OA30 Real-Life Exercise Patterns in Axial Spondyloarthritis: How Mood, Stress, and Disease Flare Influence Exercise Type and Variety

Background/Aims
The impact of daily psychological factors and disease flares on exercise behaviours in axial spondyloarthritis (axSpA) remains unclear. Exercise programmes improve disease activity, mobility, and quality of life, yet adherence is inconsistent, with patients often reducing activity during flares. Understanding daily behavioural patterns can inform more personalised, psychologically informed exercise advice, enabling healthcare professionals to support flexible, sustainable routines. Previous research shows short-term improvements in mood and well-being following exercise interventions; however, these studies report group-level outcomes and overlook daily within-person variability. Digital health approaches allow detailed investigation of these dynamics.

Methods
This study examined daily associations between mood, stress, and exercise variety, a clinically relevant indicator of flexibility and engagement in multiple exercise types, and whether exercise choices differed between flare and non-flare days in individuals with axSpA. A secondary analysis used data from Project Nightingale, a longitudinal digital health study. Participants (N = 112) self-reported daily symptoms and behaviours via the uMotif smartphone app between 2018-2022. Data included ratings of mood, stress, flare status, and exercise choices via Likert scales or tick-boxes, with exercise variety calculated as the number of activities per day. Conway-Maxwell Poisson mixed-effects regression models examined within-person associations between mood, stress, and exercise variety, controlling for gender, medication use, and flare status. This model accounts for under- and overdispersion in count data. Exercise-type engagement on flare versus non-flare days was examined using mixed-effects logistic regression models, controlling for gender and medication use. Incidence rate ratios (IRRs) and odds ratios (ORs) were reported.

Results
Better mood (IRR = 1.11, 95% CI [1.09, 1.13], p < .001) and lower stress (IRR = 1.05, 95% CI [1.04, 1.07], p < .001) significantly predicted greater exercise variety. During flare days, participants engaged significantly more in yoga (OR = 1.51, 95% CI [1.19, 1.93], p < .001) and mindfulness/meditation (OR = 1.45, 95% CI [1.24, 1.69], p < .001), but significantly less in cardiovascular exercise (OR = 0.72, 95% CI [0.66, 0.80], p < .001), weights/strengthening (OR = 0.45, 95% CI [0.38, 0.52], p < .001), and Pilates (OR = 0.75, 95% CI [0.62, 0.92], p = .006) compared to non-flare days. No significant differences were found for tai chi, hydrotherapy, or gym ball.

Conclusion
This study provides within-person evidence of how psychological well-being and flares influence exercise behaviour in axSpA. Better daily mood and lower stress were significantly associated with greater exercise variety, whilst flare days were characterised by gentler, low-impact activity. This suggests that individuals adapt their exercise routines to symptom fluctuations rather than disengaging completely. These results highlight the need to incorporate psychological support and flare-responsive exercise guidance into axSpA care to promote adherence, flexible activity choices, and improve long-term function, mobility, and quality of life.