Tele ICU: paradox or panacea?

Best Practice & Research Clinical Anaesthesiology 23 (2009) 115–126

Contents lists available at ScienceDirect

Best Practice & Research Clinical Anaesthesiology journal homepage: www.elsevier.com/locate/bean

9

Tele ICU: paradox or panacea? Adam Sapirstein, MD, Assistant Professor a, *, Nazir Lone, MD, Patient Safety Scholar b,1, Asad Latif, MD, Instructor a, James Fackler, MD, Associate Professor a, Peter J. Pronovost, MD, PhD, Professor, Departments of Anesthesiology and Critical Care, Surgery, and Health Policy and Management, Medical Director, Center for Innovations in Quality Patient Care, Director, Quality and Safety Research Group a a

Department of Anesthesia and Critical Care Medicine, The Johns Hopkins University School of Medicine, 600 North Wolfe Street, Baltimore, MD 21287, USA b The Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe Street Baltimore, MD 21205, USA

Keywords: information technology leapfrog group manpower safety organization

Telemedicine has been studied in the intensive care unit for several decades, but many questions remain unanswered regarding the costs and the benefits of its application. Telemedicine ICU (Tele-ICU) is an electronic means to link physical ICUs to another location which assists in medical decision making. Given the shortage of intensive care physicians in the US, Tele-ICU systems could be an efficient mechanism for physicians to manage a larger number of critical care patients. This chapter will examine the current state of telemedicine in an age of rapidly expanding medical information technology and increasing demand for intensive care services. While we believe that the future of TeleICU is promising, there are multiple issues that must be addressed to increase the benefit of Tele-ICU. Tele-ICU is expensive to deploy and use, it may add burdens to existing intensivists, and it requires organizational and culture changes that can be difficult to accomplish. Ó 2009 Elsevier Ltd. All rights reserved.

* Corresponding author. Department of Anesthesia and Critical Care Medicine, The Johns Hopkins University School of Medicine, 600 North Wolfe Street, Halsted 842B, Baltimore, MD 21287-7294, USA. Tel.: þ1 410 614 9920; Fax: þ1 410 614 1776. E-mail address: [email protected] (A. Sapirstein). 1 Current address: The University of Missouri-Columbia, One Hospital Drive, 65201, MO, USA. 1521-6896/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.bpa.2009.02.001

116

A. Sapirstein et al. / Best Practice & Research Clinical Anaesthesiology 23 (2009) 115–126

Introduction The application of telemedicine techniques to intensive care unit (ICU) care was first described in 1982 and has undergone continuous development since then.1 Tele-ICU is the concept and practice of using electronic means to convey information from an ICU to another location where it can be used to aid medical decision making. Given the shortage of intensive care physicians in the U.S, tele-ICU systems could help provide an efficient mechanism to bring critical care specialists to a larger number of patients. In spite of tremendous improvements in the technological basis of and investments in teleICU systems the goals of ‘‘alleviate[ing] scarcity and misdistribution of critical care services’’ remain unmet.1 Though tele-ICU services have grown substantially in the last decade, many of the questions regarding effectiveness, efficiency, and total costs of tele-ICUs systems are unanswered. This review will examine the current state of telemedicine in an age of rapidly expanding medical information technology and increasing demands for intensive care services. There are several recent reviews and reports on tele-ICUs that provide excellent and accurate information regarding the effectiveness of tele-ICU services.2–5 Our goals are more practical. In this manuscript we will consider some of the logistical aspects of tele-ICU that we have learned during an on-going evaluation process. Methods To identify information regarding tele-ICU, we conducted an informal literature review, site visits, and focus groups. To identify literature regarding tele-ICU, we used the following MeSH headings: Telemedicine, critical care, intensive care and MeSH subheadings: manpower, organization and administration, economics to perform PubMed searches of the National Library of Medicine. Additional resources were located through the University HealthCare Consortium (UHC), on-line search engines and corporate web sites. Between September 2006 and November 2007 we (AS, JF) conducted site visits to 4 organizations that have implemented tele-ICU services We conducted interviews with the 2 major vendors for tele-ICU services (iMDsoft, Nedham, MA.; VISICU Inc., Baltimore, MD.). During our evaluation a third tele-medicine vendor discontinued their product (Cerner Corp., Kansas City, MO.). Finally, we held focus groups with clinical, administrative, and technological leaders at Johns Hopkins Hospital who are considering implementing a tele-ICU service. Why consider tele-ICU? The concept of tele-ICUs has existed for over 25 years and some of the reasons to consider tele-ICU remain the same today.1 The driving force behind tele-ICUs is the promise that they improve patient care and provide an efficient mechanism to bring critical care specialists to a larger number of patients and for a longer period of time. This may occur through a number of mechanisms which include: 1) providing ICU physician staffing to ICUs that do not have intensivist staffing –the ‘‘enhanced staffing model’’, 2) extending the duration of ICU physician oversight – the ‘‘enhanced supervision model’’, 3) increasing compliance with evidence-based guidelines – the ‘‘enhanced compliance model’’ and 4) allowing earlier recognition of events in which early action can have significant impact – the ‘‘early warning model’’. These processes are not mutually exclusive and a single tele-ICU system may incorporate several but not necessarily all of them. ICU staffing There is no universally accepted optimum ICU physician staffing (IPS) model.6 While the general premise among intensivists is that more ICU physician staffing (ie. greater intensity) is associated with improved outcomes, a single study questioned this long-held belief.7 Nevertheless, the bulk of evidence and general perceptions support the model of an intensivist-led team providing daily care to all patients in the ICU. Pronovost and colleagues found that there was significant survival benefit for patients following repair of abdominal aortic aneurysms when an intensivist rounded daily in the ICU.8 A subsequent study of patients that underwent esophageal resection found that intensivist daily rounds in the ICU resulted in significant reductions in hospital length of stay and complications.9 More

A. Sapirstein et al. / Best Practice & Research Clinical Anaesthesiology 23 (2009) 115–126

117

compelling evidence came in the form of a large systematic review that compared ‘‘high intensity’’ IPS in which the ICU was closed or all patients had a mandatory intensivist consultation to ‘‘low intensity’’ IPS. In this study, Pronovost and colleagues found that high intensity staffing significantly reduced ICU mortality, hospital mortality, ICU and hospital length of stays.10 Thus the participation of an on-site intensivist appears to improve the care of the critically ill patient. Some studies also provide support for staffing ICUs 24 hours a day with intensivists though such models present substantial logistic hurdles.11 Why should IPS have such an impact on ICU outcomes? Unfortunately, we do not know. Though the relative importance of each component is unknown, Pronovost and colleagues believe that the benefits of intensivists derive from their; physical presence in the ICU, specialized knowledge, communication with patients, their families and other members of the care team, and their management of care at the unit level.12 In addition to providing care for individual patients, they help to manage the ICU by creating policies and procedures, improving quality of care and patient safety, and allocating ICU resources. In 1995 Leape and colleagues found that approximately two errors in care were committed daily to each patient in the ICU and that 20% of these were potentially serious or fatal.13 Subsequent work found that approximately 46% of ICU patients suffered an adverse event and that there were 19 preventable adverse events per 1000 patient days.14 Many of these errors are related to medications. We do not know yet if the presence of an intensivist-led team decreases the number of such preventable mistakes. One potential of a tele-ICU system, when integrated with the electronic medication administration record (eMAR), is that it may reduce preventable medication errors.15 Intensivists are also more likely to develop and implement new evidence-based guidelines and protocols to improve ICU care16,17 Standardized protocols for the management of critically ill patients can decrease the morbidity of sepsis18, mechanical ventilation19 and hyperglycemia.20 Intensivists may help ensure that patients reliably receive best practice guidelines but this has proved to be a difficult task.21 It is also possible that intensivists impact the quality of patient care through their early recognition and treatment of ICU specific diseases such as sepsis and ventilator associated pneumonia.22 The cost of ICU care is high but increased IPS may provide financial benefits to hospitals. Halpern et al estimated that in 2000 the US spent $55.5 billion on ICU care which represented 0.56% of the gross domestic product.23 Pronovost and colleagues analyzed the financial impact of IPS models and determined that high intensity IPS was likely to produce significant cost-savings for hospitals but in the worst case scenario could add approximately $1 M of costs per ICU.24 Another analysis by this same group suggested that $5 billion could be saved annually in the US by staffing every ICU at the high intensity level.25 In these models, hospital administration provided financial support for IPS. Recommendations for minimal IPS levels have existed for over a decade. In 1997, a taskforce of the European Society of Intensive Care Medicine made recommendations that a qualified intensivist provide 24-hour coverage in ICUs.26 The University HealthSystem Consortium (UHC) has stated that continuous intensivist coverage of all ICUs is the ultimate coverage goal.5 The ideal of around-the-clock intensivist coverage is seductive but the data supporting it is very limited and not compelling.11 Moreover, 24 hour in-house staffing as compared to daytime staffing (with home coverage in the evening) may not be cost effective.6 In the USA, the Leapfrog Group was formed by a consortium of major businesses to address issues of healthcare quality and cost. The Leapfrog group made three evidence-based recommendations to improve quality of patient care.27 Importantly they defined goals of high intensity IPS and computerized physician order entry systems (CPOE) in these recommendations.27 Adoption of the Leapfrog IPS standards has proved to be a challenge for hospitals largely because of limitations in the supply of fellowship-trained intensivists.28,29 A recent survey indicates an increase in the adherence to the IPS standards to 30% of ICUs in the USA.30 While this estimate is an increase from previous estimates it does not approach full adherence and a recent survey indicates that there are wide variations in the interpretation of the IPS standards by hospitals that have reported compliance.31 There are numerous reports describing the obstacles to full implementation of the Leapfrog IPS standards.29,32 There are currently approximately 10,000 intensivists in active practice in the USA.28 and ICUs with low IPS have little hope of hiring intensivists from the limited pool.33,34 Current ICUspecific manpower shortages are expected to worsen as the number of physicians in training programs

118

A. Sapirstein et al. / Best Practice & Research Clinical Anaesthesiology 23 (2009) 115–126

and critical care trained nurses is on the decline.28,35 The demand for ICU care is also expected to increase because of the dramatic aging of the baby boom generation. Between 2000 and 2030 the U.S. population older than 65 is expected to double and will account for about 20% of the total population.36 Since age is highly correlated to critical illness and use of critical care services, it is projected that this will lead to a significant increased demand for ICU services. In an assessment published in 2000 Angus and colleagues projected an additional 35% shortfall in IPS by 2035 and more current analyses have predicted worse.28,37 Given the current shortage of intensivists and the increasing demand for critical care services, efficient and effective methods to expand IPS to a larger number of patients are needed. Telemedicine has been proposed as a means to accomplish this.3 Tele-ICU core components Tele-ICU systems are composed of technology (hardware and software) and personnel that collect analyze and transmit information back and forth between the physical ICU and the tele-ICU command center. These systems can track, analyze and document patient data and allow ICU and tele-ICU clinicians to intervene in patient care. Analyses and descriptions of tele-ICU systems have largely focused on the technology but consideration of the personnel components may be even more important. We will briefly describe both. There are 3 core technology components of all tele-ICU systems. Firstly, the system relays all available patient data to the tele-ICU center. This includes nearly real-time physiological data, laboratory results, radiographic results, electronic patient records and the medication record. Secondly, a user interface software application organizes the patient data in the tele-ICU center, in an indexed manner so that the data can be easily retrieved and viewed. Finally, a communication network is established between the physical ICUs and the tele-ICU center.3,38 While these functions appear simple, making them work in the patient care environment is extraordinarily demanding. Most institutions use a large number of different hardware and software products in their ICUs. Often the same hospital will have several different products serving the same function across the hospital. System coherence and integration has often suffered in favor of a lower purchase cost. As a result tele-ICU systems may be required to collect information across a number of different and changing platforms. In addition the tele-ICU system must be secure, comply with the Health Information Portability Act (HIPA), and all access must be restricted and password protected.39 Before implementing a tele-ICU system, a clear set of institutional objectives should be formulated; changing the mission of the tele-ICU after implementation can be a formidable task. In almost all of the major tele-ICU systems the instillation of the remote ICU system has been accompanied by a major upgrade or a new implementation of electronic records in the physical ICUs. Often this has been accompanied or is coincident with the implementation of a CPOE system. The user interface application of a tele-ICU system may also incorporate advanced features including trends analysis and decision support tools. The VISICU (VISICU Inc., Baltimore, MD, http:// www.visicu.com/index_noflash.asp) system features an acuity ranking system in its user interface that ranks patients within pre-defined levels of illness. In addition the VISICU interface analyzes patient data to identify deteriorating patients early, these are called ‘‘smart alerts’’Ò, and provides an on-line library of best practices to clinicians in the tele-ICU. Both the manufacture and many users of the VISICU system believe that such diagnostic tools improve patient care. While there is some evidence that automated electronic alerts to physicians can identify illness, there is little evidence indicating they improve patient care.40,41 In our discussions with tele-ICU practioners false positive alerts remain problematic in the virtual ICU as they are in physical ICUs. The favored model to communicate between the tele-ICU and physical ICU is a bidirectional audiovisual link. The most highly developed tele-ICU systems have dedicated video cameras and monitors in every patient room. This permits the tele-ICU clinicians to see every patient and the patients’ family members and ICU staff to see the tele-ICU staff during periods of consultation. Because of the growth of tele-ICU services, the Leapfrog Group has defined specifications that allow a tele-ICU system to meet the IPS standard. The Leapfrog Group has specified that the video system must be sufficient to assess the breathing pattern of a patient.5 The commonly used systems far surpass

A. Sapirstein et al. / Best Practice & Research Clinical Anaesthesiology 23 (2009) 115–126

119

this quality and can be used to provide procedural guidance to inexperienced operators. ICU clinicians and tele-intensivists anecdotally report that two-way video systems improve acceptance and functioning of the tele-ICU (personal communications Matthew McCambridge). In the large scale applications of tele-ICUs the remote facility is placed in an off-site location. This has the dual advantage of not physically linking the tele-ICU with a single hospital and not occupying valuable hospital space. Each tele-ICU practioner occupies a work station that is equipped with a desk, a number of screens, a telephone, a direct audio and video link and dedicated tele-ICU computers. The array of computer screens typically display: the ICU census page, the current patient’s ‘‘dashboard’’, patient physiologic data, patient labs, studies or eMAR and one screen dedicated for the visual link to the patient. Although practioners seem to agree that the ability to see patients is essential it is generally required intermittently. Tele-ICU work flow The current operational models of tele-ICUs have changed little since the initial descriptions.38 In many ways the tele-ICU reconstitutes the multidisciplinary ICU in a remote location. Thus a typical teleICU is staffed by a unit clerk, non-physician providers, and an intensive care physician. In addition the tele-ICU typically has a physician medical director, a nursing director and director/manager of information technology. In the same way that ward clerks are essential in managing the flow of administrative information in an ICU, the tele-ICU clerk assures the flow of information between the tele-ICU and the ICUs that they oversee. The roles of the non-physician providers in the tele-ICU are variable but the core activity is relatively constant. The non-physician providers are usually experienced ICU nurses. In the daytime hours large tele-ICU systems staff the command center with a nurse who is available for consultation, organizes the patient census and may document compliance with protocols or evidence-based interventions. Most tele-ICUs perform the patient care functions primarily overnight. Typically one nurse can perform rounds on up to approximately 50 patients. These rounds consist of scrolling through each patient in the census and examining the current information. Thus a remote-ICU nurse will examine the current physiologic data, new lab data and medication lists of each of their patients on an almost continuous basis. The tele-ICU nurse may also be alerted to acute changes in a patient’s condition either by software alerts or directly by the staff in the ICU. Tele-ICUs report that many nurses find the intellectual and oversight processes of work in a tele-ICU a welcome change to the burden of overnight call in an ICU (personal communication and 3). In order to manage upwards of 100 ICU beds the typical tele-ICU staffs with at least 2 nurses overnight. In addition to the nocturnal nursing staffing, most tele-ICUs staff with a single nurse during some part of the day. The patient care role of this nurse is variable between tele-ICU centers. Typically the daytime level of service is less pro-active and more consultative. Thus the nurse may respond directly to questions from the bedside nurses and may also check compliance with best practice protocols on all ICU patients. In addition the daytime nurse can collect data that is essential for benchmarking the functioning of the tele-ICU and the physical ICUs. Physician staffing of tele-ICUs has been more variable than that of nurses. Because of the shortage of ICU physicians, tele-ICUs have been staffed with ICU physicians from a number of disciplines. In most systems the physician staff of the tele-ICU is drawn from the core group of the hospital intensivists that implemented the tele-ICU system. This group may be supplemented by moonlighters drawn from a pool of clinical fellows or other practitioners who are deemed to have adequate ICU experience or training. In our discussions with hospitals implementing tele-ICU services, it is clear that the interaction between the tele-ICU physicians and the covered units is best when that physician is in the normal practice rotation of the ICU. In addition tele-ICUs have employed several physician service models. In an early commercial tele-ICU model the vendor contracted independently with nurses and physicians to staff the tele-ICU. In some cases the physician is employed as a per diem worker while in others the tele-ICU coverage is by salaried physicians. As described in the Leapfrog telemonitoring standards the tele-intensivist is responsible for obtaining a sign-out for each covered patient.5 There is no standard for how this sign out is accomplished. In some tele-ICUs the tele-intensivist gets a verbal sign out from the intensivist in the ICU

120

A. Sapirstein et al. / Best Practice & Research Clinical Anaesthesiology 23 (2009) 115–126

either by phone or through a voicemail recording. One can imagine that communicating information for over 100 patients requires a significant time even when the information is cursory. When the same intensivist physician staffs the tele-ICU over a number of days in continuity, the sign-out becomes sequentially more efficient. It is also possible that written notes can be used for sign-out but this is a less well-developed system as it is burdensome and is not as current or interactive as a verbal signout. Much like other types of signouts, tele-ICU signouts need to answer three key questions: What do I need to know? , What do I need to do? , and What am I worried about?. To improve efficiency, the answers to these questions can be very general (the 30,000 foot overview level) as long as the clinicians can quickly get the information that will enable them to dive down to the ground level when needed. After sign-out the intensivist begins his/her rounds. The rounding process of the intensivist is usually guided by knowledge of the clinical status of the patients. This clinical knowledge is gained through a number of mechanisms: 1) a non-physician provider rounds on the patients before the start of the intensivist’s shift. They will then assist the physician in prioritizing the patient list, 2) the teleICU software program may prioritize the patients based on defined criteria for severity of illness 3) the tele-intensivist’s own familiarity with the patient on the census. Although each ICU system must have explicit policies regarding the roles and responsibilities of the tele-ICU, the role of the tele-intensivist is usually context dependent. In ICUs where there is an on-site intensivist the tele-intensivist defers to the pre-established care plan. Deviations from this plan may be directly communicated to the on-site intensivist or their team. In locations that lack an on-site team, the tele-intensivist may take a primary role in establishing a patient care plan. While it has been recommended that tele-ICUs follow a written process of communication and documentation, the rigor with which they do this varies across tele-ICUs. Tele-ICU systems generally track interventions that they perform. These interventions can be categorized as minor, such as potassium repletion, or major, such as requesting an intubation. Practitioners often feel that major interventions add value to the health care system and prevent catastrophic events such as cardiopulmonary arrests (personal communication and42). Proof of concept The first prospective study of a model tele-ICU compared a 16 week period of tele-intensivist coverage in a 10-bed surgical and trauma ICU to two periods of baseline data in which patients were not cared for by intensivist.43 Remote intensivists working from their homes had access to real-time patient data and video-conferencing to visualize patients and communicate with on-site caregivers. When compared to the baseline periods, the tele-ICU management period significantly reduced the Acute Physiology and Chronic Health Evaluation (APACHE) III severity-adjusted ICU and hospital mortality (60% and 30%, respectively).43 ICU complications decreased by 40% in the tele-ICU management period and this resulted in a significant decrease in the number of ‘‘outlier patients’’ (ICU length of stay .6 days).43 Following this proof of principle study the VISICU Company was formed as a commercial tele-ICU solution and was adopted by the Sentara healthcare system. The experience of this tele-ICU system was evaluated by comparing ICU benchmarks in the year preceding tele-ICU implementation to the same benchmarks in the 6 months following the implementation.38 This study of a small number of patients had several methodological shortcomings but found a 25% reduction in overall mortality (averaged for ICU and hospital mortality) as well as improvements in other outcomes.38 The major improvements in outcomes were seen in the patients in the medical but not the surgical ICU and the authors speculated that this was due to greater adoption of the tele-ICU in the medical unit.38 A recent report stated that by 2007 there were 30 tele-ICU systems in the USA covering approximately 300 adult ICU locations.2 Data from these systems is not yet in peer-reviewed publications but has been made available in reports and abstracts. The Memorial Herman Health System and the Lehigh Valley Health System implemented different tele-ICU systems and observed similar reductions in ICU mortality rates.2 In contrast Sutter Health System failed to observe an impact of tele-ICU on the predicted overall ICU mortality and attributed the lack of effect to the pre-existing high level of IPS and low mortality in their ICU system.2 An abstract from the University of Pennsylvania found profound improvements in ICU and hospital mortality and length of stays after implementation of a tele-ICU.44

A. Sapirstein et al. / Best Practice & Research Clinical Anaesthesiology 23 (2009) 115–126

121

Press releases from VISICU state that implementation of their tele-ICU products has achieved a 29% overall reduction in hospital mortality across all the ICU systems in which it is used as compared to national averages.45 In spite of their proliferation we do not know how tele-ICUs improve patient outcomes. Moreover, most of the tele-ICU programs have added tele-ICU interventions to hospitals with highly organized ICU physician staffing. It has been suggested that direct interventions in patient care by the tele-ICU improves patient outcomes though this has yet to be validated. Alternatively it has been suggested that the tele-ICU system can be used to improve compliance with evidence-based best care practices that are known to improve outcomes. Adherence to such measures has proved very difficult to achieve and clearly requires a receptive and supportive ICU culture.21 The Sutter Health Hospital system has used the tele-ICU to improve compliance with best practices for sepsis management and DVT prophylaxis. They reported that they were able to significantly lower mortality in one of their ICUs by utilizing the tele-ICU to establish timely institution of bundled sepsis guidelines.46 They have also used the system to achieve nearly 100% compliance with DVT prophylaxis in one of their ICUs.47 Some of these gains may be achieved through culture and documentation change since achieving target compliance in the ICU can be accomplished using an electronic flow sheet.48 Cost – benefit While the clinical benefits of tele-ICU can be questioned, the cost of these systems is significant. Publications and reports suggest that investments by hospital systems in tele-ICUs pay financial benefits. Breslow recently estimated that the initial cost of a tele-ICU based on the VISICU eICU model is greater than $2 million for a typically sized system though the exact size was not defined.3 This amount includes the $ 30- 50,000 cost in equipment and licensing charges for each covered ICU bed and the establishment of an off-site control center.4 Personnel expenses are ongoing and are the largest budgetary component of a tele-ICU system. A recent estimate placed staffing the tele-ICU control center costs between $ 1 – 2 million dollars annually depending on the size of the tele-ICU system.4 Unlike other new technologies, the instillation cost per bed of tele-ICU systems has not changed much in the last 10 years but we assume that manpower costs will continue to rise with scarcity of supply. Many tele-ICUs pay a premium to physicians to staff overnight shifts. The Lehigh Valley Health System tele-ICU has determined that it requires between 5-7 full time intensivists to adequately staff each teleintensivist position (personal communication, Matthew McCambridge MD); this is roughly the same number of intensivists that it takes to staff a physical ICU.49 Because tele-ICUs are not able to charge for physician services that they provide their costs can only be recouped through improved efficiency. In the Sentara study Breslow and colleagues reasoned that reductions in the patients’ average length of stay (LOS) translated to a 24.6% decrease in cost per case and this resulted in a $3.1 million benefit to the hospital over the six month study period.38 The press reported that an analysis by an independent accounting firm of the same tele-ICU over 18 months showed a net revenue benefit of $3 million per year.50 A financial analysis of the tele-ICU at the University of Pennsylvania is similar with a decrease in ICU and hospital LOS leading to an estimated reduction in costs of up to w$3.8 million dollars per year.51 The actual financial effects of a tele-ICU are difficult to measure because most hospitals’ accounting systems are designed to measure billing and reimbursement rather than actual costs. In addition, the methods of allocating overhead costs are inexact. In spite of the limitations of the financial analyses that have been conducted a large number of health care systems have and continue to make very large investments in tele-ICUs. The manpower dilemma One of the promises of tele-ICU is that it can maximize the efficiency of ICU physician resources. Put more simply this means that a single intensivist can cover a greater number of beds. In some teleICUs this number may exceed 120. Clearly the role of the tele-intensivist is different than that of an on-site intensivist who, on average, spends 9 hours a day providing direct care on the ward of a 15 bed ICU.52 In most tele-ICUs it appears that the IPS of the core hospitals already meets Leapfrog standards and remains at the same high level following the institution of the tele-ICU. IPS efficiency is

122

A. Sapirstein et al. / Best Practice & Research Clinical Anaesthesiology 23 (2009) 115–126

added to the system by the inclusion of ICU beds that were not previously in the IPS model of the health care system or in extending IPS to 24 hours. In this deployment model the number of intensivists staffing all ICU activities (both physical and tele-ICU) necessarily increases. The result is that intensivists who could provide on-site care will now also practice tele-ICU. Thus while the number of intensivists in a program that includes a tele-ICU increases, the number of intensivists per covered ICU bed generally decreases. We do not know if the tele-ICU practice model will repel or entice more trainees toward critical care or whether the practice of tele-ICU requires skills and competencies not obtained in current training programs. Fink and Suter have suggested that both regionalization and telemedicine will be necessary to improve the ICU physician shortages.53 A major tele-ICU system will become a de facto regional center for ICU care by virtue of the facility’s outreach and its concentration of intensivist manpower. Organizational change In order for a tele-ICU system to succeed, many stakeholders (clinicians, hospital leaders, payers, regulators) in the health care system must support it. Our ability to systematically evaluate the performance of the tele-ICU is challenged by the tremendous organizational and cultural changes that appear to be transformative to an ICU system when a tele-ICU is implemented.54 Tele-ICU proponents point to examples where its performance is degraded by resistance from bedside practioners.38,55 In this review we have indicated that there is a dearth of published data on tele-ICU results. This is not surprising given the difficulty in analyzing these complex systems and the relatively short time period these systems have been used. The institution of a tele-ICU brings a variety of resources and changes in the structure and process to the ICU and any of these could contribute to improvement. The executive level of hospital leadership has been instrumental in initiating and supporting the large tele-ICU systems. Without this level of support it is not likely that a critical care system could successfully implement the organizational changes required by a tele-ICU. There is a perception that the major change of a tele-ICU system is to enhance ICU physician resources. In contrast there has been little discussion about the role of the ICU nurses in the teleICUs. Based on interviews, bedside nurses at several ICUs reported that they perceived the value of the tele-ICU to be the oversight and consultations of the experienced ICU nurse. The relative value of the tele-nurses compared to the intensivist has not been evaluated. It may be that doctors and nurses benefit from both tele-physicians and nurses. Similarly most tele-ICUs have added or improved the electronic records and software tools of the physical as well as the virtual ICU. Again we do not know the relative importance of the process changes. As has been recently recognized changes in the organizational approach to diseases such as sepsis can have profound impact.56 At the University of Massachusetts the institutional ICU management structure was reorganized before the implementation of the tele-ICU.54 The goals of this change were to create a collaborative, multidisciplinary, patient-focused environment in which the tele-ICU was only one component.54 Many of these changes could be made without installing a tele-ICU but the tele-ICU seems to be a catalyst and framework for such change. Future directions It is often asked if tele-ICU is inevitable given the factors pushing our need for expanded ICU coverage and the shortage of ICU physicians, nurses, and other ICU staff. Information technology is expanding rapidly throughout the healthcare system and the ICU is no exception. ICUs now use electronic systems for viewing radiographic studies, documentation and most importantly order entry. The software systems that provide these services are currently free-standing but there is a clear demand for better systems integration. To this end software vendors are developing software systems which integrate documentation, order entry and results exposition (for example Eclypsis Corp., Atlanta, GA.) This push toward systems integration and universal access is leading toward convergence of the existing tele-ICU solutions with the other software applications. Thus we anticipate that with software convergence every ICU will have some degree of telemedicine capability. Ultimately the human and financial factors of staffing a tele-ICU are more likely to determine how tele-ICU

A. Sapirstein et al. / Best Practice & Research Clinical Anaesthesiology 23 (2009) 115–126

123

develops. Many ICU decisions are data driven and this data can now be delivered remotely. The model of radiology serves as an example of how interpretation of and decisions based on this data can be outsourced at considerable financial savings.57 The outsourcing model of radiology may also be helpful in creating a global network of intensivists physicians. As with radiology services the training and skill levels of such providers would require standardization, certification and ongoing evaluation.57 As we continue to converge toward telemedicine in the ICU it will be important to develop work environments that not only improve patient care but also the satisfaction of all members of the multi-disciplinary care team. The field of computer-aided medical decision support is in its infancy. ICU patients generate vast amounts of clinical data which can be captured in tele-ICU systems. Clearly the impact of tele-ICU could be amplified with further improvements in trend analysis, decision support and notification alarms.40,41 This is an area in which collaboration between universities, industry, and schools of medicine is essential. Effective algorithms to harness the power of archival and real-time patient data have the potential to significantly improve critical care. The automated acquisition of large amounts of patient data also creates unique research and educational opportunities for tele-ICU systems. Groups such as the Michigan Health and Hospital Association (MHA) Keystone Center for Patient Safety and Quality have demonstrated the power of creating networks for change in the ICU.16 VISICU has organized users of their eICU product into a collaborative in which performance data is compiled and benchmarks are established. In the future such networks could be instrumental in efficiently testing hypotheses and therapies. The loop will be completed when interventions based on such studies can be efficiently disseminated through the use of the tele-ICU. Conclusions Consideration of tele-ICU presents us with a series of paradoxes. While tele-ICU promises to expand ICU services to under-served and under-resourced ICUs it will draw the scarce resource of nurses and physicians from these ICUs to regional centers large enough to sustain the tele-ICU.53 While users, vendors and administrators report that tele-ICU significantly improves patient care there is very little published data to substantiate these claims. While tele-ICUs subsume millions of dollars in operational costs and have no ability to generate revenue they are reported to save millions in health care expenses. As tele-ICUs and information technology continue to mature we may see a resolution of some of these paradoxes. In the end our search for sustainable models of quality ICU care may lead us to widespread implementation of tele-ICU. It is our hope that this will be done by considered design with collaboration between intensive care providers, hospital administrators, health services researchers, and medical IT experts. Because information technology is an area where the future seems to repeatedly surpass our expectations we believe that the future of tele-ICU is promising.

Practice points  Define and reach agreement on the goals and limitations of tele-ICU within the health care system and within each physical unit before initiation  Roles and responsibilities of the tele-ICU staff relative to unit-based staff must be delineated  Plan tele-ICU with all stakeholders and implement in a staged fashion  Tele-ICUs that are staffed with unit-based practitioners are most easily accepted and integrated  Two-way telecommunication is superior to one-way observational video systems  Dedicated administrative staff to organize and maintain tele-ICU operations is essential  Track and maintain tele-ICU database with hope of continuous improvement in quality and safety

124

A. Sapirstein et al. / Best Practice & Research Clinical Anaesthesiology 23 (2009) 115–126

Research highlights and agenda  Studies of tele-ICU in single health care systems indicate improvement in patient outcomes and cost savings in excess of expenses  Centralized patient review improves compliance with best practices in tele-ICU systems where this is a goal  Study of patient outcomes in ICUs that are served by tele-ICU should be compared contemporaneously to a matched set of ICUs that are not supported by tele-ICU  A realistic manpower assessment for tele-ICU systems should be made so that the ICU community can determine how to use scarce human resources  The relative importance of alarms, decision support and provider level in the tele-ICU should be examined

Acknowledgements The authors gratefully acknowledge the assistance of Drs. C. William Hanson, Craig Lilly and Matthew McCambridge and the staff and physicians at the ICU telemedicine centers of the University of Pennsylvania Health System, The University of Massachusetts Health System and the Lehigh Valley Health System. We also acknowledge the helpful input of Peter Greene, Ronald Pauldine, Steve Mandell, Judy Schroeder, and John Ulatowski during this process. This work was supported by the Department of Anesthesia and Critical Care Medicine of Johns Hopkins University.

References 1. Grundy BL, Jones PK & Lovitt A. Telemedicine in critical care: problems in design, implementation, and assessment. Critical Care Medicine 1982; 10(7): 471–475. *2. Tele-ICUs:remote mangement in intensive care units. Cambridge, MA: New Engalnd Healthcare Institute Massachusetts Technology Collaborative Health Technology Center, 2007. 3. Breslow MJ. Remote ICU care programs: current status. Journal of Critical Care 2007; 22(1): 66–76. 4. Cummings J, Krsek C, Vermoch K et al. Intensive care unit telemedicine: review and consensus recommendations. American Journal of Medical Quality 2007; 22(4): 239–250. *5. Cummings JP. UHC technology report: intensive care unit telemedicine. In: Technology report. Oak Brook, IL: University HealthSystem Consortium, 2006. p. 24. 6. Sapirstein A, Needham DM & Pronovost PJ. 24-hour intensivist staffing: balancing benefits and costs. Critical Care Medicine 2008; 36(1): 367–368. 7. Levy MM, Rapoport J, Lemeshow S et al. Association between critical care physician management and patient mortality in the intensive care unit. Annals of Internal Medicine 2008; 148(11): 801–809. *8. Pronovost PJ, Jenckes MW, Dorman T et al. Organizational characteristics of intensive care units related to outcomes of abdominal aortic surgery. JAMA: The Journal of the American Medical Association 1999; 281(14): 1310–1317. 9. Dimick JB, Pronovost PJ, Heitmiller RF et al. Intensive care unit physician staffing is associated with decreased length of stay, hospital cost, and complications after esophageal resection. Critical Care Medicine 2001; 29(4): 753–758. *10. Pronovost PJ, Angus DC, Dorman T et al. Physician staffing patterns and clinical outcomes in critically ill patients: a systematic review. JAMA: The Journal of the American Medical Association 2002; 288(17): 2151–2162. 11. Gajic O, Afessa B, Hanson AC et al. Effect of 24-hour mandatory versus on-demand critical care specialist presence on quality of care and family and provider satisfaction in the intensive care unit of a teaching hospital. Critical Care Medicine 2008; 36(1): 36–44. 12. Pronovost PJ, Holzmueller CG, Clattenburg L et al. Team care: beyond open and closed intensive care units. Current Opinion in Critical Care 2006; 12(6): 604–608. 13. Leape LL, Bates DW, Cullen DJ et al. Systems analysis of adverse drug events. ADE Prevention Study Group. JAMA: The Journal of the American Medical Association 1995; 274(1): 35–43. 14. Bates DW, Cullen DJ, Laird N et al. Incidence of adverse drug events and potential adverse drug events. Implications for prevention. ADE Prevention Study Group. JAMA: The Journal of the American Medical Association 1995; 274(1): 29–34. 15. Shulman R, Singer M, Goldstone J et al. Medication errors: a prospective cohort study of hand-written and computerised physician order entry in the intensive care unit. Critical Care 2005; 9(5): R516–R521. 16. Pronovost P, Needham D, Berenholtz S et al. An intervention to decrease catheter-related bloodstream infections in the ICU. The New England Journal of Medicine 2006; 355(26): 2725–2732. 17. Kahn JM, Brake H & Steinberg KP. Intensivist physician staffing and the process of care in academic medical centres. Quality & Safety in Health Care 2007; 16(5): 329–333.

A. Sapirstein et al. / Best Practice & Research Clinical Anaesthesiology 23 (2009) 115–126

125

18. Gao F, Melody T, Daniels DF et al. The impact of compliance with 6-hour and 24-hour sepsis bundles on hospital mortality in patients with severe sepsis: a prospective observational study. Critical Care 2005; 9(6): R764–R770. 19. Resar R, Pronovost P, Haraden C et al. Using a bundle approach to improve ventilator care processes and reduce ventilatorassociated pneumonia. Joint Commission Journal on Quality and Patient Safety 2005; 31(5): 243–248. 20. van den Berghe G, Wouters P, Weekers F et al. Intensive insulin therapy in the critically ill patients. The New England Journal of Medicine 2001; 345(19): 1359–1367. 21. Ferrer R, Artigas A, Levy MM et al. Improvement in process of care and outcome after a multicenter severe sepsis educational program in Spain. JAMA: The Journal of the American Medical Association 2008; 299(19): 2294–2303. 22. Sinuff T, Muscedere J, Cook D et al. Ventilator-associated pneumonia: improving outcomes through guideline implementation. Journal of Critical Care 2008; 23(1): 118–125. 23. Halpern NA, Pastores SM & Greenstein RJ. Critical care medicine in the United States 1985-2000: an analysis of bed numbers, use, and costs. Critical Care Medicine 2004; 32(6): 1254–1259. 24. Pronovost PJ, Needham DM, Waters H et al. Intensive care unit physician staffing: financial modeling of the Leapfrog standard. Critical Care Medicine 2006; 34(3 Suppl): S18–S24. 25. Pronovost PJ, Waters H & Dorman T. Impact of critical care physician workforce for intensive care unit physician staffing. Current Opinion in Critical Care 2001; 7(6): 456–459. 26. Ferdinande P. Recommendations on minimal requirements for intensive care departments. Members of the task force of the European Society of Intensive Care Medicine. Intensive Care Medicine 1997; 23(2): 226–232. 27. The leapfrog group fact sheet. Available from: http://www.leapfroggroup.org/media/file/The_Leapfrog_Group_Fact_Sheet_ 03_2008.pdf; 2008. *28. Angus DC, Kelley MA, Schmitz RJ et al. Caring for the critically ill patient. Current and projected workforce requirements for care of the critically ill and patients with pulmonary disease: can we meet the requirements of an aging population? JAMA: The Journal of the American Medical Association 2000; 284(21): 2762–2770. 29. Angus DC, Shorr AF, White A et al. Critical care delivery in the United States: distribution of services and compliance with Leapfrog recommendations. Critical Care Medicine 2006; 34(4): 1016–1024. *30. The_Leapfrog_Group. Factsheet: ICU physician staffing. Available from: http://www.leapfroggroup.org/media/file/Fact_ Sheet_IPS_080327.pdf; 2008. *31. Pronovost P, Thompson DA, Holzmueller CG et al. Impact of the leapfrog group’s intensive care unit physician staffing standard. Journal of Critical Care 2007; 22(2): 89–96. 32. Kahn JM, Matthews FA, Angus DC et al. Barriers to implementing the leapfrog group recommendations for intensivist physician staffing: a survey of intensive care unit directors. Journal of Critical Care 2007; 22(2): 97–103. 33. Kelley MA, Angus D, Chalfin DB et al. The critical care crisis in the United States: a report from the profession. Chest 2004; 125(4): 1514–1517. 34. Brilli RJ, Spevetz A, Branson RD et al. Critical care delivery in the intensive care unit: defining clinical roles and the best practice model. Critical Care Medicine 2001; 29(10): 2007–2019. 35. Chalfin DB. Implementation of standards for intensivist staffing: is it time to jump aboard the leapfrog bandwagon? Critical Care Medicine 2004; 32(6): 1406–1408. 36. Humes K. The population 65 years and older: aging in America, in the book of the states 2005. Washington, DC: The Council of State Governments, 2005: 464–468. 37. Krell K. Critical care workforce. Critical Care Medicine 2008; 36(4): 1350–1353. *38. Breslow MJ, Rosenfeld BA, Doerfler M et al. Effect of a multiple-site intensive care unit telemedicine program on clinical and economic outcomes: an alternative paradigm for intensivist staffing. Critical Care Medicine 2004; 32(1): 31–38. 39. Nelson SB. Privacy and medical information on the internet. Respiratory Care 2006; 51(2): 183–187. 40. Major K, Shabot MM & Cunneen S. Wireless clinical alerts and patient outcomes in the surgical intensive care unit. The American Surgeon 2002; 68(12): 1057–1060. 41. Schoenberg R, Sands DZ & Safran C. Making ICU alarms meaningful: a comparison of traditional vs. trend-based algorithms. Proceedings/AMIA . Annual Symposium. AMIA Symposium 1999: 379–383. 42. Shaffer JP, Breslow MJ, Johnson et al. Remote ICU management improves outcomes in patients with cardiopulmonary arrest. Critical Care Medicine 2005; 33: A5. *43. Rosenfeld BA, Dorman T, Breslow MJ et al. Intensive care unit telemedicine: alternate paradigm for providing continuous intensivist care. Critical Care Medicine 2000; 28(12): 3925–3931. 44. Kohl BA, Sites FD, Gutsche JT et al. Effect of telemedicine on mortality and length of stay in a university ICU. Critical Care Medicine 2007; 35(12 Suppl): A22. 45. VISICU, Data from 185,000 icu admissions show significant reductions in mortality. 2008, PRNewswire. 46. Ikeda D, Hayatdavoudi S, Winchell J et al. The impact of using a standard protocol for the surviving sepsis 6 and 24 hr bundles in septic patients on total icu risk adjusted mortality. Critical Care Medicine 2006; 34(12 Suppl): A108. 47. Shaughnessy T. Telemonitoring helps save lives, improve care, in: CSA bulletin. 2008. 48. Wahl WL, Talsma A, Dawson C et al. Use of computerized ICU documentation to capture ICU core measures. Surgery 2006; 140(4): 684–689 [discussion 690]. 49. Pronovost PJ, Talsma A, Dawson C et al. The organization of intensive care unit physician services. Critical Care Medicine 2007; 35(10): 2256–2261. 50. Becker C. Remote control. Specialists are running intensive-care units from remote sites via computers, and at least one health system with the e-ICU is reaping financial rewards–and saving lives. Modern Healthcare 2002; 32(8): 40–42, 44, 46. 51. Kohl BA, Sites FD, Gutsche JT et al. Economic impact of EICU implementation in an academic surgical ICU. Critical Care Medicine 2007; 35(12 Suppl): A26. 52. Treggiari MM, Martin DP, Yanez ND et al. Effect of intensive care unit organizational model and structure on outcomes in patients with acute lung injury. American Journal of Respiratory and Critical Care Medicine 2007; 176(7): 685–690. 53. Fink MP & Suter PM. The future of our specialty: critical care medicine a decade from now. Critical Care Medicine 2006; 34(6): 1811–1816.

126

A. Sapirstein et al. / Best Practice & Research Clinical Anaesthesiology 23 (2009) 115–126

54. McCauley K & Irwin RS. Changing the work environment in ICUs to achieve patient-focused care: the time has come. Chest 2006; 130(5): 1571–1578. 55. Cowboy EN, Rajamani S, Jamil MG et al. Impact of remote icu management on ventilator days. Critical Care Medicine 2005; 33: A1. 56. Kahn JM & Bates DW. Improving sepsis care: the road ahead. JAMA: The Journal of the American Medical Association 2008; 299(19): 2322–2323. *57. Singh SN & Wachter RM. Perspectives on medical outsourcing and telemedicine–rough edges in a flat world? The New England Journal of Medicine 2008; 358(15): 1622–1627.