OBJECTIVE: There is debate as to the role of inflammatory disease activity in the etiology of rheumatoid arthritis (RA)-related fatigue. We undertook this study to determine the relationship of fatigue to disease activity by examining pathways associated with change in fatigue in subjects starting anti-tumor necrosis factor (anti-TNF) therapy for the first time.
METHODS: Participants who had been recruited to the British Society for Rheumatology Biologics Register for RA provided information on fatigue (Short Form 36 [SF-36] vitality scale) and other health status variables at the start of anti-TNF therapy and 6 months later. The Disease Activity Score in 28 joints (DAS28) and inflammation (erythrocyte sedimentation rate [ESR]/C-reactive protein [CRP] level) were also reported. A path analysis model comprising changes in fatigue, pain, disease activity, disability, and mental health, along with effects of sex and a history of depression, was used to examine those with high levels of fatigue at baseline (score of ≤12.5 units on the SF-36 vitality scale). The DAS28 was substituted for ESR/CRP to delineate the specific role of inflammation.
RESULTS: With a total of 2,652 participants, we identified a well-fitting model (χ2 = 0.18, P = 0.98) accounting for 40% of the variance in fatigue change. There was no direct pathway from change in inflammation to change in fatigue; instead, significant pathways to change in fatigue were observed from changes in disease activity, pain, mental health, and disability, along with effects of sex and a history of depression. A total of 82% of the effect of change in disease activity was indirect, of which ∼50% was mediated through a change in pain.
CONCLUSION: Improvements in fatigue do not appear to be driven by inflammatory disease activity; instead, they appear to result indirectly from improvements in pain. Additional significant pathways through disability and mental health suggest potentially modifiable factors that could be targeted to improve clinically important levels of fatigue.
