Psychological factors, sickness behavior and inflammatory biomarkers in longstanding pain

Abstract

Background: Longstanding pain affects a large number of adults worldwide. In addition to pain, several factors like depression, anxiety, and insomnia can also affect disability and quality of life. Some patients improve following psychological treatment with regards to symptoms, functioning, and quality of life, but there is considerable variation in outcome. Furthermore, it is not fully known for whom or why treatment is effective. Sickness behavior and inflammation are possible factors to consider in the maintenance of longstanding pain, and these factors may influence treatment outcome.

Aims: The specific aims of this thesis were to: Evaluate aspects of construct validity and reliability of the SicknessQ in patients with longstanding pain (Study I); Investigate how sickness behavior in patients with chronic conditions (chronic pain and ME/CFS) differed from that in participants with experimental acute sickness, primary care patients, the general population and healthy controls. In addition, to explore how sickness behavior was related to self-rated health and health-related functioning (Study II); Investigate if levels of depression, anxiety, insomnia, pain intensity, self-rated health, and sickness behavior were related to low-grade inflammatory biomarkers (Study III); Investigate if low-grade inflammation affected the outcome of ACT concerning maximum pain intensity, psychological inflexibility as well as pain interference and whether there were changes in ongoing inflammatory activity following ACT (Study IV).

Methods: Study I. Construct validity of the SicknessQ was evaluated by performing a confirmatory factor analysis (CFA) and by hypothesis-driven analyzes. Reliability was evaluated by analyzing the internal consistency of items. Study II. Correlations and linear regression analyzes were used to investigate associations between sickness behavior, self-rated health, and health-related functioning. Study III. Associations between the factors described in the aims were analyzed using bivariate Spearman rank correlation coefficients and regression analyzes. ANOVA was performed to investigate potential differences between subgroups. Study IV. The treatment effects and moderating effects of IL-6 and TNF-α on alterations in outcomes were analyzed using linear mixed models.

Results: Study I. In the final CFA, the Chi-Square test was not significant (χ2 [32, N = 190] = 42.95, p = 0.094), indicating a perfect model fit for the one-factor model. Internal consistency was adequate, as indicated by a Cronbach's α value of 0.82 for the entire questionnaire. Study II. Patients with chronic pain (M = 16.1), patients with ME/CFS (M = 16.1), LPS-injected individuals (M = 16.3), and primary care patients (M = 10.7) reported significantly higher SicknessQ scores than individuals from the general population (M = 5.4) and healthy controls (M = 3.6), all p’s < 0.001. Higher levels of sickness behavior were significantly associated with poorer self-rated health and health-related functioning (p’s < 0.01) in the general population and chronic pain sample, but not significantly in the ME/CFS sample. Study III. There were significant correlations between insomnia and hsCRP (p < 0.05); sex and ESR (p < 0.05); age and IL-6 (p < 0.05) and IL-8 (p < 0.05); BMI and IL-6 (p < 0.001), hsCRP (p < 0.001) and ESR (p < 0.001). Sickness behavior and anxiety (p < 0.05 and p < 0.001, respectively) contributed significantly, explaining 49% of the total variance in depression. Similarly, sickness behavior (p < 0.05) contributed significantly, explaining 34 % of the total variance in insomnia. Inflammatory biomarkers, however, did not contribute significantly to the models. There were significant differences between subgroups of depression regarding age, self-rated health, anxiety, insomnia, and sickness behavior (p < 0.001, respectively) as well as hsCRP (p < 0.05). In subgroups of insomnia, there was a significant difference in BMI, pain intensity, self-rated health, anxiety, and IL-6 (p < 0.05, respectively) as well as depression and sickness behavior (p < 0.001, respectively). Study IV. Mean baseline levels of IL-6 and TNF-α tentatively moderated improvement in psychological inflexibility during treatment (p = 0.044), but not in pain interference (p = 0.205) or pain intensity (p = 0.536). Cytokine levels did not change over the course of the treatment (IL-6/TNF-α p = 0.086/0.672).

Conclusion: The results indicated that the SicknessQ is a brief questionnaire with reliable and valid statistical properties to assess sickness behavior in adults with longstanding pain. Patients with chronic pain and ME/CFS reported similarly high levels of sickness behavior, higher than primary care patients, and comparable to levels in experimental inflammation. Participants rated a relatively high symptom burden, but the included symptom variables had weak associations with the included inflammatory biomarkers. Higher levels of baseline inflammatory biomarkers (IL-6 and TNF-α) were related to less improvement in psychological inflexibility.