Search

Background: It is estimated that over 62% of people with a spinal cord injury (SCI) experience chronic pain, (Ullrich, Jensen, Loesser & Cardenas, 2007). Much research has demonstrated that a variety of biopsychosocial factors can impact on pain outcomes (Tran, Dorstyn & Burke, 2016) and, consequently on adjustment to injury. It is also well established that cognitive appraisal of SCI impacts on psychological adjustment during rehabilitation (Eaton, Jones & Duff, 2018). SCI pain is unusually resistant to standard pain management programmes (Perry, Nicholas & Middleton, 2010). However, the development of a tailored programme requires a profile of the biological, psychological, and social characteristics of chronic pain sufferers with SCI, but the existing knowledge base is fragmented. This study aimed to investigate how biopsychosocial factors interact to impact on pain-related outcomes for people with SCI. Method: A longitudinal, multiple assessment-point design was used with 60 spinal cord injured in-patients at the NSIC, Stoke Mandeville. Participants were asked to complete a set of two pain and six psychological assessments at three different time points over a nine-month period, and to provide salivary samples on each occasion to assess concentration levels of cortisol. Additionally, a cross-sectional study using the same questionnaires and cortisol sampling was undertaken with 47 out-patients, who had been discharged a minimum of two years previously from the NSIC. Cohen's (2009) power primer was used to calculate sample size. Independent t-tests measured differences between in-patient and out-patient groups on each questionnaire. Multiple regression was used to determine which biopsychosocial factors have greater predictive power in accounting for a range of functional, affective and sensory pain outcomes, highlighting how variables may individually and in combination influence the pain experience. Repeated measures ANOVAs were used to assess how the data changed over time and each measure was additionally correlated with time since injury. Additional exploratory analyses were undertaken to see whether pain catastrophising, appraisal of injury and pain acceptance mediated the effects of the other biopsychosocial variables on the pain outcomes. Lastly, multiple regressions explored whether the psychosocial variables predicted the way in which the injury was appraised. Results: Out-patients appraised their injury more negatively (p = .04) and had lower determined resilience (p = .05) than in-patients at time one. They also demonstrated less pain acceptance (p = .04) and received fewer solicitous responses from a significant other person (p = .01). In-patients at time three had higher depression scores than out-patients (p = .006). In the multiple regression analyses, negative psychological variables predicted pain intensity (p = .002-.005), interference from pain (p = .001), and pain-related distress (p = .001). Positive psychological variables did not predict pain intensity but did predict pain interference (p = .029 - .071) and distress (p < .001). The way a significant other responded to the individual in pain did not predict pain intensity but did predict life interference (p = .001) and distress (p = .018 - .049). Cortisol did not predict any of the pain outcomes directly. Of all the variables, cortisol concentration was only significantly related to pain catastrophising (p = .008). The in-patient longitudinal analysis showed that over the three time points determined resilience decreased (p < .001), and depression scores increased (p = .025). The magnification sub scale of pain catastrophizing also increased between the first and second time point (p = .021). Time since injury was positively correlated with mental defeat (p = .047) and the helplessness sub scale of pain catastrophizing (p = .010), and negatively correlated with cortisol concentration levels (p = .001). In the mediation analyses, pain catastrophizing and appraisal of injury mediated the effects of most of the biopsychosocial variables on a wide range of pain outcomes. Pain catastrophizing was most influential on sensory and functional pain outcomes, and injury appraisal had greater effects on affective and functional outcomes. Pain acceptance was not influential as a mediating variable. In the final multiple regression analyses, the psychosocial variables were entered into regression models to see if they would predict the way the spinal injury was appraised. The psychological variables model (catastrophizing, acceptance, perceived stress, anxiety and mental defeat) predicted catastrophic negativity (p < .001) and determined resilience (p < .001). The way a significant person responded to the individual in pain did not predict catastrophic negativity with regard to injury appraisal but did predict determined resilience (p = .001). Conclusion: The results of this study clearly indicate that biopsychosocial variables combine and interact to affect the consequences of pain for people with spinal cord injury. Pain treatment programmes that fail to take account of each of the components of the biopsychosocial model will not be addressing all of the factors associated with the pain experience, and this will have a negative impact for those in pain. This is especially concerning as the study found that psychosocial variables worsen on transition to the community. Appraisal of injury and pain catastrophizing are particularly influential, both as predictors and mediators, so focusing on these factors in pain management could improve pain outcomes and injury adjustment for people with spinal cord injury.