Abstract: FR-PO0410
StepNet: A Personalized, Real-Time Wearable Intervention to Promote Physical Activity in Patients on Hemodialysis
Session Information
- Dialysis: Measuring and Managing Symptoms and Syndromes
November 07, 2025 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 801 Dialysis: Hemodialysis and Frequent Dialysis
Authors
- Malhotra, Rakesh, University of California San Diego, La Jolla, California, United States
- Dasgupta, Subhasis, University of California San Diego, La Jolla, California, United States
- Ahmadi, Armin, University of California San Diego, La Jolla, California, United States
- Larsen, Britta, University of California San Diego, La Jolla, California, United States
- Ix, Joachim H., University of California San Diego, La Jolla, California, United States
Background
Hemodialysis (HD) patients have markedly low physical activity levels, contributing to poor quality of life, frequent hospitalizations, and increased mortality. Most physical activity programs rely on in-person counseling, which is resource-intensive and difficult to scale. We developed StepNet, a personalized, low-touch mobile health system that leverages wearable activity data and behavioral messaging. We tested its ability to promote physical activity in HD patients.
Methods
We conducted a 12-week single-arm pilot study in 25 maintenance HD patients using Fitbit activity trackers. Each participant received personalized weekly step goals and adaptive text messages informed by autoregressive modeling of prior step counts, dialysis schedule, and psychosocial factors. Text messages were delivered three times weekly: goal setting (start of week), progress update (mid-week), and reinforcement (end of week). The primary outcome was change in average daily step count from baseline to the end of the 12-week intervention period.
Results
Of 25 enrolled participants (mean age 64 years, 61% male, 57% Hispanic), 23 completed the 12-week intervention. The mean daily step-count at baseline was 2959 ± 1129 steps, which increased by 1,449 steps/day over the 12-week period (P = 0.02). Fitbit adherence was 94%, with over 90% syncing at least once every 3 days. Unsupervised clustering of weekly step trajectories revealed distinct behavioral phenotypes, highlighting heterogeneity in response patterns (Figure 1).
Conclusion
An intervention using wearable step count technology linked to text-based personalized adaptive text-based exercise guidance is feasible, well-tolerated, and associated with increase in physical activity among HD patients. These findings support the use of wearable-guided interventions in exercise trials in high-risk HD patients.
Funding
- NIDDK Support