A multilevel path model analysis of the relations between sleep, pain, and pain catastrophizing in chronic pain rehabilitation patients

Background and aim: Pain catastrophizing is linked to heightened pain and poorer coping among individuals with chronic pain, yet little is known about how pain catastrophizing associates with sleep and pain over the course of treatment for chronic pain. Previous research employing a cross-sectional design suggests that sleep mediates the association between pain catstrophizing and pain, but there have been no longitudinal studies examining the directionality of these associations. Thus, the aim of this study was to test two competing theoretical models. The first model specified that pain catastrophizing leads to increased pain via poor sleep. The second model specified that poor sleep leads to increased pain catastrophizing via increased pain.

Methods: This study examined the relations between pain catastrophizing, sleep, and pain among 50 consecutive patients (36 female, 14 male) ages 20-80 (M=45.96, SD=13.94) with chronic, non-malignant pain who were admitted to the Cleveland Clinic, Chronic Pain Rehabilitation Programme (CPRP). The CPRP, within the Neurological Centre for Restoration, Neurologic Institute at the Cleveland Clinic, is a comprehensive, interdisciplinary programme designed to treat patients with disabling chronic pain. As part of their daily, morning update with their case manager, patients completed self-report ratings of their previous night's sleep time (TST), and their current pain, anxiety, and depression. Pain catastrophizing was assessed at admission and discharge.

Results: Over the course of treatment, daily TST increased from approximately 5h and 20min per night to nearly 6h and 30min per night, and average daily pain, daily depression, and daily anxiety decreased over the course of treatment. As the data in this study has a multilevel structure, with daily reports nested with in patients, we conducted multilevel path models to examine the longitudinal relations between pain catastrophizing, sleep, and pain. Multilevel path analysis permits the analysis of interdependent data without violating the assumptions of standard multiple regression. Models were conducted for pain catastrophizing and each of its subscales: rumination, magnification and helplessness. The findings were uniform across the composite pain catastrophizing scale and its subscales. There was an indirect path from sleep to pain catastrophizing (post-treatment) via pain, but not from pain catastrophizing (pre-treatment) to pain via sleep. There were also direct effects of sleep on pain and from pain to pain catastrophizing (post-treatment). Additionally, decreases in pain over the course of treatment were related to lower pain catastrophizing post-treatment.

Conclusion and implications: These results call into question previous evidence that pain catastrophizing indirectly affects pain by way of its impact on sleep. Rather, our findings suggest that pain mediates the relationship between sleep and levels of pain catastrophizing. These results therefore underscore importance and value in collecting longitudinal data and potential influence on the conclusions gained with regards to sleep, pain and psychological variables. These findings may be of clinical importance when tailoring interventions for individuals with chronic pain and perhaps even more so for those with comorbid pain and sleep disturbance; prioritizing the treatment of sleep difficulties could result in improvements to pain-related outcomes.