Associations between smoking history, baseline pain interference and symptom distribution, and physical function at discharge, in individuals seeking care for musculoskeletal pain

Objective: Study of the association between smoking and pain intensity has produced conflicting results; with less focus on pain interference. Different pain constructs could have varying associations with smoking behaviors. This study sought to investigate the association between smoking history and not only pain intensity, but also pain interference, symptom distribution and physical function. Methods: Smoking history (current, past, or none), pain interference (Pain, Enjoyment of Life, and General Activity scale), symptom distribution and physical function scores were extracted from medical records of patients seen in physical therapy for common sites of musculoskeletal pain (lumbar and cervical spine, knee, or hip). Generalized linear models assessed the relationship between smoking history and pain/function. Results: 833 patients from an integrated healthcare system were included (mean: 57.6 years, SD=16.3; 43 % male). After controlling for several variables, current smokers had significantly higher baseline pain interference scores compared to never and former smokers (beta [B]: 0.65, 95 %CI: 0.13 to 1.18, P=.02). Smoking was not a significant predictor of symptom distribution at baseline [B: 0.17, 95 %CI −0.06 to 0.42, P=.16] or physical function scores at discharge [B: -0.03, 95 %CI: −0.08 to 0.02, P=.25]. Conclusion: Smokers experienced a greater impact of pain at baseline. However, symptom distribution at intake and function upon discharge were similar between all smoking groups. These findings suggest smoking cessation and abstinence may be important recommendations to help curb pain interference.