Personal Health and Well-Being Determinants Associated with the Day-to-Day Variability of Sedentary Behaviour in Community-Dwelling People with Stroke

Background: To improve personalized behavioural interventions for people with stroke, it is crucial to understand which factors influence fluctuations in sedentary behaviour. This study aimed to explore the association between health and well-being determinants and daily variability in sedentary behaviour over time within community-dwelling people with stroke. Methods: An n-of-1 study design was conducted to examine the associations between determinants and sedentary behaviour during the RISE-intervention randomized multiple baseline study. The percentage of sedentary behaviour was measured daily with the ActivPAL activity monitor. The Visual Analogue Scale scores of the determinants were collected with diaries. Dynamic regression modelling (time-series analysis) was performed, starting with univariable and followed by multivariable linear regressions. Results: The analysis included twelve community-dwelling people with stroke (median age 65 years), with daily sedentary behaviour ranging from 55.5 to 81.4 percent. Objectively measured sleep length was positively associated with the percentage of sedentary behaviour in three participants (p = 0.001, p < 0.0001, and p = 0.045), and negatively associated in one participant (p = 0.002). Subjective sleep length (p = 0.016), fatigue (p = 0.013) and pain (p = 0.0098) were exclusively associated with the percentage of sedentary behaviour in one participant. No significant associations were found for sleep quality, happiness, stress or time pressure in five participants. Conclusions: The findings indicate inconsistent association patterns between health and well-being determinants and the day-to-day variability of sedentary behaviour across participants. This highlights the need for a personalized approach, as the determinants associated with the daily variability of sedentary behaviour in one individual may differ from those influencing another individual.