Dialysis-imposed, weekly and seasonal patterns of physical activity: a multi-center prospective study in patients using a wearable activity tracker

Background Hemodialysis patients are often sedentary, and their life is structured around the dialysis schedule. Wearable activity trackers present an opportunity for long-term continuous monitoring of physical activity. We aimed to characterize dialysis-imposed, weekly and seasonal patterns of physical activity in hemodialysis patients. Methods In this prospective observational study, patients on in-center hemodialysis in New York City wore the Fitbit (R) Charge 2 (TM) for 1 year. Physical activity was assessed by weekday, dialysis versus interdialytic days (post-hemodialysis day and second interdialytic day), dialysis start time and season. Linear mixed-effects models with random intercepts between patient were constructed to determine the effect of time-patterns and determinants of physical activity levels. Results A total of 109 patients on hemodialysis were included (mean age 54 +/- 11.6 years, 72% male, 23% diabetic). The observed number of steps was 6590 +/- 4014 (mean +/- standard deviation) per day; 44 (40%) participants walked <5000 steps/day. Participants walked 912 (95% confidence interval 768, 1057) fewer steps on Sundays and 284 (129, 440) fewer on the second interdialytic day vs dialysis day. Winter activity was reduced by 321 (162, 478) to 455 (312, 598) steps compared with other seasons. Older age, diabetes and higher equilibrated Kt/V were associated with lower physical activity levels; higher albumin was associated with increased physical activity levels. Conclusion Most hemodialysis patients walked less than recommended level of 10 000 steps/day and physical activity levels vary greatly between and within patients. Dialysis-imposed, weekly and seasonal patterns affect physical activity. Lower physical activity levels on second interdialytic days and Sundays could provide opportunities for improvements of physical activity in shared the decision-making process.