COVID-19 highlighted many challenges the U.S. healthcare system faces when healthcare workers and their resources are tested by an overwhelming number of patients in need of care.
As treating patients with COVID-19 became a top priority, non-essential medical care was deferred or conducted via telehealth to reduce the burden on providers, lower risk of exposure for patients and slow the spread of the virus. Telehealth services became a lifeline for millions of Americans seeking care during shelter-in-place and quarantine orders.
Hospital-at-Home (HaH) Care Models Gain Traction
In November 2020, a major hurdle to telehealth adoption was removed when the Centers for Medicare and Medicaid Services (CMS) announced the Acute Hospital Care at Home (AHCaH) program. The program lifted constraints on reimbursements for the hospital-at-home (HaH) model allowing patients to be voluntarily “admitted” to their homes, while clinicians remotely manage their care, freeing up hospital beds for the most critical cases. Indeed, by early 2021 the use of telehealth was 38 times higher than before the pandemic.
Evidence shows that patients under care in the HaH model receive an equivalent quality of care to brick-and-mortar hospitals, with lower admission rates, lower mortality risk and high patient satisfaction. Hospitalization for any reason can be disruptive for a patient. Anyone who has spent the night in a hospital can attest that restful sleep is elusive in an unfamiliar environment with frequent interruptions to check vital signs. Increased stress levels and lack of sleep can hinder the healing process. The average hospital stay for U.S. patients is 4.5 days, and studies have shown that the longer a patient stays in the hospital, the less likely they will have a positive outcome because their risk for hospital-acquired infection (HAI) and other complications increases.
Spurred by the AHCaH program and accelerated reimbursement by payers, next-generation technologies are enabling the deployment of innovative HaH programs delivering many tangible benefits, including:
- Improved access to care
- Improved overall care and outcomes
- Fewer visits to the emergency department
- Lower risk of adverse events and infection
- Improved patient mobility and satisfaction
- Better preventative health management
- Reduced overall costs
With this expanded coverage, CMS outlines rules for physiologic monitoring services, including the requirement that platforms used at home are regulated by the U.S. Food and Drug Administration (FDA) and data must be electronically and automatically collected and transmitted, rather than self-reported by the patient. Additionally, continuous remote patient monitoring (cRPM) may be deemed medically necessary for patients with certain acute and/or chronic conditions.
Stay up to date on the latest news, regulations and innovations in medtech design and development. Subscribe to the MedTech Intelligence weekly newsletter.
Remote patient monitoring is not a new care modality—hospitals monitor patients remotely in hospital beds, as clinicians cannot remain bedside with each patient 24/7, and telehealth monitoring is deployed in many forms to manage supplemental care of patients in their homes—but not all monitoring technologies are created equal. Some systems involve “spot-check” measurements taken and entered manually into a digital system by the patient or their caregiver. Relying on intermittent physiological measurements, which are often inconsistent and experience lag times, can limit a provider’s ability to make informed treatment decisions.
In contrast, combining wearable biosensors with machine learning-based analytics generates a continuous stream of clean, high-quality data that may serve as an early warning system. The benefits are significant with studies showing better clinical outcomes, shorter average length of stay, higher patient and family satisfaction, fewer complications and significant cost savings compared to traditional inpatient hospital care.
AI-Driven cRPM for Post-Acute Care
Medical-grade cRPM assets utilizing artificial intelligence (AI) can collect and analyze wearable biosensor data and extract highly personalized clinical insights from real-world continuous data streams. A growing portfolio of FDA-cleared AI-driven algorithms convert raw data into actionable insight. Personalized physiology analytics (PPA) can track and integrate digital biomarker signals to detect clinically meaningful changes against an individual’s baseline, rather than comparing that individual to population-based norms.
Designed to improve physiologic monitoring accuracy by tapping the power of deep neural network computational approaches, PPA algorithms must be validated in accordance with medical-grade standards and FDA 510(k) clearance. These sophisticated embedded analytics enable a multivariate baseline to be developed for each patient to detect subtle yet meaningful changes in a patient’s physiology and extract actionable clinical insights, so clinicians can make more efficient decisions and allocate resources as early as possible. Other key benefits include the capability to:
- Follow and treat patients across the entire care continuum
- Receive early alerts to help avoid hospital admissions
- Engage and empower patients in their own care to improve disease management
The market for wearable consumer devices is booming with more than 80% of U.S. consumers expressing willingness to wear devices. The factors that matter most to patients when engaging with health care include: ease of access (89%), personalized care (88%), healthcare costs (85%) and the ability to have their health monitored remotely using wearable devices and apps (75%) while maintaining privacy.
A recent clinical trial used AI-driven cRPM to manage ambulatory patients diagnosed with COVID-19. Participants wore the adhesive biosensor patch 90% of the time, and many participants expressed reluctance to give up the patch after the prescribed 28 days, citing they felt safer knowing they were being monitored 24/7 by a care team.
AI-Driven cRPM in Chronic Disease Management
Identifying illness early, often before the patient notices symptoms, can allow providers to take action to prevent further exacerbation.
Those with chronic medical conditions account for over $1 trillion annually in healthcare costs including: in-hospital care, pharmaceuticals, medical devices and at-home care. And people with chronic medical conditions along with those suffering from diabetes and heart failure rank among those with the highest hospital readmission rates.
Early intervention can prevent a severe exacerbation or decompensation when treating people with chronic conditions. Between the start of the pandemic and June 2020, over 40 % of U.S. adults delayed or avoided seeking medical care due to the risk of contracting COVID-19. Patients with chronic conditions were among those most likely to avoid seeking treatment. By using at-home options like cRPM, healthcare providers can empower patients to stay engaged in their own care and maintain disease management and overall health.
Studies support the safe and cost-effective use of HaH interventions as an alternative to hospitalization for people with post-acute and chronic disease, with no greater risk of mortality and a lower rate of hospital readmission. AI-driven cRPM epitomizes personalized precision medicine and can strengthen the influence of early interventions aimed at helping overall health of patients.
Medication therapy, for example, is an essential aspect in monitoring and treating ambulatory patients with chronic conditions. Those with poor medication adherence do not feel better and have poorer outcomes. Yet, about one-third of patients never fill their prescriptions, and three out of four do not take their medications as directed. The cost of preventable medical procedures resulting from medication non-adherence is estimated at up to $300 billion annually.
For many medications, adjusting a drug to the optimal dose requires multiple patient visits. This can be burdensome and create delays in achieving optimal treatment. Based on objective AI-driven cRPM data a clinician can quickly identify a patient who may be experiencing medication side effects, skipping doses or forgetting to take medications, so they can address causes of non-adherence and get the patient back on track.
Building on the Momentum
During the height of the pandemic, remote monitoring of COVID-19 patients delivered vital benefits and supported public health safety protocols. AI-driven cRPM HaH programs remained available, even when face-to-face studies and other health services were suspended. HaH also opened the door to caring for more patients across the country. The industry has learned that when revenue streams are in line with in-person care, HaH can provide a safe, efficient care delivery model for many patients.
Meeting healthcare systems where they are by integrating the capabilities of novel digital AI-driven cRPM can deliver proactive solutions to identify and engage with patients in need of critical care. Future growth depends on embracing innovation to intervene before patients fall ill and to ensure the highest quality of care.
 Bestsennyy, O., Gilbert, G., Harris, A., and Rost, J. (2022, February 28). Telehealth: A quarter-trillion-dollar post-covid-19 reality? McKinsey & Company. Retrieved June 9, 2022.
 Federman, A. D., Soones, T., DeCherrie, L. V., et al. (2018, August 1). Association of a bundled hospital-at-home program and 30-day post-acute transitional care program with clinical outcomes and patient experiences. JAMA Internal Medicine.
 Clarkson, K. Why reducing length of stay is critical for Patient & Hospital Wellbeing. Pulsara. (2021, June 23).
 Leff, B., Burton, L., Mader, S. L., et al. Hospital at home: Feasibility and outcomes of a program to provide hospital-level care at home for acutely ill older patients. Ann Intern Med 2005;143:798-808.
 Phaneuf, A. (2021, May 21). 5 examples of popular wearable devices in healthcare. Business Insider.
 US healthcare industry 2022: Sectors, Trends & Statistics. Insider Intelligence (2022, January 11).
 Richards, D. M., Tweardy, M. K. J., Steinhubl, S. R., Chestek, D. W., Vanden Hoek, T. L., Larimer, K. A., and Wegerich, S.W. (2021, November 8). Wearable sensor derived decompensation index for continuous remote monitoring of COVID-19 diagnosed patients. NPJ digital medicine.
Reed, T. (2019, April 16). Report: U.S. economic burden of chronic diseases tops $3.8 trillion-and expected to double. Fierce Healthcare.
 Weiss, A. J. and Jiang H. H. Agency for Healthcare Research and Quality – User support. Overview of Clinical Conditions With Frequent and Costly Hospital Readmissions by Payer, 2018 #278. (2021, July).
 Siwiki, B. How remote patient monitoring is moving into the mainstream. Healthcare IT News. (2022, March 9).
 Arsenault-Lapierre, G., Henein, M., Gaid, D., et al. (2021, June 8). Hospital-at-home interventions vs in-hospital stay for patients with chronic disease who present to the emergency department – A systematic review and meta-analysis. JAMA Network Open.
 Iuga, A. O., and McGuire, M. J. (2014, February 20). Adherence and Health Care Costs. Risk management and healthcare policy.