Effect Modifiers of the Association Between Physical Activities and Subsequent Low Back Pain Flares.

OBJECTIVE: This study evaluated whether concurrent actions, work-relatedness, and analgesic use modify the associations between physical activities and subsequent LBP flares.

SUMMARY OF BACKGROUND DATA: Prior longitudinal case-crossover studies found transient risks of various activities on LBP flares. These studies have focused on the main effects of exposures for LBP. However, little is known about contextual factors that could modify activity-flare relationships.

METHODS: This was a secondary analysis of a 1-year prospective case-crossover study evaluating transient risks of 10 common physical activities on LBP flares. Participants completed scheduled surveys and self-initiated flare surveys inquiring about potential modifiers and activity exposures. Multivariate-adjusted conditional logistic regression analyses evaluated various concurrent actions, work-relatedness, and analgesic use as potential modifiers of the association between each activity and subsequent flares.

RESULTS: Of 416 participants, the mean age was 47.6 years and 77% had LBP duration >5 years. Many concurrent actions (15 of 21 candidate effect modifiers) were significantly associated with greater activity-flare associations, including but not limited to twisting (odds ratio [OR]=1.48, [95% confidence interval [CI] 1.23-1.78]), bending (OR=1.32 [95% CI 1.10-1.59]), lifting weights of 20-29 pounds (OR=1.70 [95% CI 1.30-2.23]), and vibration exposure (OR=2.00 [95% CI 1.24-3.21]). Work-relatedness was associated with lower flare risk for 3 out of 10 activities (lifting, climbing, and squatting), but these findings were not directionally consistent in secondary analyses using 2-hour exposure windows, suggesting that work-relatedness does not modify activity-flare associations. Analgesic use did not meaningfully modify activity-flare associations.

CONCLUSIONS: Concurrent biomechanical actions were generally associated with greater activity-flare risk. Work-relatedness and analgesic administration did not modify activity-flare associations. These findings provide the first epidemiologic evidence that contextual factors, and not solely the type of activity, may influence activity-associated LBP flare risk for some activities.

STUDY DESIGN: Prospective case-crossover study.