Predictive analytics takes center stage in the fight against COVID-19 and the Critically Ill

Highlighting the vulnerabilities and the challenges

The COVID-19 epidemic has brought the vulnerability of healthcare systems and how they can be rapidly overloaded in excess of the available ICU bed and ventilator capacity sharply into focus.  Additionally, COVID-19 has highlighted the significant shortage of Intensivist and qualified and experienced ICU nursing staff.

ICUs are currently being challenged on multiple fronts including resource limitations, infection control, protection of healthcare workers (HCWs), and adaptation of services to a rapidly evolving pandemic situation. 1 Goh These demands combined with decreased revenue generation from elective procedures has placed profound strain on health systems both logistically and financially.  These stressors combined with under and overutilization of manpower have created an environment that many consider the most challenging in the history of American healthcare.

One area that may assist in the current environment are real-time prediction models.  Rarely used in American healthcare, such models may offer an opportunity to change practice and enhance efficiency in the operation of clinical enterprises.  Prediction models, ranging from rule-based scoring systems to advanced machine learning models, that combine a few to hundreds of variables or features to estimate the risk of experiencing deterioration or a poor outcome could assist medical staff in triaging patients when allocating limited healthcare resources. 2 Wyants

As an overall response, predictive models have been proposed to support healthcare authorities in early planning of resources, personnel, ICU, and hospital bed capacity.

In the ICU, predictive markers of patient deterioration enable clinical decision support to facilitate resource utilization and enhance patient outcomes.  Given the complexities associated with COVID-19 treatment, such models may enhance patient interaction, ICU staff capacity, and virtual ICU care. 1 Goh, 3 Aziz, 4 Alban

Reducing the frequency and duration of times that staff enter the rooms of patients with COVID-19, is generally thought to reduce healthcare worker exposure. Potential strategies include the use of telemedicine to monitor patients along with other remote communication and monitoring devices. 3 Aziz However, implementing such a workflow without impacting outcomes remains a delicate balance.  Predictive models of decompensation may be a key element in ensuring appropriate and timely interaction with patients while still having adequate time to conform to best personal protective equipment practices.

Predicting and upgrading capacity

In general, most American ICUs are equipped to withstand an approximate 20% surge in demand.  However, in the face of the pandemic, a significantly higher ICU surge capacity has been needed and critically ill patients may need to receive care outside of a traditional ICU, depending on available resources.  It is now becoming common for critically ill patients to be cared for in locations that were previously anesthesia recovery areas or general wards.  At the same time decisions on timely discharge to step-down care areas have become exceptionally important in effectively managing ICU resources and ensuring availability. 1 Goh 

Although such logistic decisions may provide increased physical capacity for the critically ill, they do little to address critical shortages in competent manpower.  With such shortages of human resources, in extreme situations like those recently seen in New York City, intensivists and ICU nurses may have to take up a supervisory role, with non-critical care HCPs providing direct care to patients. 5 Kumarajah 

Redeployment of resources

Tele Critical Care has come to the forefront in the pandemic as a beneficial tool that expands the reach of experts in critical care and allows skilled critical care physicians and nurses to work in areas from which they are geographically remote. 3 Aziz Aziz et al, reported on a reorganization strategy of team structures and workflow to adapt to this rapidly increased resource demand while still enabling quality critical care. This requires placing senior skilled ICU physicians in oversight positions with non-ICU trained physician staff at the bedside (Figure 1) 3 Aziz In such situations, Tele Critical Care solutions can act as a significant workforce multiplier.  While an expanded model like Aziz describes can provide care to perhaps 50 patients with a single intensivist, Tele Critical Care technology commonly expands intensivist coverage to as many as 150 to 200 patients per physician.

Figure 1: Care team model for extending the capacity of ICU clinicians

Adapted from Aziz S, et al 3

Determining deterioration or improvement and planning accordingly

In cases of COVID-19 critical illness, patients have progressed rapidly to acute respiratory failure, acute respiratory distress syndrome, and septic shock. Early identification of risk factors for critical illness can facilitate rapid access to the ICU when required. For patients with mild and moderate illness, general isolation and medical treatment is required.  ICU-care is not needed unless their condition worsens. Awareness of early deterioration is a vital tool in helping to reduce mortality and alleviate the shortage of medical resources. 6 Shang

By closely observing patients in areas with lower levels of staffing and then triaging them to ICUs only when requiring critical care treatment, ICU resources can be dedicated to those patients that truly need them.  This is a different model of operation compared to many facilities where patients that are at risk are admitted to the ICU for monitoring.

What about predicting deterioration?

A model of close observation outside of the ICU introduces profound risk to patients and the institution if not implemented correctly as such models are dependent on clinicians being situationally aware of the active deterioration and being resourced to act and provide critical care therapies at a moment’s notice.  Such models require appropriate technology for monitoring and medical team resources be readily deployable to areas outside of the ICU.

Failing to recognize a significant deterioration can result in further deterioration and/or cardiac arrest; therefore, early recognition of decompensation is important as early intervention may avoid intubation. It is not only the respiratory parameters that need monitoring in COVID-19 disease. Intensive hemodynamic monitoring should be considered for patients with COVID-19 as there have been multiple case reports of patients suffering profound hemodynamic collapse.  The etiology of such events is not clear as some cases appear linked to embolic phenomena and others appear linked to profound myocardial depression.

The possibility of predicting which patients will decompensate to require critical care therapies hours ahead of time offers the opportunity to intervene early and minimize the need for maximal critical care support.  In the event of a 4-hour advanced warning of respiratory failure for a patient it is possible to prepare.  A patient can be transferred to an ICU, receive a trial of high flow nasal cannula and self-proning in addition to optimization of volume status and medical therapy.  Such interventions may eliminate the need for invasive mechanical ventilation, likely reducing ICU length of stay and morbidity associated with mechanical ventilation.  One could likely see similar interventions in the event of such a warning regarding hemodynamic compromise.

How does it all come together and integrate?

For such models to work, institutions much have technology in place to assess, risk stratify and monitor patients.  With the current ICU shortage of beds and qualified providers, one critical question remains. Which skilled intensivist and what technology will be at the center of monitoring the real time status of all the patients, integrating all the predictive parameters and anticipating deterioration or clinical improvement to support further patient management decisions as well as assisting in planning daily ICU capacity?  CLEW’s monitoring technology offers such a definitive solution.  Whether your facility operates exclusively with on-premises providers or utilizes Tele Critical Care services, CLEW’s unique technologies can help you enhance your providers’ situational awareness and better utilize your ICU capacity.


  1. Goh KJ, Wong J, Tien H-C C, Ng SY, Wen SD, Phua GC, et al. et al Preparing your intensive care unit for the COVID-19 pandemic: practical considerations and strategies. Critical Care 2020;24:215. Available from: Accessed July 14,
  2. Wynants L, Van Calster B, Collins GS, Riley RD, Heinze G, Schiut E, et al. Prediction models for diagnosis and prognosis of covid-19: systematic review and critical appraisal. BMJ 2020; 369: m1328.doi: 10.1136/bmj.m1328: 10.1136/bmj.m1328 Available from: Accessed August 3, 2020
  3. Aziz S, Arabi YM, Alhazzani W, Evans L, Citerio G, Fischkoff K, et al. Managing ICU surge during the COVID-19 crisis: rapid guidelines. Intensive Care Med 2020. Available from: Accessed July 14,2020.
  4. Alban A, Chick SE, Dongelmans DA, Vlaar APJ, Sent D and Study Group. ICU capacity management during the COVID-19 pandemic using a process simulation. Intensive Care Med 2020; Available from:  Accessed July 14, 2020.
  5. Kumarajah D, Yip N, Ivascu N, Hill L.Innovative ICU Physician Care Models: Covid-19 Pandemic at NewYork Presbyterian. NEJM Catalyst Innovations in Care Delivery 2020 DOI: 10.1056/CAT.20.0158 Available from: Accessed August 3, 2020.
  6. Shang Y, Pan C, Yang X, Zhong M, Shang X, Wu Z, et al. Management of critically ill patients with COVID‑19 in ICU: statement from front‑line intensive care experts in Wuhan, China. Ann Intensive Care 2020 10:73 Available from: Accessed July 14, 2020







Like what you see? Drop us a line.

Schedule a time with us here:

* We want to see you. Register before we’re filled.

(*) indicates mandatory field

One moment while we direct
you to the CLEW website

Patient View
Rapid, clear, consistent patient data summary and visualization

  • Patient risk stratification (high, moderate and low risk patients)
  • Intubated patients
  • Vasopressors/inotrope support
  • Empty beds
  • Key patient and unit information
  • Occupancy, ventilator status and vasoactive medication
  • User selectable vital sign display
  • Summary of all relevant lab results
    Dynamic and collaborative caregiver task list
  • Aggregate patient demographicsand PMH (past medical history) andpredicted risk level

Patient View
Rapid, clear, consistent patient data summary and visualization
High Risk Patients

  • Tasks
  • New Admissions (pulled from a left sidebar)
  • User selectable vital sign display
  • Summary of all relevant lab results
  • Dynamic and collaborative caregiver task list
  • Aggregate patient demographics, PMH and risk level

Unit View
Full TeleICU situational awareness, displaying patient data and predicted state from all units and highlighting all notifications

  • Patient risk stratification (high, moderate and low risk patients)
    Intubated patients
  • Patients on vasoactive medication
    Unit occupancy
  • Patients on vasoactive medication
  • High-level estimation of the overall unit acuity

    Key patient and unit information


Work List
Integration with the TeleICU workflow by gathering all notifications, tasks, new admissions and high-risk patients

  • Notifications
  • High Risk Patients
  • Tasks
  • New Admissions (pulled from a left sidebar)

Unit View by Layout

Full ICU situational awareness, displaying patient data and predicted risk level, for one unit or multiple units. The display highlights all notifications, low risk patients and other key clinical information, to provide multi-dimensional situational awareness.

  • Patient risk stratification (high, moderate and low risk patients))
  • Intubated patients
  • Vasopressors/inotrope support
  • Empty beds
  • Key patient and unit information
  • Occupancy, ventilator status and vasoactive medication
Thank you for requesting our demo.

You’ll be hearing from our specialist shortly.