Abstract Background/Aims Understanding the causal relationship between physical activity and pain is needed to design effective and sustainable physical activity interventions for people with knee osteoarthritis. This study aimed to explore the temporal association between day-to-day changes in physical activity (step count) and pain in individuals with knee osteoarthritis. Methods Using smartwatches, with a bespoke app-based symptom questionnaire, twice-daily knee pain Numerical Rating Scale (0 - 10) and step count data were collected contemporaneously from 26 participants with knee osteoarthritis for ∼3-months. Generalised linear mixed models were used to explore bi-directional associations between step count and pain measured on the same day, the prior day, and the following day. Results We analysed 1,473 daily pain and step count measurements, recorded over a median 58 days in 26 people of median age 64. There were considerable day-to-day changes in individuals’ step count (range 423 - 7,142) and pain (range 0 - 9). A higher step count was associated with higher average pain on the same day (0.04 points/1000 step increase, 95%CI 0.01 - 0.06), and the following day (0.05 points/1000 step increase, 0.03 - 0.07). We found no association between change in step count from prior day and the following day’s pain, nor did we find association between prior day’s pain and change in pain from prior day and the following day’s step count. Conclusion There was a modest positive association between step count and pain assessed on the same day and the following day. This study demonstrates how smartwatches can capture daily data to enable the exploration of complex time-varying relationships in people with knee osteoarthritis. Disclosure A.E. Olanrewaju: None. M.J. Parkes: None. J.C. Sergeant: None. E. Pritchard: None. S. Shao: None. S.R. Filbay: None. S.N. van der Veer: None. D. Wong: None. W.G. Dixon: None.
Olanrewaju et al. (Wed,) studied this question.