Abstract
Objectives: This study explored the relationship between sleep quality and next-day pain intensity for people with low back pain and investigated whether there was any evidence that this relationship was causal.Methods: We conducted a secondary analysis of an observational study that investigated sleep quality in people with low back pain. People with low back pain were recruited from primary care and the community. Sleep quality was measured with subjective (self-report) and objective (polysomnography (PSG)) measures. PSG analysis classifies sleep into stages, of which slow-wave sleep (SWS) is thought to have a key role in maintaining or increasing pain intensity. We drew directed acyclic graphs to identify possible confounders of the relationship between both measures of sleep quality, and pain intensity. We constructed two linear regression models to explore the effect of subjective and objective sleep quality on next-day pain intensity before and after confounder adjustment.Results: Thirty-nine participants were included in the study. For participants with low back pain, self-reported better quality sleep beta=-0.38 (95% CI -0.63 to -0.13), or spending a greater proportion of time in SWS beta=-0.12 (95% CI -0.22 to -0.02) was associated with lower next day pain intensity. After confounder adjustment, the effect reduced and was no longer significant for either self-reported beta=-0.18 (95% CI -0.46 to 0.10), or SWS beta=-0.08 (95% CI -0.18 to 0.03).Conclusions: Sleep quality, whether measured by self-report or proportion of time in SWS, was associated with next day pain intensity for people with low back pain. However, this relationship is likely to be confounded and therefore not likely to be causal.