Pilot and Feasibility Deployment of an Advanced Remote Monitoring Platform for COVID-19 in Long-Term Care Facilities.

Introduction:

Older adults in long-term care facilities (LTCFs) are at a significantly higher risk for hospitalization and death due to COVID-19.^1,2 Prolonged stays, close aggregation, and low functional status compound poor prognosis of this vulnerable population.³ Wearable devices have increasingly emerged as novel tools to track and mitigate outbreaks given continuously collected and wirelessly transmitted physiological data.^4,5 Devices for older adults in LTCFs must consider low technical literacy, medical complexity, cognitive decline, skin fragility, overburdened caregivers, and limited technical staff expertise. Though recent studies have used wearables like AppleWatch and FitBit to identify COVID-19 infections, they systematically underrepresent older adults and fail to measure key symptoms including cough, fever, and shortness of breath.^4,6 This study evaluated the feasibility of remotely deployed bio-integrated wireless sensors to comprehensively measure vital signs in high-risk, residential older adults.

Methods:

This is a fully virtual single-arm, prospective observational study of older adults in two LTCFs (Chicago, IL) of the ANNE One (Sibel Inc.), a FDA-cleared physiological monitoring system (Figure 1A), consisting of two medical-grade silicone patches. A chest sensor is positioned at the suprasternal notch by biocompatible, conductive adhesive, while the second is wrapped around the index finger. The chest unit has a 1-lead electrocardiogram (ECG), 3-axis accelerometer, and temperature sensor, capturing continuous heart rate (HR), respiratory rate (RR), chest wall movement, snoring, respiratory sounds, seismocardiography, body position, and temperature.⁷ The ANNE limb sensor has a photoplethysmograph (PPG), and temperature sensor, capturing pulse oxygenation, HR, perfusion index, and peripheral arterial tonometry (PAT). The two units are time synchronized generating pulse transit time (PTT) and continuous blood pressure.⁸ Sensors record data and automatically download to the cloud and tablet via encrypted Bluetooth.⁷

Figure 1.

The ANNE sensor system is shown deployed on an older adult LTCF resident (A). In panel (B), the data outputs in a 60-second snap shot are shown with continuous 1–lead ECG for heart rate and respiratory rate), photoplethysmograph (PPG) for SpO₂, seismocardiograph (SCG) for heart sounds and respiratory sounds, accelerometry (ACC) for motion, and temperature (TMP). (C) Scalable model for remote deployment of an advanced wireless continuous monitoring system for LTCFs during pandemic situations.

Patients with active skin conditions, major psychiatric disorders, or inability to consent were excluded. LTCF staff were virtually trained and applied the system to patients, after virtual consent. Participants wore sensors for up to 14 days and completed questionnaires regarding comfort and usability. Skin was evaluated for injury after sensor removal. Patient demographics, HR, RR, pulse oxygenation, cough count, position, and snoring were collected. Advarra Institutional Review Board (STU00213706) approved the trial.

Results:

22 patients (median age 84 years old; 65–95 years) enrolled between January 27 and May 13, 2021. Hypertensive (66%) and neurocognitive disorders (55%) were common. 87 gigabytes of data were collected from 2,738 monitoring hours. Participants completed an average of 6 sessions (131 total sessions, averaging 161 hours or ∼6 days per participant). Figure 1B demonstrates the system’s outputs. Patient and population level summary statistics are generated, including heart rate, respiratory rate, SpO², blood pressure, snoring, cough and fall events. No adverse events or data transfer failures occurred. Overall, 45% of participants reported remote monitoring made them feel safer, while 61% felt it provided helpful information for their physicians. 55% described sensor use as easy or very easy, with 61% preferring the ANNE One system to wired hospital monitors.

Discussion:

We demonstrated feasibility, safety, and acceptability of remote monitoring of medically complex, residential older adults with low profile, bio-integrated wearable sensors. Though nearly 70% of participants were 80 years or older with multiple comorbidities, the system was correctly applied over 100 times by LTCF staff without injury or device failure. The design, deployment, and validation of remote monitoring systems intended for older adults should consider age, comorbidities, cognitive decline, limited mobility, skin fragility, and low technical literacy.⁹ The sensor system presented here addresses many of these challenges. Figure 1C illustrates the model for sustainable and scalable programs for remote staff training and patient onboarding with automatic transmission of real-time data for healthcare decision making at both the patient and population level. The ergonomic, waterproof, and wireless design is unobtrusive providing continuous, ICU-grade vital sign monitoring without interrupting sleep or requiring confinement to bed or bulky bases, minimizing fall risk, disorientation, and delirium. Elderly skin is 50% thinner with increased vulnerability to shear or frictional injuries and slower healing. There were no skin injuries as a result of the hydrogel adhesives—prior work using the sensors in premature neonates demonstrated similar safety profiles.¹⁰ This study is not without limitations. It was a small pilot at two different urban LTCFs. Therefore, our findings may not be generalizable. Conventional consumer wearables may not provide relevant biomarkers with sufficient accuracy to achieve disease prevention and surveillance. We demonstrate successful deployment of the ANNE ecosystem with a contactless trial design, virtual training, and device reusability cumulatively suggesting a viable path for scalable, sustainable distribution in LTCFs.