For many clinical practitioners in the emergency department, the introduction of information technology has been an ordeal. The early versions of the technology proved more of an impediment to smooth and efficient workflow than anyone could have imagined. Many practitioners felt duped: The promises of what a comprehensive ED information system could be and do just did not jibe with the realities of what was happening on the front lines.
Physicians saw disruptions in work flow and flawed systems that required more work and workarounds to get basic ED tasks done. It seemed that many processes that had functioned quite well with paper and pen were cumbersome and faulty when embedded in the new technology. Once- efficient departments watched as productivity (measured by time data and patients per hour) plummeted in the name of new information systems and IT support. With this background, it should come as no surprise that most emergency physicians reluctantly embrace the introduction of IT into their facilities.
ED staff work under pressure in a hectic environment characterized by frequent interruptions. They cope with an unpredictable, highly variable patient mix, often with limited access to important information about patients’ previous conditions and treatments. ED patient care is a collaborative and complex process, requiring exchanges of information among the different administrative and clinical staff within and outside the ED. And yet, the ED work environment operates as a zero-sum game, playing relatively fixed resources against variable demand. These finite resources must be distributed based on need and judgment with regular adjustments. These adjustments depend on the acquisition and organization of ever-changing sets of data. One study (Acad Emerg Med 2004;11:1114) characterized the ED environment as “highly variable, evanescent, contingent, uncertain, poorly bounded, resource constrained and beholden to many external influences. For example, patients just ‘show up.’ In order to reconcile care resources with patient care demands, ED teams routinely engage in fluid, dynamic cognitive activities that require flexible, reliable artifacts to support them. They seek, track, digest, calibrate, probe, evaluate, verify, and share information. They also plan, speculate, re-plan, and make trade-off decisions. No circumstances make it acceptable to suspend patient care. Because of this, practitioner teams have developed a range of sophisticated strategies to manage the balance between work and demand. In order to succeed, automation needs to be a team player in the ED setting.”
In this special feature, the authors lay out the aspects of workflow unique to the ED and explain how IT can facilitate them. A vision of a futuristic ED that uses information systems and technology to make the ED safer and more efficient is described and offered for your imagination and optimistic consideration. The ED work environment has elements that are idiosyncratic and unique to the setting:
Multiple patients and interruptions: When compared with office-based physicians, the emergency physician manages more patients at the same time and experiences more interruptions and breaks in task. An emergency physician is interrupted 10 times an hour, and spends the majority of his time multitasking. Health IT solutions need to enable the physician to recall where he was working, to begin work on a divergent path, and then to return to the site of the previous task. (Ann Emerg Med 2001;38:146.)
Coordination and sequence of care: Care is provided to a high volume of patients whose problems and acuity vary widely, often with large information deficits. Variation in practice leads to inefficiencies and errors. Fortunately, much of the care given in the ED can be organized around chief complaint-driven clinical pathways. While there are limited evidence-based pathways, locally crafted pathways can reduce error and improve efficiency. Technology should support these protocols by prompting the clinicians when information suggests a certain pathway and cueing the next step in the pathway. (Quality Matters: Solutions for the Safe and Efficient Emergency Department. Chicago: Joint Commission Resources Publishing; 2009.).
Time constraints: Many of the clinical entities that present to the ED involve diagnoses that are “on the clock.” Acute MI, acute stroke, and sepsis are just a few in which the outcome directly correlates with timelines of care. (Ann Emerg Med 2007;50:489; Emerg Med J 2008;25:403; Acad Emerg Med 2008; 15:190.) Add to that impetus the patient's expectations for timeliness and the capacity constraints of the department, and you have all the reasons you need to make every encounter expeditious. (Ann Emerg Med 1993:22:586.) Information technology could be used to cue staff when specific time intervals are looming regarding particular time-sensitive presentations.
Information management: The sheer amount of information exchanged during an ED encounter continues to grow and must be managed. Information comes from many sources including old records, offices, radiology, the lab, the pharmacy. (Ann Emerg Med 1998;32: 65.) An information system that collates the data by patient reduces physician time and energy spent on the quest for relevant information and facilitates workflow and expedites patient flow.
Surges and crowding: The majority of emergency departments nationwide are spending part of each day at overcapacity, with more patients than care spaces. Information technology that helps forecast crowding facilitating resource reallocation might expedite patient flow, reducing crowding. (Acad Emerg Med 2008;15:171.)
Handoffs and transitions of care: Many interfaces and transitions of care occur in the course of an ED visit. The ED transitions between outpatient and inpatient settings, and these transitions are a common source of lost information. (Int J Med Inform 2007;76 [11-12]:801.) ED information systems that are part of system-wide information systems should improve health care transitions.
In the ED of the future, IT will be harnessed to expedite care, make clinical care more standardized, and facilitate the transfer of information required for health care delivery. Hold onto your hat, and ride along as a futuristic ED encounter uses technology to the max. Many of these innovations are not yet published, but imagine the very technology that bedevils you in the ED today making your job easier tomorrow, with fewer process defects and more predictability.
Intake: Starting the Encounter
Imagine an ED encounter in the near future. Patients will present to the ED and self-register using smart cards. They will self-triage using a computer-prompted interview. This technology is already being tried at London Health Sciences Centre in Ontario. Ninety-four percent of first-time users of a computer-based triage process liked the model and would use it again. Intake took less than five minutes on average, and the computer outperformed physicians in historical elements associated with the triage of patients with abdominal pain.
In an even more futuristic model, the information will reside in a “cloud,” and biometrics will be used as the ultimate authentication mechanism; no card required. Cloud computing is an emerging technology that uses the Internet and central remote servers to maintain data and applications. It allows consumers and businesses to use applications without installation and to access their personal files at any computer with Internet access.
Three other novel IT applications that will affect ED intake deserve mention. At the Medical College of Georgia, a telemedicine triage project is being piloted. The ED was subject to high numbers of transfers from surrounding area nursing homes for problems that could be managed without transferring the patient. Using telemedicine technology, some unnecessary transfers were avoided, and problems were managed without transport. Another novel IT application, palmar scanning for patient identification, is being tested as a bio-identification technique at Carolinas Medical Center. The technology allows the instantaneous creation of a unique patient ID using infrared scanning of the patient's palm. This technology may be used on unconscious or incapacitated patients, and it prevents duplicate identifications from being generated for the same patient. This ID process takes less than 15 seconds, and has reduced the door-to-physician time to 45 seconds. A third novel IT application, assessment over the Internet, may abort the ED encounter altogether. Microsoft and Emory University developed a web-based system that interviews patients online to determine whether they need to see a health care provider. (www.h1n1responsecenter.com)
Generating the Medical Record
Another problem inherent at the start of the ED encounter involves the retrieval of existing health information to facilitate care. Imagine — finally — the existence of an electronic health record that allows seamless connection of all available health information instantaneously. Drs. Theodore Chan and James Killeen at the University of San Diego have built interfaces between information databases so that the ED record is immediately populated with the most current health data on the patient. This expedites intake, and saves human time and effort on the front lines. Imagine starting an ED encounter with a patient in which all current and historical information on the patient is summarized and pushed to you, rather than chasing the data through unreliable historians and incomplete or unavailable records.
Health care providers are increasingly coming around to the idea of standardization. By reducing variation, health care in the ED becomes more efficient, safer, and less costly. When evidence-based guidelines are lacking or inconclusive, the development of local protocols that standardize diagnostic and therapeutic steps will become the norm for the common ED chief complaints. At HCA in Denver, Dr. Dickson Cheung is leading his system in chief complaint-driven ED protocols which streamline and standardize care. The early data are quite promising. Imagine clinical decision support that goes a step further. Imagine a futuristic system that prompts the clinician to consider certain patient pathways as the ED encounter progresses. How about a personal communication device worn on your belt to cue you: “Dr. Smith, Patient X is a 24-year-old woman with abdominal pain. Her pregnancy test is positive. Order an Rh factor and a pelvic ultrasound? YES, NO.”
IT also will undoubtedly assist emergency physicians in managing the growing volume of information associated with ED encounters. Often physicians are faced with pages and pages (or screen after screen) of data embedded in long columns with normal data ranges included. The significant findings are often lost in the morass. Imagine IT capable of pulling out crucial data and organizing it in a meaningful way for the clinician.
Now imagine this IT support going a step further and suggesting the next clinical step. Diagnostic or therapeutic interventions are proposed with prompts that the clinician can simply turn on or off as order sets. Try to imagine you as a practitioner with a complete electronic health record at your fingertips. A patient checks in, and you immediately have access to every piece of health information available on the patient. It is organized for you with easy displays of information in the form of logical summary data, and there is no need to read through complex detailed charts. Technology assists you with pattern recognition, one of the strongest skills the emergency physician hones in day-to-day practice.
Clinical Decision Support
In the near future, you will come to work and log on to a computer, setting up the screens and real estate just the way you like it. As you pick up patients, their health information is organized under separate tabs that you may toggle among them. A computer tablet lets you begin your data collection about the patient with a template touch screen that populates a medical record embedded in the information system. Historical and physical exam data entered lead to clinical decision support suggestions for diagnostic and therapeutic interventions. Documentation is done in real time as the encounter progresses, allowing capture of important information that self-populates the medical record, with, for example, lab and imaging data automatically appearing in the record under “hospital course” without needing separate entry by the physician. Clinical decision support also suggests a differential that the physician can address in the medical record and the latest treatment plan for common diagnoses. All of these improvements not only aid the physician with prompts, reminders, and data organization, they are built into the work flow without disrupting it. This is the key to successful IT systems: They must be built on a sound understanding of ED workflow and function seamlessly within that framework.
Finally, as patients leave the department, there is opportunity for IT to support this process. Patients may take their “smart cards” (or their retinas or palms if bio-identification is in place) to a discharge kiosk. There discharge instructions and prescriptions are dispensed, freeing up staff for patient care tasks. Informational videos also could be shown to patients needing further tests or treatment. The computer could even talk the patient through scheduling or complicated treatment regimens.
Technology also affords an opportunity to facilitate the handoff with admitted patients, with telemedicine used to aid in communication between caregivers. In a study of handoffs in high-risk settings (high-speed rail, nuclear power, air traffic control), the verbal interaction at handoff identified significant issues 25 percent to 30 percent of the time. (Int J Qual Health Care 2004;16:125.) Technology could facilitate the exchange, and smart data transfers with self-populating handoff forms could make the transfer of care seamless and eliminate the human hand in data entry.
IT systems can help support work flow and patient flow. Unfortunately and too frequently, actual implementations have failed to do so. One truth that must be understood and accepted by users is that even the best technology when laid upon faulty operations will yield predictably poor results. In the past, implementers, emergency physicians among them, “paved the goat paths” rather than reengineering processes to take advantage of the new technology. Technology is an enabler; it can be used to make change in processes and often consequently changes in culture. Or through the missed opportunity, further embed poor processes and poor results into a dysfunctional culture.
You have every right to be skeptical given the propensity for information technology to disrupt workflow and cause mayhem in your department. But keep the faith. This vision of a future with information technology as a team player may not be as far off as it seems.