Working for a Remote Intensive Care Unit: How Does It Differ?

Antonios Liolios, MD; Timothy G. Buchman, PhD, MD; Cheryl Hiddleson, MSN, BSN; Teresa A. Rincon, BSN, PhD; David Kirk, MD; Sherstin T. Lommatzsch, MD; Craig M. Lilly, MD


March 18, 2019

Antonios Liolios, MD

The first attempt at remote intensive care unit management (eICU) was reported in 1977, when Grundy and colleagues[1] used a two-way audiovisual link between a small private hospital and a large university medical center. Today, 42 years later, hundreds of ICUs are managed remotely, primarily in the United States.

So how cost-effective is this model? Does it have an impact on mortality or morbidity?[2] Does it reduce interhospital transfers?[3] Can patients and families relate to their "teledoctor"? Is the therapeutic patient-doctor relationship preserved or lost?

To answer these and other pertinent questions, we have asked some experts in the field to give us their opinions.

eICUs in the United States: Models and Provider Roles

Antonios Liolios, MD: Please tell us about your eICU model . How many patients do you typically cover? Is it mostly at night? How many hours is the typical shift?

Timothy G. Buchman, PhD, MD: The Emory eICU Center (Atlanta, Georgia) provides remote patient monitoring and care to critical care patients across five hospitals. The use of advanced audiovisual equipment and secure data information transmission enable that care. We have nurses working 24/7 and board-certified intensivists at night (7 PM-7 AM) and all day on weekends and holidays.

Timothy G. Buchman, MD, PhD

There are insufficient numbers of intensivists to provide 24/7 bedside coverage. Our model now relies heavily on critical care nurses, advanced practice providers (APPs), and intensivists at the bedside during weekday/daylight hours. Nights, weekends, and holidays, we use the eICU intensivist to support the APPs.

David Kirk, MD: Our WakeMed eICU Service (Raleigh, North Carolina) covers and provides nursing support to all ICU beds in our health system. The primary goal of our eICU is to get accurate information into our databases and to serve as a resource to our hands-on bedside staff. The eICU software system monitors clinical data, and nurses will check on a patient if an alert goes off. Under our model, a single eICU physician could cover at least 100 ICU beds.

Physicians provide coverage from 3 PM-7 AM in two 8-hour segments. An eICU secretary helps with patient flow, generating checklists, and coordinating tasks.


Teresa A. Rincon, BSN, PhD: At Sutter Health (Sacramento, California), the purpose of the eICU is to leverage critical care nurse and physician expertise across a large health system that spans more than 500 miles. It includes very small hospitals, large tertiary care centers, and several rural critical access hospitals. A secondary objective is to increase critical care standardization.

The Sutter model is nurse-driven and staffed 24/7 with experienced critical care nurses and clerical support personnel. Intensivist coverage ranges from 16-20 hours a day depending on the day of week. The two eICUs, located 90 miles apart, provide services to more than 400 adult ICU beds across 500 square miles. Nurses cover 40-60 patients and providers cover as many as 150 patients.

Sherstin T. Lommatzsch, MD: At National Jewish Health (Denver, Colorado), the eICU multidisciplinary team is composed of critical care physicians, nurses (usually RNs who have ICU bedside experience), pharmacists, and health unit coordinators. Some teams also include advanced nurse practitioners or physician assistants.

A typical shift is 12 hours, and the more common model is to have night coverage only. The number of patients a physician covers varies depending on the presence and number of bedside physicians. For example, for hospitals without critical care physicians at the bedside during night hours, an eICU physician may be responsible for 50 or more patients. More providers are available during the day, and an eICU physician may not be needed. This frees the eICU physician to work with bedside hospitalists at facilities that don't have bedside critical care. Together they review patients' clinical status, labs, and bedside nurse notes, and develop a daily plan.

Craig Lilly, MD: The integrative UMass Memorial Health Care eICU model (Worcester, Massachusetts) allows a team of specially trained critical care nurses and intensivists to monitor ICU patients 24/7 from a remote support center. The intensivist is responsible for approving the admission of new patients into our supported ICUs and performs case reviews for all newly admitted off-hour patients who have not been reviewed by bedside intensivists. Established patients are signed out by on-hour intensivists using a telephonically augmented electronic process.

This approach reduces operational costs because it does not require nurses to transcribe clinical information and lowers the amount of time-based rounding by the clinical staff. It also allows leveraging of our ICU intensivist with APP support. This is efficient because APPs can prescribe and thus can handle more transactions as single events than nursing extenders.

Liolios: What specialty support do you typically have, especially at night? Who performs invasive procedures (eg, central line, chest tube placement) if the need arises?

Buchman: Typically, one of our APPs will perform the invasive procedures. One hospital does not have APPs, and we ask an emergency medicine physician to carry out the procedures.

Kirk: We have bedside intensivists and APPs at some centers. Our top-tier APPs can perform all ICU procedures. At our smaller hospital, we collaborate with anesthesiology for intubations, arterial lines, and central lines.

Rincon: Critical care nurse practitioners, physician assistants, other APPs, and emergency department physicians generally provide these services. Sometimes an on-call physician will be called to the bedside.

Lommatzsch: Specialty support varies greatly from one facility to another. Most hospitals have consultative services (at least by phone) for the most commonly needed specialties, such as general surgery, cardiology, nephrology, and others. Radiology is always available even if radiologists are not reading images on-site.

Invasive procedures can be performed by a variety of individuals. Many of our hospitals have a procedure team of nurses, respiratory therapists, nurse practitioners, or physician assistants have been trained to intubate, place arterial lines, perform bedside echocardiography, and so on. At times, the in-house hospitalist or emergency department physician is asked to assist with intubation or chest tube placement. Finally, some facilities have an anesthesiologist on-call for central venous catheter line placement, arterial line placement, and/or intubation.

eICU Pros and Cons

Liolios: What are the major advantages and disadvantages of the eICU?

Cheryl Hiddleson, MSN, BSN (Emory eICU Center): The major benefit is the ability to expand available resources across a larger footprint. This allows better care for more patients than the standard brick- and- mortar model.

Timothy G. Buchman, MD, PhD, and Cheryl Hiddleson, MSN, BSN

Buchman: An advantage is that we can be anywhere at the speed of light (or close to it). A disadvantage is that we have only two out of the five senses that are available at the bedside.

Rincon: An eICU is a conduit for knowledge management and governance, which accelerates the translation of evidence into practice (something that takes 10-12 years on average in the United States). It also introduces new ways to conduct purposeful and continuous patient surveillance in order to catch " badness" before it happens. Why not put expert nurses who have astute situational awareness, experiential knowledge, and high-level cognitive skills into a controlled environment, where they have clinical decision support tools designed to help them predict and prevent decompensation?

Major disadvantages are the upfront costs associated with starting up an eICU and the ongoing costs associated with keeping the program running. It would help if the Centers for Medicare & Medicaid Services and third- party payers allowed reimbursement for services. Another big disadvantage is that not all states allow centralized credentialing; to practice across state lines requires a different license.

Sherstin T. Lommatzsch, MD

Lommatzsch: An eICU provides care through a central command center otherwise known as the "bunker model." This is a dedicated off-site central monitoring station designed with multiple workstations, each of which is equipped to allow meticulous monitoring of ICU patients. A team member is able to view live telemetry and radiographic images, access electronic medical records, and enter patient care orders. The eICU software platforms can quickly identify patients whose clinical status is deteriorating, and alert providers. Using the software platform, a physician has immediate audiovisual access to patients and can discuss clinical concerns with bedside staff. Cameras in a patient's room allow the provider to zoom in and view such details as a patient's ID bracelet and records displayed on monitors.

  • Rapid access to a critical care team 24/7 for patients in any type of hospital or extended care facility—rural hospitals, community hospitals, and tertiary care hospitals.

  • Ability to track changes in patient clinical status using computer software alerts.

  • Larger critical care trained staff to assist in care management, from therapy changes to family conferences.


  • It can be awkward communicating with patients and families about sensitive issues —for example, transitioning to comfort care —through a video camera.

  • Sometimes it may be difficult for a patient or the family to understand what the eICU practitioner is saying, so the bedside nurse may need to repeat statements. Patients or family members may also have a hearing impairment.

  • In acute clinical situations, there may be a lot of background and side conversations, making it challenging to speak to a family member also present in the room.

  • The initial financial investment is high.

Kirk: I could talk about the advantages all day. Tele-critical care allows expensive ICU resources to cover more beds accurately. Moreover, the data it generates and tracks are essential for improving complex ICU care.

The greatest disadvantage is that an ICU still needs someone who is physically available and can perform critical care procedures. The major issue I see with eICU development across the United States (and internationally) is a race to the cheapest system possible. Typically, this is accomplished by pooling physicians from multiple different regions without consistency. I believe that successful eICU integration requires the bedside team to become very familiar and comfortable with a core set of eICU physicians. Teamwork requires familiarity. The eICU and ICU workflows and processes must be integrated to provide superior outcomes.

eICU Impact on Critical Care Patients

Liolios: What is the impact of eICU on mortality/morbidity, length of stay, or other endpoints of patient care?

Hiddleson: We have experienced a decrease in mortality rate based on APACHE IVa, of 52% in the ICU and 40% for patients after they move to the ward.[4,5]

Kirk: After the initial rollout, we saw a significant reduction in mortality and length of stay, which has been sustained. The more accurate data measurements provided by the eICU system enabled us to identify other challenges, such as glycemic control, in-hospital renal failure rates, and low tidal volume ventilation compliance. When those occult issues were identified, we dramatically improved our outcomes. In fact, when our eICU software benchmarks us against other eICU centers, we are number one in the country in best practice measures.

Lilly: Effective eICU requires competent bedside nursing and the availability of on-hours physicians to perform ICU bedside procedures and create viable care plans. Implementation without proper bedside support can be counterproductive.

Craig Lilly, MD

The eICU system reports provide a real- time second-level expert review of current care plans that targets patients at the time of, and often before, physiologic instability requires emergent unscheduled evaluation and management service delivery. Consistent with common sense, mechanisms of targeted secondary review and early intervention are effective with adverse events, preventable mortality, and differences in the ability to deliver the same off-hours services that are provided during on-hours. Outcomes also depend on the ability of the institution to enable the eICU team to perform real-time interventions by having the off-site team accepted as part of the on-site ICU care team.

Liolios: Does eICU reduce the need for hospital transfers of critically ill patients?

Hiddleson: Four of our hospitals have expertise that allows them to keep most patients unless there is no bed available. In the first 6 months after implementation of the program, one small rural hospital we provide services to reported the following:

  • 40% decrease in transfers to other hospitals for severity of patient condition;

  • 15.3% increase in admissions;

  • 21.5% increase in daily census;

  • 0.5- day decrease in length of stay; and

  • 8% increase in patient days.

Kirk: Other than reducing the amount of in-house intensivist coverage required, the biggest financial benefit for many eICUs is preventing hospital transfers. Controlling which patients go where is extremely important to many large hospital systems.

Rincon: It has done that at UMass, according to our studies.[6,7] If needed, the eICU can assist in stabilizing patients for transfer and play a role in finding a bed at a hospital that can provide any additional treatment patients need.

Lommatzsch: That is one of the goals of eICU medicine.If a patient requires emergent hemodialysis, the patient will be transferred to a facility that can provide this service.

On a Personal Level, Is Working in an eICU Stressful?

Liolios: How stressful is working in an eICU? What is most challenging in the eICU environment for the healthcare provider?

Hiddleson: All of our physicians and nurses will say that working in this environment is much less stressful than being in the actual brick- and- mortar ICUs. There is something about being removed from the heightened emotions and activity at the bedside. It allows us to be more objective and observe from a different viewpoint. We also have a complete view of the room and everyone in it, so we can see what those in the room cannot.

The biggest challenge is not being in total control of a patient's treatment. We offer suggestions and a plan of action, but the bedside physicians can change that plan because they are the primary physician of record and guide the plan of care.

Buchman: The other challenging aspect is that it is primarily night work. Candidly, intensivists—especially older ones—find the night work exhausting. This is why we set up the Night-Into-Day program, so that we deliver nighttime care from as far away as Australia.

David Kirk, MD

Kirk: Critical care is incredibly stressful in general. What causes the most physician burn out aren't the stressful patient encounters ; it's being paged excessively, the inability to get into a great workflow, constant interruptions, and being pushed too hard when physically exhausted. The eICU reduces all of these stressors. The unique challenge for an eICU physician is sitting for long periods.

That's why many eICUs like ours have treadmills and standing desks to keep our staff fit.

Some eICU providers find it a bit awkward to communicate over a camera instead of in person. Nothing is ever going to beat the human touch, but most eICU providers find telecommunication better than what a simple phone call would allow.

Lommatzsch: eICU work can become stressful, because a physician is responsible for a large number of patients. One needs to respond with detailed knowledge when asked clinical questions. The physician may receive several questions at one time or multiple pages in a short period. This requires quick and strong triage skills and efficient time management.

Furthermore, one should remember that the level of patient acuity is the same regardless of provider location. Running a code or stabilizing a deteriorating patient carries the same degree of care intensity and may be required throughout an entire shift.

Liolios: How do on-site providers of the local ICU feel about being monitored by an eICU team?

Hiddleson: In the beginning, they see us as intruders. As time goes on, they begin to realize we are there to support, not supplant, them. Attitudes change, and we have had some real champions emerge in all of our ICUs.

Buchman: The patients are being monitored, not the staff. It takes a few weeks for staff to realize that the camera is off and facing the wall most of the time. Once they recognize that we are trying to help them do their work more safely and effectively, many of the concerns vanish.

Kirk: We are all part of a single team. If we have an issue, we talk it out. Ultimately, the bedside team makes the final call, but it is rare that it escalates to a severe situation. The health system as a whole has a good reputation for respectful feedback and dialogue, so it's not a pain point for us.

Teresa A. Rincon, BSN, PhD

Rincon: As with most new care delivery systems, eICUs have taken some time to catch on. It is difficult for some clinicians to understand how someone can provide high-level care without being in the same room with the patient. eICU technologies have allowed intensivists to become part of the care in hospitals that didn't traditionally have intensivists.

In addition, eICU technologies have given expert ICU nurses and intensivists transparency into how care was being delivered across multiple environments where critically ill and injured patients were cared for. Although this transparency enabled health systems to find opportunities to improve care, bedside teams were not always accepting of this. Donald Berwick has described how clinicians honor and protect unscientific variations in care because that is the way they have always done it.[8] I believe eICUs exposed these variations, and that made people uncomfortable. It is important to develop policies and procedures that include a "chain of command" and an accountability structure.

Lommatzsch: There have been mixed impressions of telepresence in the ICU. Initially, some providers thought it led to an improvement in the overall ICU team performance; knowing others are watching increases motivation to perform better. Others believed their clinical decisions were being questioned or challenged. Over time, as relationships are built and both the bedside and eICU providers learn to maximize the benefits of each service, the concept of being monitored tends to dissipate.

Malpractice and Medical Liability

Liolios: Is the malpractice risk higher when practicing in an eICU? Any particular risks for physicians and patients? How does medical liability work in the eICU environment?

Buchman: The malpractice risk is not particularly high and certainly is less than being at the bedside. Remember, we carry out the care plan of the local attending physician. Moreover, when we have a new situation or a new finding, we have tools to immediately document it. I tell the physicians to keep the documentation simple; keep it to what you saw/heard/learned (what was happening, what actions you took, what the outcome was of those actions).

Kirk: At WakeMed, we are all members of the same group, and we are self-insured, so it's not an issue with us. The eICU provides an additional layer of checks and balances, so it reduces complications and at-risk situations. Our compensation and benefits are all the same.

Rincon: Any time a physician's name is on the medical record, they can be named in a lawsuit. Good documentation and having good escalation processes in place are always important. This helps patients and families feel like they are getting better care.

Patient Perceptions of eICU Care

Liolios: How do patients perceive the telepresence of their doctor or nurse? What impact does this have on the patient-physician relationship? On families?

Hiddleson: We have found almost 100% of the patients and families view this as a positive experience because there is someone else watching out even when the nurse or physician in the ICU is busy with something or someone else. We live in a time when, walking down a sidewalk anywhere in the world, you see someone using a cell phone; many of them are having a face-to-face conversation with someone while walking. Telepresence is new to healthcare, but not new to our everyday personal lives.

Buchman: Patients and families seem to enjoy the interactions and knowing that there are additional eyes and ears keeping an eye out for the safety of their loved one.

Kirk: Patients love the backup. The fact that someone is monitoring them when the nurse walks out of the room is very comforting. Our eICU nurses explain the system to the family. I have never heard of a family complain about it.

Rincon: I have been involved in providing critical care from an eICU since 2003, and I can count on one hand how many times patients or their families had a problem with having an extra set of expert eyes on them. But I have lost count of how many times I have had to deal with upset bedside clinicians.

Lommatzsch: My personal impression is that the millennial generation seems to work better with telemedicine, but patients who are used to the medical team being at their bedside take more time to become comfortable with the concept.

Liolios: eICU does not allow you to physically examine the patient or perform an ultrasound. How limiting is it? Is telerobotics a solution?

Hiddleson: This is less of a problem than you might think. We can hear a Doppler instrument. We can see an ultrasound screen. What is needed is some basic training of bedside staff so that they can put the probes on reliably.

Rincon: Even without telerobotics, a physician anywhere in the world can use streaming video to watch an ultrasound and interpret it. The audiovisual solutions we have today enable surgeons to provide guidance and advice to other surgeons. We are lagging behind in technology, and this leads to clinicians and patients pulling out their cell phones when they need advice, which introduces issues with security and potential breaches in patient confidentiality. Ultrasound, radiology, and other diagnostic imaging technicians have been sending images to radiologists for decades.

Lommatzsch: None of these are terribly prohibitive. Many hospitals have in-house ancillary teams trained to perform bedside procedures. Also, physicians have traditionally relied on a nurse or other provider to perform the bedside physical exam. Telerobotics is a helpful solution to this concern. The use of telerobotics is even being explored in surgery —physicians performing surgery remotely with the use of an on-site robot.

Financial and Confidentiality Concerns of an eICU

Liolios: eICU implementation requires extensive infrastructure and Health Insurance Portability and Accountability Act (HIPAA) compliance, among other things. How cost-effective do you believe the concept is overall?

Buchman: It is costly, but the alternative is finding and hiring onboarding intensivists. There are not enough intensivists, especially ones willing to work nights.

Kirk: The hardware components and interfaces are the most expensive initial cost, but that cost is a true barrier to many systems. My hope is that as audio and video solutions-makers compete, it will drive down the costs. Overall, I believe the long-term return on investment is easy to justify. The need for fewer intensivists is the most obvious value, but many studies have shown that eICU systems allow for larger case volume, higher case revenue relative to direct costs, and shorter lengths of stay.[7] In our system, we saved a lot of money by reducing complications, such as the need for transfusions, and by improving our quality goals.

Lilly: The costs of imperfect adult critical care are so large compared with the current costs of ICU telemedicine programs that even modest improvements in the availability of higher- quality care result in favorable financial outcomes for nonfederal hospitals. In the same way that improved telecommunication technologies have improved our daily lives, they are becoming more ubiquitous in ICU practice because they are often more efficient than older technologies. The key issue is how ICU telemedicine is implemented. Also, eICU intensivists need fair compensation for the evaluation and management services that they provide.

Liolios: What are the confidentiality issues? Almost all online systems are vulnerable to cyberattacks, so how safe is eICU from this perspective?

Buchman: We are probably more secure than most health IT systems, but no one is invulnerable. We have secure lines at all of our sites.

Hiddleson: When we set up this system, we extensively explored the security of patient information and so has the vendor we work with. Our eICU monitoring platform is inside the Emory network and can only be accessed by those granted permission. The video sessions are not stored and are similar to using Skype. The session is live while we are in the room, then it is disconnected. Each room also has its own IP address that we connect to.

Kirk: I don't know of any unique confidentiality issues. All aspects of a health care system are targets for digital criminals. I don't believe an eICU system provides any larger an attack vector than any other technology.

Rincon: All systems are susceptible to cyberattacks. The eICU is as safe as any other health information system or electronic health record.

Liolios: How often do you face network issues, dropped connections, and system malfunction in general?

Hiddleson: Technological difficulties do arise; some are local and others are inherent in the software/hardware used for providing the care. We have had very few "network/dropped connection" issues as opposed to software/hardware issues. The latter is something we deal with on a monthly basis, but the technology is updated regularly and we feel we can overcome that obstacle with the newer in-room system being developed.

Buchman: The technology is more reliable than most I have seen. The network and software are stable. Our biggest challenges have been maintaining the in-room systems. Fortunately, there are "next- generation" in-room systems coming.

Kirk: With the support of our excellent IT teams, we have had very few problems with the hardware or software.

Rincon: In both of the programs I have been associated with, dropped connections and system malfunctions have been rare. California spent around $300 million on a project to increase broadband coverage more than a decade ago. Investing in a hosted platform for data center resources or in an entity-owned/operated data center are the two options. High- functioning and secure data centers require large capital and ongoing investments. Many telehealth platforms purchase very secure data center hosting from such entities as Amazon (cloud-based technologies).

Liolios: The concept of eICU has flourished in the United States compared with the rest of the world. Why do you think this is the case?

Buchman: We think it is a model that will grow even more quickly when Medicare starts paying for eICU services. eICUs are expensive. Few hospitals in nations outside the United States have the funds necessary to support eICU hardware, software, IT, and operations.

Hiddleson: Each entity, be it an ICU or a country, believes it is unique and different from all of the others and has its own set of workflows and standards for what it believes is the best patient care. Meeting with clinicians and healthcare leaders from many different countries, I have learned that although there are different practices and workflows, we all have a shared objective, and that is to provide the best possible care to our patients. In the United States, a hospital can make an independent decision to implement and pay for these services rather than rely on or wait for a government agency or authority to approve the decision.

Lommatzsch: ICU staffing in the United States has been a challenge over the past two decades for several reasons, including a deficit of critical care providers and a high degree of provider burnout. Multiple supervisory groups have strongly recommended or required proof of quality measure compliance and improvement in patient outcomes, such as hospital length of stay and mortality. The eICU was designed not to replace the bedside presence, but rather provide a means to relieve the on-site staff of some responsibilities by being a "second set of eyes" when managing an increasingly complex patient population. In addition, the software platforms can collect data to monitor compliance when providing treatment (for example, deep venous thrombosis prophylaxis).

I don't have a great answer for this, as I'm not as familiar with the needs for ICU care in Europe versus other countries.


It appears that the concept of eICU is rapidly expanding. Despite the lack of the on-site presence of the physician or nurse, there is evidence that current eICU models offer cost-effective and safe medical care. There are some concerns that further research is needed before drawing definite conclusions, yet the number of eICUs is increasing primarily in the United States, counterbalancing the nationwide physician shortage.[9] As technology is constantly evolving, the future appears very promising. We live in interesting times.

I would like to thank our expert interviewees for their time and for sharing their experience with us.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: