Wearable-derived sleep and activity data reporting is technically feasible in inpatient psychiatry but sustainable clinical uptake requires reliable near-instantaneous data transfer, electronic medical record integration, and shared implementation ownership across staff levels.
Key Findings
Results
A majority of admitted patients accepted wearable actigraphy devices when offered, indicating patient-level feasibility.
155 patients were admitted during phase 1 of the implementation at a single 21-bed adult inpatient psychiatric unit.
88 of 155 patients (56.8%) were offered a GENEActiv wrist-worn actigraphy device upon admission.
68 of 88 patients offered a device (77.3%) accepted it.
Sleep and activity reports were successfully generated for 42 of 68 device-wearing patients (61.8%) during phase 1.
Results
Automation in phase 2 dramatically reduced the time required to generate sleep and activity reports.
In phase 1, report generation took approximately 5 days.
Following automation implemented in phase 2, report generation time was reduced to under 24 hours.
This reduction in report generation time was identified as a key facilitator for potential clinical adoption.
Despite this improvement, clinicians still emphasized the need for reliable 'last-night' sleep data, suggesting sub-24-hour turnaround remained a barrier.
Results
Clinical use of the wearable-derived reports was highly concentrated, with only one of three psychiatrists regularly incorporating them into routine care.
Only 1 of the 3 psychiatrists on the unit regularly used the reports in routine care.
This psychiatrist served as an 'early adopter and project champion' (identified as AGY in the paper).
The other two clinicians expressed conceptual interest but did not adopt the reports into routine practice.
Non-adopting clinicians cited lack of electronic medical record (EMR) integration, insufficient data speed and reliability, and overly complex report design as barriers.
Results
The sleep and activity reports were most clinically useful for reconciling discrepancies between patient and nursing sleep estimates and for supporting clinical conversations.
Reports were described as most useful for reconciling discrepancies between patient self-report and nursing observational sleep estimates.
Reports also supported clinical conversations about sleep patterns and medication adherence between clinicians and patients.
This use case reflects a practical niche where objective actigraphy data provided added value over existing subjective monitoring methods.
Inpatient monitoring typically relies on brief observational checks described as 'subjective, variable, and sometimes disruptive.'
Results
Multiple implementation barriers were identified through qualitative interviews with clinicians and unit staff.
Semistructured qualitative interviews were conducted with clinicians and unit staff; interview data were coded and analyzed by a team of 2.
Barriers included challenges in the speed, reliability, and clarity of the data.
Variable staff buy-in across the unit was identified as a barrier.
Disconnects between the research and clinical teams running the implementation were cited as an organizational barrier.
Clinicians emphasized the need for EMR integration and simplified report design as prerequisites for broader adoption.
Methods
Raw accelerometry data were processed using an automated pipeline to derive sleep and activity metrics reported to clinicians.
Patients wore the GENEActiv wrist-worn actigraphy device, which captured continuous raw accelerometry data.
Raw data were processed using the DPSleep pipeline to derive daily sleep and activity metrics.
Reports combined graphical summaries and natural language summaries of sleep, activity, and medication data.
Reports were iteratively refined based on clinician and staff feedback gathered during the implementation.
Conclusions
Sustainable and broad clinical uptake of wearable-derived reporting is contingent on several system-level factors beyond technical feasibility.
The authors identified reliable, near-instantaneous data transfer as a key requirement for sustainable use.
Electronic medical record integration was emphasized as necessary for broad clinical uptake.
Shared implementation ownership across staff levels—not just research or champion-driven adoption—was identified as essential.
The implementation was conducted at a single 21-bed adult inpatient unit at a psychiatric hospital in Massachusetts, limiting generalizability.
What This Means
This research suggests that it is technically possible to use wrist-worn activity trackers to generate objective sleep reports for patients admitted to an inpatient psychiatric unit, and that many patients are willing to wear these devices. In a study conducted at a 21-bed psychiatric unit in Massachusetts, more than three-quarters of patients who were offered a device agreed to wear one, and automated data processing was eventually able to produce reports within 24 hours of data collection. The reports were found to be most useful when doctors needed to resolve disagreements between what patients said about their own sleep and what nursing staff observed during their checks—a real clinical problem since traditional monitoring involves brief, sometimes disruptive, and inherently subjective observations.
However, the research also highlights significant challenges to making this technology a routine part of clinical care. Despite being available on the unit, only one of three psychiatrists regularly used the reports, and that individual served as an enthusiastic early champion of the project. The other clinicians expressed interest but pointed to practical obstacles: the reports were not integrated into the electronic medical record system where doctors normally work, data was not always available quickly enough to inform same-day clinical decisions, and the report format was seen as too complex. Staff buy-in across the whole team was inconsistent, and there were organizational gaps between the research team running the project and the clinical staff using it.
This research suggests that wearable sleep monitoring in psychiatric inpatient settings has real clinical potential—particularly for giving clinicians an objective window into patients' sleep that doesn't require waking them up—but that technology alone is not enough. For tools like this to be widely adopted, they need to be seamlessly woven into existing clinical workflows, deliver data fast enough to be actionable, and have buy-in from staff at all levels, not just individual champions.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.