Clinical Implementation of Wearable-Derived Sleep and Activity Reporting for Inpatient Psychiatric Monitoring (Preprint)

Background: Sleep is a core component of psychiatric assessment, yet inpatient monitoring typically relies on brief observational checks that are subjective, variable, and sometimes disruptive. Wearable devices offer a means of capturing continuous, objective sleep and activity data without disturbing patients. Although digital health technologies are increasingly used in psychiatric research, little is known about how wearable-derived data can be integrated into routine inpatient workflows or used meaningfully by clinicians. Objective: This implementation aimed to evaluate the feasibility, usability, and workflow integration of a wearable-derived sleep and activity reporting system within an adult psychiatric inpatient unit. Methods: The implementation unfolded in 2 phases at a single 21-bed adult inpatient unit at a psychiatric hospital in Massachusetts. Patients were offered a wrist-worn GENEActiv actigraphy device upon admission. Raw accelerometry data were processed using the DPSleep pipeline to derive daily sleep and activity metrics for patients participating in the implementation. Sleep and activity reports combining graphical summaries and natural language summaries of sleep, activity, and medication data were iteratively refined and delivered to psychiatrists providing patient care. Semistructured qualitative interviews were conducted with clinicians and unit staff to gather feedback on the sleep and activity report prototype and discuss barriers to and facilitators of implementation. Interview data were coded and analyzed by a team of 2. Unlabelled: During phase 1 of the implementation, 155 patients were admitted, of whom 88 (56.8%) were offered a device and 68 (77.3%) accepted it. Sleep and activity reports were generated for 61.8% (42/68) of patients wearing a device during this phase. During phase 2 of the implementation, automation reduced report generation time from approximately 5 days to under 24 hours. Only 1 of the 3 psychiatrists on the unit regularly used the reports in routine care. Reports were most useful for reconciling discrepancies between patient and nursing sleep estimates and for supporting clinical conversations about sleep patterns and medication adherence between clinician and patient. Clinicians who had not yet used the reports expressed conceptual interest but emphasized the need for integration in the electronic medical record, reliably available "last-night" sleep data, and simplified design. Barriers included challenges in the speed, reliability, and clarity of the data; variable staff buy-in; and disconnects between the research and clinical teams running the implementation. Conclusions: This implementation suggests that wearable-derived sleep and activity data reporting is technically feasible in inpatient psychiatry. This data reporting potentially offers clinically meaningful insights. Use of the reports was concentrated in 1 of the 3 psychiatrists on the unit, who served as an early adopter and project champion (AGY). Sustainable use and broad clinical uptake are more likely with reliable, near-instantaneous data transfer; electronic medical record integration; and shared implementation ownership across staff levels.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.