/lifestyle (e.g. physical activity), diet (e.g. coffee intake), physical examination/blood test (e.g. squat endurance, c-reactive protein)), measured at multiple timepoints (14-, 17-, 18-, 20-, 22-, 27-, 28-year follow-up) were modelled using machine learning (Light-GBM) to predict membership of the minimal pain trajectory.

Results: A Light GBM model predicting minimal pain trajectory membership was derived (ROC AUC .86, precision .80, recall .68, F1-Score .74, R²: .28). Protective variables (timepoint) contributing most to the model were: 11-20% of work time sitting (27 years), very low perceived risk of pain persistence (27 years), high perceived ability to decrease pain (27 years), activities of daily living (walking, light work, shopping, household chores) without pain interference (22 years) and no time off work due to pain (22 years). Smaller contributions were derived from: socioeconomic indices, resting heart rate, forced expiratory volume, vegetable intake and 37 other variables.

Conclusions: This study suggests protection against musculoskeletal pain is complex, emergent and multidimensional, involving protective factors across sociodemographic, work, psychosocial, health/lifestyle, diet and physical dimensions. Comparisons across different machine learning models and different longitudinal cohorts are underway to deepen our understanding of positive pain trajectories and associated protective factors.

References: Aili K, Campbell P, Michaleff ZA, Strauss VY, Jordan KP, Bremander A, et al. Long-term trajectories of chronic musculoskeletal pain: a 21-year prospective cohort latent class analysis. PAIN. 2021;162(5).

Bartley EJ, Palit S, Fillingim RB, Robinson ME. Multisystem Resiliency as a Predictor of Physical and Psychological Functioning in Older Adults With Chronic Low Back Pain. Front Psychol. 2019;10:1932.

Bartley EJ, Makhoul M, Palit S, Robinson ME, Fillingim RB. Examining Physical and Cognitive Function in Chronic Low Back Pain using a Multisystem Resilience Framework. Pain Med. 2022.

Beynon AM, Hebert JJ, Beales DJ, Jacques A, Walker BF. Multi-trajectory analysis of C-reactive protein and low back pain from adolescence to early adulthood. Eur Spine J. 2021;30(4):1028-34.

Coenen P, Smith A, Paananen M, O'Sullivan P, Beales D, Straker L. Trajectories of Low Back Pain From Adolescence to Young Adulthood. Arthritis Care & Research. 2017;69(3):403-12.

Presenting Author: Rabey
Submitting Author: Martin Rabey

Senior Author: Rabey

IASP 2026 Global Year theme of Neuropathic Pain: No

• Anna Marcuzzi: PhD - Norwegian University of Science and Technology
• Phillip Melton: PhD - University of Tasmania
• Darren Beales: PhD - Curtin University
• Niamh Moloney: PhD - Curtin University
• Matthew Bagg: PhD - University of Notre Dame