P174 Identifying patterns of fatigue across flares in patients with axial spondyloarthritis

Abstract Background/Aims Fatigue is a common and debilitating symptom of axial spondyloarthritis and has a significant negative impact on quality of life for these patients. Previous qualitative and quantitative research indicates that fatigue significantly worsens during flares. There remains uncertainty around the quality and intensity of fatigue before a flare compared with during and after. This understanding could help in diminishing the unpredictability of flares and fatigue which has been identified as a key issue for patients and may aid self-management. The main aim of the current analysis was to determine whether there would be any significant differences in the distribution of fatigue scores depending on patients’ self-reported flare status. Methods The current project involved secondary analysis of patient data collected through the Royal National Hospital for Rheumatic Diseases. A Kruskal Wallis test was conducted utilising anonymised Project Nightingale data from 286 patients collected via a mobile app. Data from a single time point for each patient were analysed. Fatigue was measured via responses to a 5-point scale. Patients’ data were organised into five categories depending on their self-reported flare status. This included flare started, flare continued, flare stopped and no flare. A fifth group was created where patients had reported their fatigue scores the day before a flare started. Results 43% percent of patients were female and 57% percent were male. There was at least one significant difference in rank totals between categories, H (4, n = 286) = 48.38, p < .001. Dunn’s post hoc tests were subsequently performed with Bonferroni adjusted p values for pairwise comparisons and are displayed in the table below. Fatigue was significantly higher in the flare continued category than in the no flare (p <.001), day before flare started (p = .008) and the flare stopped categories (p = .028). Inspection of medians and interquartile ranges suggested that fatigue appeared to be slightly higher the day before a flare started compared with no flare, however there were no other statistically significant differences between categories. Conclusion The results indicate that fatigue was highest during the middle of a flare as opposed to the start or end of a flare or the day before a flare. Future studies could include data at more time points including prior to, during and after flares. As previous literature implies that there may be two types of flare associated with different levels of fatigue, future studies could compare levels of fatigue between patients within these subgroups. This research could facilitate the refinement of interventions with the aim of minimising the unpredictability and impact of fatigue in axSpA patients. Disclosure S. Patel: None. R. Barnett: None. C. Cavill: None. R. Sengupta: Honoraria; R.S. has received honoraria from Abbvie, Biogen, BMS, Eli Lilly, MSD, Novartis, Pfizer, Roche and UCB, all unrelated to this abstract.