Presentation Authors: Necole Streeper*, Hershey, PA, Edison Thomaz, Austin, TX, Ashley Sanders, Alexandra Dubnansky, David Conroy, University Park, PA
Introduction: Compliance with increasing fluid intake to produce at least 2.5L of urine daily for stone prevention is commonly below 50%. We have previously shown that patients are interested in utilizing sensors to provide automated lapse detection in fluid intake to improve adherence to recommendations. Wrist-worn inertial sensors can detect drinking behavior with accelerometers and gyroscopes. The purpose of this study was to devise a protocol to test the ability of a wrist sensor to refine the detection of drinking behavior in everyday activities and to evaluate the recruitment process for future clinical trials.
Methods: Patients with kidney stones were recruited to participate in the 70-minute lab sessions. Wrist sensors were worn bilaterally and participants were videotaped performing a variety of daily activities, including drinking from various vessels.
Results: 36 patients with a history of kidney stones were enrolled to participate out of a total of 56 who were eligible. Six participants cancelled their appointment and we had a total of 30 participants (87% female, average age 33) that completed the lab study over 18 sessions. Table 1 highlights the amount of time spent performing protocol activities.
Conclusions: Sensor data can be collected in the laboratory to classify drinking gestures. The digital biomarker being developed in this protocol can be applied to automate lapse detection and trigger fluid intake for patients with kidney stones. Combining this automated lapse detection with evidence-based behavior change techniques may improve adherence to fluid intake recommendations and prevent recurrence of kidney stones.
Source of Funding: The project described was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant UL1 TR000127 and TR002014. The content is solely the responsibility of the authors and does not necessarily repre