| Abstract: |
Rationale: Relationship dissatisfaction has been linked with worse health outcomes in many patient populations, though the mechanism(s) underlying this effect are unclear. Among patients with chronic fatigue syndrome (CFS) and their partners, there is evidence for a bi-directional association between poorer relationship satisfaction and the severity of CFS-related fatigue. Objective: Here, we hypothesized that relationship dissatisfaction negatively impacts fatigue severity through greater depression and less patient satisfaction about communication about symptoms to partners. Method: Baseline data were drawn from diagnosed CFS patients (N = 150) participating in a trial testing the efficacy of a stress management intervention. Data derived from fatigue severity (Fatigue Symptom Index, FSI), depression (Center for Epidemiologic Survey-Depression, CES-D), relationship quality (Dyadic Adjustment Scale, DAS) and communication satisfaction (Patient Symptom Disclosure Satisfaction, PSDS) questionnaires were used for bootstrapped indirect effect analyses using parallel mediation structural equation modeling in Mplus (v8). Age and BMI were entered as covariates. Results: Greater relationship satisfaction predicted greater communication satisfaction (p < 0.01) and lower CES-D scores (p < 0.01), which in turn were each significantly related to greater fatigue severity (p < 0.05). Tests of the indirect paths indicated that relationship satisfaction had a significant effect on fatigue severity through both constructs, but primarily via depression. There was no direct association between relationship satisfaction and fatigue severity after the intermediate variables (depression, communication satisfaction) were included in the model. Conclusion: Results highlight the importance of considering depression and communication-related factors when examining the effects of relationship satisfaction on CFS symptoms such as fatigue. Further mechanism-based, longitudinal research might identify relationship-related mediating variables that can be targeted therapeutically. © 2019 |
