Association between musculoskeletal pain at multiple sites and objectively measured physical activity and work capacity: Results from UK Biobank study

Objectives To describe the cross-sectional association between musculoskeletal pain at multiple sites and physical work capacity (PWC) and objectively measured physical activity (PA). Design Observational study. Methods Data from a subsample of the UK Biobank were utilised (n = 9856; mean age 58.5 years, mean body mass index 30.2 kg/m2, 62% female). PWC was measured by a bicycle ergometer and PA by an accelerometer. Pain experienced in hip, knee, back and neck/shoulder was collected by questionnaire. Linear regression modelling was used with adjustment for potential confounders to estimate the association between pain and PWC and PA. Results Increase in number of painful sites was associated with lower PWC, moderate and vigorous PA and increased low intensity PA in a dose-response relationship (all p-values for trend ≤0.001) before and after adjustment for confounders. In site specific analyses, hip pain was associated with an increased low intensity PA (β 52.8 min/week, 95% CI 2.3–103.2) and reduced moderate PA (β −50.1 min/week, 95% CI −98.5 to −1.8). Knee pain was only associated with vigorous PA (β −5.7 min/week, 95% CI −10.0 to −1.3). Pain at neck/shoulder pain and back were not independently associated with PWC and PA. Conclusions Greater number of painful sites is consistently associated with poorer PWC, increased low intensity PA and reduced moderate to vigorous PA. Clinicians should address the critical role of being physically active in managing chronic musculoskeletal pain and interventions targeting musculoskeletal pain may be needed to increase PA levels