Objectives: Differences in pain severity among acutely injured people may be related to the perceived stress of the event and pre-existing vulnerabilities. In this study, we test the hypotheses that pretrauma life stress influences posttrauma pain severity, and 2 potential mediating pathways, 1 biological (C-reactive protein, CRP) and 1 contextual (sleep quality).
Materials and methods: Data collected from participants within 3 weeks of a noncatastrophic musculoskeletal trauma were used in this observational cross-sectional mediation analysis. The primary outcome was pain severity as measured using the Brief Pain Inventory. Predictors were posttrauma CRP assayed from plasma, sleep interference measured by the Brief Pain Inventory, and a study-specific "General Life Stressors" scale. First, the sample was split into low and high life-stress groups, and mean differences in the pain and the predictor variables were explored by t test. Next, a mediation model was tested through a regression-based path analysis. The base model explored the predictive association between pretrauma life stress and posttrauma pain. Sleep quality and CRP concentration were then entered as possible mediators of the association.
Results: The sample of 112 participants was 54.6% female, and 52.7% reported high pretrauma life stress. Mean differences in pain severity, sleep interference, and CRP was significant between the high-stress and low-stress groups. In path analysis, life stress explained 8.0% of the variance in acute pain severity, 6.3% of the variance in sleep interference, and 8.0% of the variance in CRP concentration, all P-value <0.05. In mediation analysis, the association between life stress and pain severity was fully mediated by sleep interference. CRP did not mediate the association.
Discussion: Pretrauma life stress predicted pain severity, sleep interference, and plasma CRP. In mediation analysis, pretrauma stress was associated with pain severity only through its association with sleep interference, while CRP did not mediate the association. Implications of these results are discussed.