Welch, Shari J. MD
Dr. Welch is the quality improvement director in the emergency department at LDS Hospital in Salt Lake City, a clinical faculty member at the University of Utah School of Medicine, faculty at the Institute for Healthcare Improvement and the Urgent Matters Project for the Robert Woods Johnson Foundation, a quality improvement consultant to Utah Emergency Physicians, and a member of the Emergency Department Benchmarking Alliance.
Often with the busy paces kept by emergency and office physicians, there is contention over who should hold the phone for whom when the two try to connect. The office physician is with a patient when the ED calls, and then the emergency physician is doing a procedure when the office attending returns the call, and so forth. Harsh words sometimes pass as the office physician exclaims, “I'm as busy as you are!” But in fact, there are few specialties in medicine that require a physician to multitask as often and be interrupted as frequently as the emergency physician.
Two studies by Chisholm et al help quantify the differences between the work and pace of an office-based physician and an emergency physician. In a 2000 article appearing in Academic Emergency Medicine, the authors reported the results of a time study that revealed that the average emergency physician has 12.1 patient encounters in a 180-minute cycle. During this cycle, he will be interrupted 30.7 times and have 20.7 breaks in task. (Acad Emerg Med 2000;7:1239.) We know from work on human cognition that interruptions are one of the biggest sources of human error. Because the average individual can keep no more than seven items in his short-term memory, small wonder that interruptions often force the emergency physician off task and that these tasks are forgotten completely as he moves through a busy shift.
Some hospitals have created interruptionfree zones for nurses mixing medications
These findings were replicated by Chisholm et al in 2001 in the Annals of Emergency Medicine. He again found roughly 10 interruptions per hour for the emergency physician compared with 3.9 interruptions per hour for the office physician. Even more telling was the fact that the emergency physician in this study spent two-thirds of his time managing three or more patients while the office physician in practice spent less than one minute per hour managing more than a single patient. (Ann Emerg Med 2001;38:146.)
The ED work environment seems designed to promote human error. Noise, level of activity, heat, visual stimuli, and fatigue contribute to this problem. Waits and delays, medication errors, wrong tests performed on patients, falls, and other safety issues occur with discomfiting frequency in most emergency departments.
In the nursing community, interruptions during mixing and administering medication have been recognized as an etiology for medication errors. Many departments are testing the concept of the interruption-free zone. One ED began by taping off an area around the medication counter in the ED. A rule was enacted that a nurse in that zone could not be interrupted, and the results were so convincing that the hospital administration laid tile to permanently mark the interruption-free zone.
Shouldn't such interruption-free zones be created for other members of the ED team, including the physician? While recently acting as an operations consultant for a South Dakota ED, I noticed that a nurse accompanied the emergency physician to the vending machines for an early morning snack during the graveyard shift. They were discussing a patient's care. When the physician stopped to make a selection from the snack machine, the nurse stopped talking to allow him to concentrate and select an Almond Joy from the machine.
When that same physician was looking at lab results and fielding a phone call an hour later, the nurse thought nothing of shoving a normal EKG in front of the physician's nose, causing him to lose concentration with the multitasking and critical thinking he was doing at the moment. Why is the etiquette surrounding interruptions at the vending machine different from those at the computer in the ED? I also have observed two physicians putting in a technically difficult central line under extremely stressful resuscitation conditions when a staff member shoved an ABG slip in front of them. The two physicians looked up for a split second and lost their timing and their line!
As we move forward with the work of operations improvement in emergency medicine, a legitimate focus ought to be in creating a work environment more conducive to error-free performance. Part of this would include spaces for physicians to synthesize and make sense of patient data and perform high-level cognitive functioning. Most physicians work in an environment with high noise levels and frequent interruptions. (Am J Emerg Med 2005;23:332; Acoust Soc Am 2007;121:1996.)
On a related topic, the work of the emergency physician is increasingly data-driven and computer-dependent. Shouldn't ED workers in general and physicians in particular have a computer with a single log-on process at the beginning of the shift, and have free access to physician data, medical reference material, and decision support? At LDS Hospital where I work, the attending physicians have a computer bank, and each physician claims a work station for his shift. He can set up his station for maximum efficiency according to his work habits. These types of arrangements to facilitate workflows should be part of every new ED design. Shouldn't we design spaces conducive to high-level performance as we do our important work?
▪ Emergency physicians are interrupted 30.7 times in every 180-minute cycle.1
▪ They experience 20.7 “breaks in task” per cycle.1
▪ EPs are interrupted 9.7 times per hour while office-based physicians are interrupted 3.9 times per hour.2
▪ Emergency clinicians spend two-thirds of their time managing multiple patients (three or more) while office-based physicians spend less than one minute per hour managing multiple patients.2
Source: (1) Acad Emerg Med 2000;7(11):1239; (2) Ann Emerg Med 2001;38(2):146.
© 2007 Lippincott Williams & Wilkins, Inc.