Introduction
Timely assessment and treatment of intensive care unit (ICU) patients is a major goal in intensive care. This requirement can be difficult given the shortage of available intensive care physicians and surgeons and ICU beds. Delays in both patient assessment and physician response can result in lost opportunities to improve patient outcome and can result in increased morbidity and length of stay. These problems particularly affect neurocritical care patients, because the dichotomy between the need for rapid intervention and the shortage of highly trained experts is most acute. Most neurocritical care patients are at high risk for further secondary brain injury such as brain ischemia or elevated intracranial pressure (ICP), and the time window for intervention is quite short. Hence there is a great need to quickly identify and rapidly respond to secondary brain insults once they occur. However, there is a nationwide shortage of neurointensivists. The use of information technology may hold the answer to facilitate this goal of quick diagnosis and rapid response despite the workforce shortage.
This chapter outlines the role of information technology to change the paradigm for patient care using robotic telepresence (RTP). Robots have been used increasingly in a variety of surgical procedures, in large part to augment surgical visualization, dexterity, and precision, although the effect on patient outcome and cost efficacy rather than on surrogate endpoints is still debated. In recent years there also has been a rapid expansion of ICU telemedicine that combines audiovisual technology with the electronic medical record to provide critical care from a remote location. This includes continuous offsite monitoring for the entire ICU or all ICUs in a hospital, or as is the focus of this chapter, targeted individual patient evaluation. This chapter outlines how RTP works, how it is used, how it impacts patient care, and how developers anticipate it will be used in the future in the neurointensive care unit. There are essentially two components to the RTP system: the robot and the computerized information system. These two elements have synergy and facilitate high-quality, timely clinical care. RTP may represent the next clinical care paradigm in the NCCU. This approach, perhaps through implementation of evidence-based practice in the ICU applied by an intensivist who can conduct remote rounds has the potential to improve patient outcomes.
Physician Response Paradigms
Rapid response and implementation of goal-directed treatment improve outcomes for critically ill patients with sepsis. Physician attention to patients is divided into prospective and responsive attention. In most neurointensive care units (NICU), prospective physician attention is conducted on morning bedside rounds. Morning bedside rounds are made to evaluate the patient’s clinical condition (the physical examination), and to evaluate the laboratory studies, brain images, and other data that accumulate over time. Morning rounds in the NICU are a mixture of retrospective analysis of trends in data and prospective planning of treatment for the day. These rounds are usually based on scheduled periods of time and result in a well-formulated “plan of attack” for the day, which likely will change several times during the course of the day based on intercurrent changes in patient condition. For a typical brain injury patient with ICP monitoring, it has been calculated that the intensivist evaluates more than 650 data points during each episode of bedside rounds. These data points consist of hourly vital signs, waveform morphology values, brain monitoring values such as brain tissue oxygen and microdialysis, radiologic studies, clinical examination findings, laboratory data, verbal story-line information, and visual-cue information from body language and appearance. These data points involve integration of graphic information, longitudinal trends of data, and cross-sectional review of data at fixed points in time. This analysis is done prospectively during morning rounds to make a reasoned clinical treatment plan.
Physicians also respond to changes in the clinical condition of the patient by being paged by the nurse or other health care professionals. This can be called an urgent or emergent response , and occurs throughout the day and night. During such an urgent response, the physician often needs to consider the same set of data points and to determine the evolution of the patient over time. However, the physician may not be at the bedside when paged and may not have all these data points available to him or her; yet he or she still is required to make a clinical decision. The response is by telephone and, at best, a delayed face-to-face evaluation of the patient, often hours later. This means at the time of the initial call, the physician may not have access to visual, graphic, or waveform information, and perhaps the electronic medical records or electronic charting data. Thus the physician is at a disadvantage in processing the clinical information and reaching a clinical decision during the time of urgent patient need. This disadvantage may result in poor clinical judgment or medical errors. Medical errors can occur when incomplete information is presented and/or verbal information alone is used. One way to avoid medical errors is to increase intensivist coverage in the ICU. Because there is a shortage of intensivists, it is difficult to increase the number of intensivists, per se. However, using RTP technology, one may be able to increase the percentage of time in which the intensivist is telepresent in the ICU. As discussed later in this chapter, robotic telepresence can help reduce or avoid medical errors.
Robotic Telepresence Technology
InTouch Technologies Inc. (InTouch Health), a privately held company based in Santa Barbara, CA, manufactures the robotic telecommunication system currently in use in the author’s ICU. The company has pioneered Remote Presence technology for hospital providers. Through its RP-7 Remote Presence Robotic System, a proprietary mobile robotic and communications platform, health care professionals are able to consult with hospital-based patients and staff more easily and frequently. The InTouch Health solution leverages the time and expertise of health care professionals across multiple care facilities to improve the efficiency and effectiveness of care delivery. The RP-7 Robotic System enables “remote presence” that allows individuals to project themselves from one geographic location to another, via the wireless mobile robot, such that they can interact from that remote location.
There are two main components to the Remote Presence System: the RP-7 Robot ( Fig. 42.1 ) and control station ( Fig. 42.2 ). The technology is dependent on the use of broadband internet connectivity and existing wireless network within the hospital. Most hospitals already have the necessary network infrastructure in place. The system incorporates two-way audio and video that permits observation and face-to-face interaction between health care professionals and patients. The robot operates in the hospital environment on an 802.11 Wi-Fi network. It is controlled remotely by a health care professional using a laptop or desktop computer, called a control station , which can be linked to the Internet through secure broadband connection. Security protocols enable Health Insurance Portability and Accountability Act (HIPAA) compliant audiovisual communication across the public internet, via a hardwire or cellular connection.
From the control station, the health care professional can drive the robot throughout the facility and communicate in real time with patients, their families, and medical staff. The RP-7 is 5.4 feet tall, weighs 220 lb, and travels up to 2 miles per hour. The anthropomorphic design consists of a “head,” which contains a flat screen monitor, microphone, and digital camera with pan, tilt, and 10× zoom features. The central body houses the speaker, robot-side volume control, and expansion bay that can include a handset for private conversation, a printer for physician’s notes, a digital stethoscope, and other video inputs. The control station has a share-and-display feature that permits images on the control station to be displayed remotely on the head/face of the robot, visible to the audience in the ICU (see Fig. 42.2 ). The robot rides on a patented holonomic platform about 20 inches wide, made up of three balls that allow it to move in any direction, thus greatly increasing maneuverability in tight spaces. An array of infrared proximity sensors placed around the lower section of the robot prevents it from colliding with people or objects. The robot is powered by a rechargeable battery with a life of 5 to 6 hours depending on usage. The robot moves from bedside to bedside under the control of the physician, is capable of traveling on elevators, and has been safely used in operating and procedure rooms. Direct feeds from clinical information systems, such as digital radiology systems and lab systems, are routinely done using the doctor’s workstation. At UCLA, caregivers use the Global Care Quest (GCQ, Mission Viejo, CA) clinical information system to view all types of electronic medical record data. In this fashion, the physician can discuss cases with colleagues, evaluate radiology, and evaluate the bedside environment ( Fig. 42.3 ).
HIPAA compliance and consent for telemedicine is required for any system. There are formal consent forms for all ICU procedures, including RTP. Information technology and compliance officer review of RTP is done routinely at centers that implement the technology; acceptance of RTP technology has been uniform across various regions of the country. Telemedicine has enjoyed a robust experience with medical-legal circles. Increased physician participation and enhanced data review actually makes medical-legal positions better compared with telephonic communication. Multiple systems can be used to provide medical documentation and computer order entry, depending on the particular hospital. A separate internal documentation system, called Stroke-Respond , is currently available, and can integrate into a hospital-based electronic medical record system for transparency. In July 2008, Medicare initiated special ICU billing codes for telemedicine, called 0188T and 0189T. These codes are currently under study by Medicare.
Changing the Physician Response Paradigm
The response to an urgent condition using RTP makes use of visual information and real-time audiovisual communication with the bedside nurse. In a recent study, the importance of visual information was highlighted. The main types of critical data acquired from the telemedicine interaction were judged by the attending physician to be important in the decision pathway. The distribution of these critical data is as follows: 67% visual and 33% verbal information. Of the visual information, the breakdown was as follows: physical examination (40.6%), physiologic monitor graphics (11.5%), printed information from the medical record (5.8%), body language of the nurse (5.7%), information about a tube or catheter (3.8%). In a randomized controlled trial of telemedicine for acute ischemic stroke, telemedicine provided important visual and verbal information, improved diagnostic accuracy, and enabled more complete data assessment as compared with the telephonic paradigm. Similar reliance on visual and graphic information has been reported using the electronic-ICU system (Visicu, Baltimore, MD). These data speak to the need for visual information to be part of the physician response paradigm.
Uses of Robotic Telepresence
Rapid Response to Unstable Intensive Care Unit Patient Condition
The RTP system enables the clinician to make a rapid response to a sudden deterioration in patient condition. In a recent study, Vespa et al demonstrated a marked reduction in physician response time for a face-to-face evaluation of the patient. Overall there was a marked reduction in the attending physician response latency using RTP. The mean response time was significantly shorter for patients with brain ischemia (7.8 ± 2.8 vs. 152 ± 85 minutes) and elevated intracranial pressure (11 ± 14 vs. 108 ± 55 minutes) ( P < .001). Univariate analysis demonstrated a significant reduction in the response latency for each category and for the composite overall ( P < .001). Figure 42.4 illustrates a comparison of face-to-face response times using the RTP system by specific categories.
