The Leapfrog Initiative

As neurocritical care has come to the forefront in the care of seriously ill or injured neurologic patients, so has the concept of specialty critical care medicine in general become more widely accepted over the past two decades. In its 1999 report, “To Err Is Human: Building a Safer Health System,” the National Academy of Sciences’ Institute of Medicine (IOM) estimated at least 44,000 and as many as 98,000 people die each year in hospitals from preventable medical errors. In a subsequent report, “Crossing the Quality Chasm: A New Health System for the 21st Century,” published in 2001, the IOM called for a redesign of the U.S. health care system to help achieve the goals stated in the first report and stated: “You can’t get there from here.” This means that the problem of quality improvement will not be solved by doing more of what has been done in the past, or what is being done currently, but rather by using new tools to help solve core problems. The report also called for use of information technology to help drive the quality improvements. In response to the IOM reports, The Leapfrog Group, an organization of Fortune 500 companies and large private and public health care purchasers, identified four hospital quality and safety practices that could help prevent errors and save lives, including intensive care units (ICUs) staffed by intensivists 24 hours a day and all week long. Estimates suggest that 50,000 lives and $4.3 billion may be saved each year in the United States through 24-hour intensivist staffing.

Too few ICU patients are cared for by critical care specialists; not every hospital has the patient volume or the financial resources to hire dedicated intensivists. In addition there is a shortage of trained intensivists. About 6000 intensivists are estimated to be in practice in the United States today, and less than 15% of all ICU beds have dedicated intensivists. There are approximately 6000 ICUs in the United States, which represents 5% to 10% of hospital beds. On a daily basis, these ICUs care for approximately 55,000 patients. Only one third of these patients are treated by an intensivist, either as the primary physician or as a consultant. In addition, there is a current nursing shortage and aging of the most experienced critical care nurses. The first of the “baby boomers” are reaching retirement age at a time when technology provides for unprecedented life expectancy and recovery from illness and injury. However the relative lack of intensivists and critical care nurses is not likely to improve.

ICU Outcomes with an Intensivist Model

Many variables determine outcomes in critically ill patients, including the primary diagnosis that leads to admission, patient characteristics, the evolution of secondary disorders, and the care that these patients receive. There is a wealth of data to suggest that intensivists, having been trained specifically in the care of the critically ill patient, are better at management of comorbidities and avoidance of complications, and so can improve outcomes in ICU patients. This has stimulated considerable interest in ICU performance. Patients managed by a dedicated critical care service in a single hospital surgical ICU have a shorter length of stay (LOS), fewer complications, and lower hospital charges than patients cared for by general surgeons and house staff. In 1999, a study by Pronovost et al found that daily rounds by an intensivist were associated with improved outcomes in patients who underwent abdominal aortic surgery. Others have demonstrated reduction in mortality, including a systematic literature review in which 16 of 17 studies showed a decrease in hospital mortality with high-intensity ICU staffing (the intensivist was the primary attending or there was a mandatory critical care consultation for all ICU patients). However, Levy et al have debated the benefit of the intensivist model and suggest that intensive care provided by critical care specialists does not improve outcomes and may be detrimental to patients. In this study, hospital mortality was higher for patients managed by critical care physicians, even after adjustment for severity of illness and other factors that may have influenced the probability of being cared for by a critical care physician. However, this study is observational in nature, making cause-and-effect conclusions difficult. Furthermore, heterogeneity beyond that accounted for by applying adjustments may limit result interpretation.

Most studies identify ICU physician staffing as an opportunity to reduce in-hospital mortality. Assuming 3.5 million patients are admitted each year to ICUs in the United States and a 40% reduction in mortality with dedicated intensivists, it is estimated that 54,000 lives could be saved with a full intensivist model as described by the Leapfrog Group. However, only 10% to 15% of U.S. hospitals have such a program in place, in large part because of a shortage of intensivists. This shortage of intensivists is expected to worsen in coming years. The technology tools of the e ICU program, including virtual presence through audiovisual technology, provides an alternative means to bring intensivists expertise to hospitals that lack sufficient numbers of intensivists.

Telemedicine to Leverage Scarce Resources

Grundy and colleagues first introduced telemedicine as a tool for critical care in the late 1970s. Interactive television was used to provide consultation with university-based critical care physicians for patients in the ICU of a 100-bed hospital that lacked physicians with critical care expertise. During an 18-month period, 1548 telemedicine “visits” were made to 395 patients. The process was thought to be a reliable method to extend the availability of specialist expertise and to be beneficial to patient care. However it was another 15 years before the concept of telemedicine in critical care reemerged, albeit in a very different form. In 2000, Rosenfeld reported that introduction of around-the-clock intensivist oversight through telemedicine produced reductions in mortality and LOS similar to those reported by implementation of an intensivist program. Subsequently, Breslow and Rosenfeld founded VISICU, a start-up company that provided the first continuous, remote, centralized, computer-assisted intensivist program for critical care. Sentara Healthcare in southeast Virginia was the first to implement this program in 2000. After the first year, hospital mortality decreased by 27%, ICU LOS by 17%, and hospital LOS by 13% compared with the previous year. Variable costs decreased and hospital revenue increased because of increased patient throughput. These improvements have been sustained over subsequent years.

What is telemedicine? Because onsite intensivist staffing is associated with reduced mortality, telemedicine in the ICU has evolved in part to address limited clinician supply through a technologic solution. ICU telemedicine can come in several forms that range from consultations regarding individual patients to continuous offsite monitoring for the whole ICU, but all combine audiovisual technology (videoconferencing technology and telemetry) and electronic medical records to provide critical care from a remote site. It is estimated now that about 10% of the total ICU beds in the United States use this technology. The telemedicine coverage may come in several forms, including (1) type of coverage (proactive, reactive, or mixed), (2) structure (a single consulting “hub” that covers multiple ICU “spokes,” a “parent” hospital that covers another hospital in its system, or a network of clinicians who monitor from multiple sites), and (3) hours of operation (24 hours or night only). In part, the telemedicine setup depends on physical ICU staffing (open or closed), intensivist availability at physical ICUs, number and type of ICUs, and hospital resources or affiliation with a tele-ICU vendor.

Telemedicine and General ICU Outcome

Several factors may influence how a telemedicine program affects clinical outcome: (1) integration of the information systems of the tele-ICU and the monitored units, (2) physician acceptance, and (3) how the telemedicine system is used. In addition, other delivery systems and work-process innovations rather than technology alone can impact outcome. Several studies have examined how telemedicine in general ICU care may influence outcome. Some suggest a benefit in the sickest patients, because these may be the patients most likely to have unexpected events. In 2011, Young et al published a systematic review and metaanalysis that described the clinical impact of ICU telemedicine. They identified 13 studies, only seven of which were published in peer-reviewed literature that examined the association between ICU telemedicine and clinical outcomes. Each study used a before-and-after design, but severity of illness and case-mix was not always adjusted for. On metaanalysis there was an association between reduced ICU mortality and LOS and a telemedicine ICU. However, there was no association between admission to an ICU under a telemedicine model and in-hospital mortality or hospital LOS. The confidence intervals were wide, and so a clinically significant effect could not be excluded. In addition, most of the studies included in the analysis used telemedicine at night to enhance patient safety in ICUs that already had high-quality daytime staffing. In some of the studies, authority was delegated to the telemedicine physicians only in life-threatening events. The value of telemedicine may not be so much in improved safety but rather in the ability to provide evidence-based care on a 24-hour basis in a proactive fashion. Furthermore the benefit may be greatest in smaller ICUs with limited staffing where the telemedicine intensivists can interact with bedside staff to optimize proactive evidence-based management rather than reactive care.

Evolution of VISICU

Since the original Sentara project, VISICU has implemented more than 35 programs in 26 states, bringing the e ICU program to more than 200 hospitals and more than 8% of the adult ICU beds in the United States. More than 300,000 patients now have their ICU care supplemented by a remote care team annually ( Fig. 44.1 ). These e ICU programs can contribute to potential clinical and financial benefits including at large academic medical centers and level I trauma centers ( Table 44.1 ). For example, in the 30-bed surgical trauma unit at the Hospital of the University of Pennsylvania, Kohl et al compared care in the year before implementation of the e ICU program with care in the years following. In an ICU that already was staffed with dedicated intensivists, fellows, and residents, significant reductions in ICU (63%) and hospital (45%) mortality were observed; that is, 336 lives were saved. Hospital and ICU LOS also were reduced with an estimated $5.8 million savings in direct costs. Studies have also demonstrated an e ICU program to be useful in rural hospitals. For example, in the Avera-McKennan hospital of the Avera Healthcare System in South Dakota, significant reductions in standardized (APACHE III) ICU and hospital mortality ratios (0.63 and 0.67 to 0.25 and 0.53, respectively) and ICU and hospital LOS were observed with use of an e ICU program.

e ICU programs are used to monitor critically ill patients in intensive care units, emergency departments, and postanesthesia care units, as well as other areas in university hospitals, community hospitals, and critical access hospitals.

Setup of an e ICU

A variety of clinical process changes accompany the introduction of an e ICU. The e ICU program is not a single technology or a single product. It is a set of tools and an innovation stream that results in a technology-supported team approach to the care of ICU patients. It can be transformational because it reorganizes work processes and care models. The e ICU program does not replace bedside doctors and nurses or significantly alter their roles and responsibilities, but instead adds an additional layer of support for the care of the most critically ill patients whose conditions can, and often do, change rapidly. Critically ill patients require frequent assessment throughout the day and night, which is probably an important mechanism by which intensivists improve care by providing regular availability and focus on the constant titration of critical care. This allows for more rapid interventions such as fine-tuning therapies and prompt detection of emerging problems, which if not recognized early can lead to life-threatening complications. These tasks (assessment/titration/intervention) represent a core activity of the remote care team and allow health systems to leverage existing intensivists to provide 24/7 coverage to all critically ill patients.

The e ICU program is designed to ensure intensivist involvement in care for every ICU patient, and to provide the intensivists with tools that improve efficiency. Cardiologists frequently say “time is heart.” Stroke specialists now say “time is brain.” In trauma there is the concept of “the golden hour” to prevent major systemic damage. In the ICU, there frequently are many important things going on such that only emergent situations truly generate rapid responses and many pathophysiologic processes progress further than is ideal before they are readdressed. Often the traditional systems that are in place to take care of patients become overwhelmed. In this setting, improvements in efficiency lead to improved care as patient-related events are addressed at an earlier stage in their evolution.

e ICU Technology

An essential part of the e ICU system that leads to improved efficiency is the technology infrastructure that manages the flow of clinical information from the hospitals to the e ICU center. This includes the real-time data from bedside monitors, and all the supporting data (i.e., lab results, medications, patient registration and administration data, and medical images such as x-rays and magnetic resonance images), and audio and video streams for remote observation of patients ( Fig. 44-2 ). Using evidence-based algorithms and clinician-set parameters, the software parses patient data in real time to identify incipient clinical events and alerts clinicians before many significant adverse events develop. Smart Alert prompts evaluate patients’ electronic monitoring data in real time and flag potentially worrisome trends and readings that exceed specified thresholds ( Fig. 44-3 ). This focuses the intensivist’s attention and can promote timely intervention to avoid crises. The e CareManager System provides a “dashboard-like” interface that helps the remote intensivists coordinate patient care with doctors and nurses at the bedside. An online decision-support tool called The Source draws on the latest evidence-based care guidelines. In addition, the e CareManager system is built upon a relational database, which allows data organization to drive best practices such as simplified multidisciplinary rounds and the tracking of performance according to evidence-based best practices.

e ICU clinicians monitor patient data from a remote site, using Smart Alerts, real-time monitor data, laboratory results, and other ancillary test results to intervene on the patients’ behalf.

Smart Alerts software continuously track vital signs and trends, and notify e ICU clinicians of aberrancies, enabling quick action and thereby potentially preventing problems from escalating. H, High; HR, heart rate; L, low; MAP, mean arterial pressure; O ₂ Sat, oxygen saturation; RR, respiratory rate.

Clinical Transformation Process

The real significance of the e ICU program lies in the way it changes care delivery. The technology is only enabling; the key is a clinical process transformation that accompanies the technology. The hospital admitting or managing physician remains the attending physician of record and is responsible for the patient care plan and determining which on-site consultants assist with care delivery. This concept is particularly important in understanding how an e ICU program staffed with intensivists can be of great assistance in the care of subspecialty patients such as the neurocritical care patient. It is not the role or responsibility of the e ICU team to determine the specific needs of an ICU patient or to define the primary course of treatments; those evaluations and decisions are best done at the bedside by the physicians and surgeons who know the patient. The role of the e ICU team is to closely monitor the patient’s data for evidence that care needs are changing or that expected data are available, and then to alter or initiate care in accordance with the previously expressed plan and direction of the bedside health care providers. If there is a question whether the original plan is still best considering the changes or new data, the e ICU intensivist seeks clarification from the managing bedside team and then resumes giving support accordingly.

e ICU Team Responsibilities

The e ICU team can regularly fine-tune the acute care of ICU patients through observation of trends. Consider a patient who presents with an intracerebral hemorrhage. The plan of care following initial evaluation by a neurologist and neurosurgeon may include admission to an ICU with an ICP monitor, frequent follow-up neurologic evaluation, blood pressure control, and intubation for airway protection. The remote e ICU physician monitoring the patient follows the plan of the bedside team and monitors ICP and cerebral perfusion pressure (CPP) in the patient, and regularly confers with the bedside nurse via camera to determine stability in the patient’s neurologic findings. Several hours into the patient’s ICU course, the e ICU staff is alerted via Smart Alerts (see Fig. 44-3 ) that, although the patient’s vital signs are not out of the set range of acceptable values, the trend has changed such that the heart rate has dropped 25 beats per minute and the mean arterial pressure has risen 30 mm Hg. Inspection of the ICP trend reveals that it has increased. Concerned about a potential Cushing response, the e ICU physician contacts the bedside nurse, orders an emergent repeat head computed tomography (CT) scan, and alerts the attending physician of record, consulting neurologist, and neurosurgeon. While awaiting formal input from the neurologic experts, the e ICU physician begins simple measures to treat elevated ICP, such as sedation and elevation of the head of bed. When the CT scan is completed and demonstrates extension of the hemorrhage with worsening edema, the e ICU intensivist confers with the neurologist or neurosurgeon immediately to discuss further patient treatment.

The e ICU team also can intervene in emergencies and manage best practices (e.g., glucose control, deep vein thrombosis (DVT) and stress ulcer prophylaxis, low tidal volume ventilation). In many e ICU programs, the attending physician defines in advance the requirements for communication by the e ICU center before nonemergent actions are taken for their patients. This may range from specific review of all nonemergent decisions to allowing the e ICU intensivist to act as a surrogate decision maker, calling the bedside team as indicated by specific conditions and degree of expertise in those decisions. Examination of communication methods suggests better outcomes when the remote care team is allowed more autonomy to address new problems and institute best practices. Independent of communication preferences, the e ICU physician reviews all patient data at regular intervals (from hourly for the sickest patients to every 4 to 6 hours for the most stable). Each e ICU caregiver monitors and delivers care from a workstation comprised of several desktop monitors that provide access to the eCareManager application suite, including audiovisual tools, the hospital information system, picture archiving and communication system, and remote-view applications for the bedside monitors (see Fig. 44.2 ). Caregivers consist of physicians, nurses, clerical staff, and, in some programs, a pharmacologist or pharmacist. Staff numbers vary based on the number of beds in the network. With smaller networks (less than 70 beds being monitored), the remote care team is comprised of one intensivist and one critical care nurse. As the number of monitored beds increases, the number of care providers increases, with additional nurses typically added after 60 to 70 beds are reached and a second physician (not necessarily a second intensivist) added after 100 to 120 beds. There can be variability in how tasks are allocated among the various care providers, however, several core features must be in place to achieve quality goals, such as a comprehensive daily care plan that addresses key clinical issues (e.g., ventilator weaning, nutrition, risk assessment, social issues). The care plan is developed by the on-site team and optimally encompasses the input of all relevant care providers. The e ICU team monitors the care plan and assists with implementing the various components, and can take primary responsibility for processes such as compliance with best practices. High compliance rates can be achieved by centralizing these essential activities and assigning them to dedicated personnel. Because the remote team has no other clinical responsibilities, it is efficient and continuously attentive. The responsibilities of the remote team include (1) continuous monitoring of the progress of each patient; (2) titration of therapies, as necessary, to achieve care plan objectives; (3) identification of emerging problems and initiation of appropriate countermeasures; (4) facilitation of communication among members of the care team; and (5) responsibility for best practice compliance.

Several lines of evidence suggest that allowing the remote team to fulfill these responsibilities is associated with improved ICU practices in both urban and rural hospitals, including many evidence-based practices, such as DVT prophylaxis compliance, glycemic control, and 6- and 24-hour care bundles in patients with sepsis, often resulting in reduced mortality. In addition to best practice improvements, improvements in other aspects of patient care have been observed with an e ICU program. For example, Shafer et al observed a reduction in the number of cardiopulmonary arrests in their ICUs and the number of patients who died of cardiopulmonary arrest once an e ICU program was instituted at their hospital.

The Impact Beyond the ICU

There are several potential advantages to a tele-ICU that extend beyond the ICU. For example, tele-ICU care may contribute to more judicious decisions about which patients are to benefit from ICU admission and which patients should receive palliative care, that is, better resource use. Furthermore, any telemedicine program means that there is a large volume of electronic data that is collected in real time. Used correctly, this can drive clinical practice improvement through “instantaneous” quality metrics. Alternatively, the aggregated data can be used to answer important research questions. For this to be a reality, however, informatics infrastructure and protection of patient privacy must be addressed.

A remote e ICU monitoring program can help increase attending physician oversight and availability. This may be particularly helpful in hospitals with residency programs and enhance education. House officers have an immediate point of contact to discuss complex cases or even ask simple questions. Often a new house officer may be hesitant to call an attending physician in the middle of the night, especially if it is not clear to the house officer that the question is urgent. The e ICU physician can serve as a buffer for the house officer, even if it is simply to advise the house officer whether an issue for the bedside attending physician should be addressed immediately or can wait until morning. Intensivists in the e ICU center have time to educate house officers and can quickly email or fax relevant information including publications, using The Source for topical information or as a conduit to the medical literature and papers of interest. Staff in the e ICU center should be regarded as a resource for house staff, but should not usurp house staff autonomy. In that regard, e ICU physicians are trained to include house officers in any decision making that is of educational value. In addition, the presence of the e ICU clinicians can help reduce the stress that critical care nurses often experience by providing them a constant safety net. Together these resources for nurse and house staff can provide better outcomes for ICU patients by eliminating the wait time often needed to reach a physician. An e ICU program allows an “ICU without walls”; it is already available in some emergency departments (EDs) and postanesthesia care units (PACUs) and is being deployed through wireless mobile units to assist with rapid response in general acute care settings. Several e ICU programs also are adding stroke assessment and intervention programs using their e ICU technology and operational backbones. Office-based or at-home stroke specialists then can work collaboratively with the existing e ICU teams and so more rapidly assess and intervene in acute stroke cases from the ED through the ICU care stages.