Feelings of fatigue, annoyance, anger and hostility are predictive for reductions in pain-related disability following a multidisciplinary rehabilitation program for chronic musculoskeletal pain

Background and aims: Chronic musculoskeletal pain is a major health care problem worldwide that comes together with physical and emotional burdens, resulting in substantial financial costs for society. Several treatment options are known for patients with chronic pain. Knowledge about predictive factors for therapy response is essential to refer the patient directly to the most efficient treatment program. This retrospective observational study strives to identify predictive factors for the change in disability after an interdisciplinary rehabilitation program. Methods: A dataset of 653 patients with chronic musculoskeletal pain who attended either a 10 or a 20-week interdisciplinary rehabilitation program at the Centre for Integral Rehabilitation (CIR) in four different outpatient care centers, in the Netherlands was analyzed. The therapy was personalized and consisted of pain neuroscience education, lifestyle education, coping strategies, cognitive behavioral therapy, relaxation therapy, acceptance and commitment therapy, graded activity, mental coaching and physical coaching, according to the needs of the patient. To assess the effect of the therapy on disability, the TampaPain Disability Index (PDI) was chosen as the outcome. The PDI was conducted at the screening (T0), after 10 weeks (T1), and after 20 weeks (T2) if the therapy continued. Questionnaires conducted at the screening were assessed as possible predictive factors: Tampa Scale of Kinesiophobia (TSK) was used to assess fear of movement, Checklist Individual Strength (CIS) examined different aspects of fatigue, Beck Depression Inventory (BDI) evaluated different symptoms of depression, Utrecht’s Coping List (UCL) assessed the different coping strategies used by the participants and Symptom Checklist-90 (SCL-90) evaluated physical and psychological symptoms. The specific subcategories of the UCL and SCL-90 were also assessed individually. A generalized linear mixed model (GLMM) was used with the PDI as the dependent variable over time (screening, 10-week, 20-week). The model controlled for age, gender, treatment location and pain medication. The baseline questionnaires were separately added to the GLMM to analyze their possible predictive value. Results: The mean age of the population was 44.6 (± 11.9) years and consisted for 73% out of women. Most patients were treated in Amsterdam (28.9%). Pain medication was taken by 64.2%. No predictive value was found for the TSK, BDI, total UCL and total SCL-90 score at screening. However, significant interaction effects with ‘Time’ were found for the CIS (p<.05), subcategory ‘Expression of emotion’ of the UCL (p<.01) and subcategory ‘Hostility’ of the SCL-90 (p<.05). An increase in the CIS score of one point at T0, resulted in a greater decrease of the PDI score at T2 with 0.08 (95% CI = 0.01 – 0.14) compared with T0. Furthermore, an increase of one point in the subcategory ‘Expression of emotion’ of the UCL at T0 resulted in a greater decrease of the PDI score at T2 with 1.10 (95% CI = 0.34 – 1.86) compared with T0. Lastly, the PDI score decreased by 0.50 points (95% CI = 0.10 – 0.90) more at T2 compared to T0, for every one point increment in the subcategory ‘Hostility’ of the SCL-90 at T0. It is noteworthy that the change in PDI from screening towards the end of the therapy exceeded the minimal clinically important change value. Conclusions: Patients who experience more severe fatigue, who demonstrate more annoyance or anger, and who show more signs of hostility prior to starting treatment are more likely to experience improvement in pain disability after an interdisciplinary treatment program