Key points are not available for this paper at this time.
Chronic low back pain (CLBP) is prevalent. Understanding its progression and identifying related predictors is essential to guide its management. This study aimed to characterize CLBP trajectories over two years and identify baseline predictors of these trajectories. Data from 2713 participants enrolled in the Quebec Low Back Pain Study were analyzed. This cohort represents the analytic sample drawn from a larger cohort (n= 3894) where 30.3% (n = 1181) were excluded due to missing follow-up data. Baseline variables were collected using the Canadian adaptation of NIH Minimum Dataset. Pain intensity (0-10 scale) was assessed at 3, 6, 12, and 24 months after enrollment. Latent Class Growth Analysis (LCGA) was used to identify pain trajectories, and machine learning classifiers were trained to determine baseline predictors of trajectory membership. LCGA revealed moderate stable pain (40%), mild stable pain (30%), high persistent pain (16.4%), and gradually improving pain (14.6%) clusters. Most individuals with CLBP experienced stable pain, whereas gradually improving pain cluster showed clinically meaningful improvement (mean pain intensity: from 4.08 to 2.10 over two years). In all clusters key prognostic factors included pain intensity, pain interference, physical disability, and catastrophizing. Higher levels of these factors were associated with increased likelihood of membership in higher-pain trajectories, whereas lower levels were associated with lower-pain trajectories. Some factors were unique to each cluster; younger age, lower sleep disturbance, and shorter pain duration were associated with the gradually improving pain trajectory. These findings should be interpreted cautiously given the exclusion of participants with missing follow-ups. PERSPECTIVE: Our analyses of a large cohort of CLBP identified four trajectories: high persistent pain, moderate stable pain, mild stable pain, and gradually improving pain over a two-year period. The most influential factors associated with each trajectory help to better understand the characteristics associated with pain stability or improvement over time.
Rabiei et al. (Tue,) studied this question.