SEARCH

SEARCH BY CITATION

As a specialty, emergency medicine (EM) prides itself on being able to handle multiple patients at once, and most providers would probably characterize themselves as multitaskers. Indeed, it is difficult to argue that the physician who must manage the well-being of five, ten, or even more patients at once is not multitasking in some way. However, it has been suggested that humans—even emergency physicians (EPs)—are simply not capable of “true” multitasking; in other words, humans cannot perform one task in parallel with another.1 So, can we multitask or not? In this commentary, we will explain that humans do in fact have powerful multitasking abilities, but that we also experience predictable limitations of short-term memory as well as predictable patterns of error in task-switching and task-stacking. These capabilities and limitations need to be addressed when training residents and designing emergency department (ED) systems.

Our everyday experience tells us that for some combinations of tasks, multitasking is possible, while for others it is not. For example, EPs can only carefully consider one patient’s diagnosis at a time. Diagnosis is an example of a controlled process, requiring conscious, effortful thought that cannot be done in parallel with other tasks.2,3 However, EPs can skillfully suture a wound while simultaneously giving verbal discharge instructions and precautions to the patient or while giving verbal orders for another patient whose nurse has just entered the room. This is possible because the suturing task, after sufficient practice, becomes automatic and is easily parallelized. Further understanding of the controlled versus automatic processing model will help us teach EM residents which tasks they should expect to be able to perform in parallel and which tasks they should not.

Given that controlled tasks cannot be parallelized, the ability to switch controlled tasks intelligently and with minimal error is central to EPs’ performance. However, there are predictable patterns of error that occur when managing multiple controlled tasks. For example, suppose that an EP orders complete blood counts and basic electrolyte panels on 10 of his or her patients during a typical shift and thyroid function tests on only one of these patients. As the thyroid tests often takes much longer to be processed, and most EDs lack feedback loops that actively indicate pending (but not completed) lab results, the function of remembering to follow up on the thyroid test often rests with the ordering physician. Physicians must then try to either remember this abnormality or habitually recheck for abnormalities at discharge time. However, the former requires short-term memory, which is subject to well-known limitations.4 The latter task, habitually rechecking for abnormalities, requires vigilance and tends to break down as task loads increase, a condition common in the ED setting. Vigilance also experiences a well-established and consistent decrement over time, even in the best of situations.5,6

Adding in environmental factors such as frequent interruptions, task-switching, and task-stacking further complicates both these issues.7–11 In this specific case, there is a risk that the physician will discharge the patient before the thyroid results are available. Without robust predischarge checks for incomplete or abnormal tests, this could result in a delay in diagnosis because of a missed abnormal result. Most EDs, even those with ED information systems, lack a robust method to provide reminders of outstanding labs, so instead current systems rely on the ability of physicians to overcome well-known limitations of human cognition, including short-term memory.

While such predictable limitations may exist in humans and need to be managed, an important capability demonstrated by many humans is intelligently deciding when to switch tasks.12 This in itself is a form of multitasking: it requires the automatic, constant querying of the periphery of the sensory environment to determine when to switch from one controlled task to another. EPs use task-switching as a method to juggle multiple patients, many of whom are at different points in the process of care. The ability to prioritize tasks that must be completed and store others for eventual completion (referred to as task-stacking11) while still remaining sensitive to an unplanned need to switch tasks, requires clinicians to multitask. From this perspective, it is clear that EPs are constantly multitasking. For example, an EP is almost always processing nearby sensory input from the environment, such as noticing a sudden cluster of activity in a patient’s room, an abnormal bedside alarm, or a patient vomiting in the background. This specific form of expert multitasking requires training and experience, as well as a system and an environment that provides adequate support for it. The effect of the environment on this type of multitasking was demonstrated by one study that showed chest pain patients had longer door-to-doctor time intervals when they were placed in rooms further from the physician workstation or in rooms with a solid door.13 Both the distance and the door reduced the opportunity for environmental stimuli which, once processed, would likely have led to task-switching based on priority. This type of multitasking is one of the most important ways in which humans successfully multitask both in everyday life and especially in the ED work environment.

This all suggests that EM residents should be trained to recognize where their natural abilities allow them to become expert multitaskers, while still being cautious of those task-switching and task-stacking pitfalls that are common to most, if not all, humans. We can and should teach strategies to help residents avoid these pitfalls. For example, in the case of vigilance, we could teach residents the importance of using cognitive artifacts, such as notes on a whiteboard or entries in an EDIS system,14,15 or arranging cues from other ED staff, such as asking a nurse to “let me know how the p.o. trial goes so I can reevaluate.” Although such strategies may have started as workarounds in lieu of system fixes, this type of adaptability is a form of resilience and is characteristic of EM providers.16 We could also explicitly train residents on strategies to avoid certain interruptions when it would be detrimental to their work processes, while also keeping an open eye or ear to know when switching tasks is appropriate to prioritize another patient over the current one. These skills are critical to successful management of multiple patients. The idea of a hierarchy of interruptions is relatively unstudied and is an important area for future study, so that someday ED staff may have a common understanding about when interruptions are appropriate and when the information they provide can and should wait.

The design of the ED environment and systems is critical to the EP’s multitasking and multiple patient management abilities. Information technology systems, bedside monitors, communications devices, and physical layout should be designed to match natural human cognitive abilities and to capitalize on human strengths, such as the capability to notice events in the sensory periphery that require switching tasks.12 If ED systems are designed (intentionally or inadvertently) to rely on human cognitive abilities that are known to be more error-prone, such as task vigilance in an interruption-driven environment, or the multitasking of two controlled cognitive processes, then predictable and inevitable patterns of error will be the result. However, it is clear that some interruptions are unavoidable; more research into how changes to support systems and training of residents could mitigate these interruptions is warranted.

In summary, humans can multitask and are actually quite adept at one form of multitasking: the ability to collect and process peripheral sensory information and to use this information to decide when to switch from the current controlled task to another controlled task. This ability, along with the task-stacking and task-switching abilities that it enables, is a large part of how EPs successfully perform their work and therefore is a concept that is very important to convey during EM residency training.

The authors acknowledge Lindsey Clark, MA, for her editorial review. Dr. Fairbanks is supported by a Career Development Award from the NIBIB (K08-EB009090), and Dr. Stephens is supported by a grant through EMF from the Emergency Medicine Patient Safety Foundation.

References

  1. Top of page
  2. References
  • 1
    National Public Radio. Former Google Executive On Getting Organized [Radio Interview]. Available at: http://www.npr.org/templates/story/story.php?storyId=125005439. Accessed Nov 6, 2011.
  • 2
    Schneider W, Shiffrin RM. Controlled and automatic human information processing: I. Detection, search, and attention. Psychol Rev. 1977; 84:166.
  • 3
    Shiffrin RM, Schneider W. Controlled and automatic human information processing: II. Perceptual learning, automatic attending, and a general theory. Psychol Rev. 1977; 84:12790.
  • 4
    Baddeley A. Working memory. Science. 1992; 255:5569.
  • 5
    Grier RA, Warm JS, Dember WN, et al. The vigilance decrement reflects limitations in effortful attention, not mindlessness. Hum Factors. 2003; 45:34959.
  • 6
    Mackworth NH. The breakdown of vigilance during prolonged visual search. Q J Exp Psychology. 1948; 1:621.
  • 7
    Chisholm CD, Dornfeld AM, Nelson DR, Cordell WH. Work interrupted: a comparison of workplace interruptions in emergency departments and primary care offices. Ann Emerg Med. 2001; 38:14651.
  • 8
    Chisholm CD, Pencek AM, Cordell WH, Nelson DR. Interruptions and task performance in emergency departments compared with primary care offices [abstract]. Acad Emerg Med. 1998; 5:470.
  • 9
    Coiera EW, Jayasuriya RA, Hardy J, Bannan A, Thorpe ME. Communication loads on clinical staff in the emergency department. Med J Australia. 2002; 176:4158.
  • 10
    Fairbanks RJ, Bisantz AM, Sunm M. Emergency department communication links and patterns. Ann Emerg Med. 2007; 50:396406.
  • 11
    Wolf LD, Potter P, Sledge JA, Boxerman SB, Grayson D, Evanoff B. Describing nurses’ work: combining quantitative and qualitative analysis. Hum Factors. 2006; 48:514.
  • 12
    Woods DD, Sarter NB. Capturing the dynamics of attention control from individual to distributed systems. Theoret Iss Ergonom. 2010; 11:728.
  • 13
    Hall KK, Kyriacou DN, Handler JA, Adams JG. Impact of emergency department built environment on timeliness of physician assessment of patients with chest pain. Environ Behav. 2008; 40:23348.
  • 14
    Bisantz AM, Pennathur PR, Guarrera TK, et al. Emergency department status boards: a case study in information systems transition. J Cog Engin Decision Mak. 2010; 4:3968.
  • 15
    Pennathur PR, Cao D, Bisantz AM, et al. Emergency department patient tracking system evaluation. Int J Indust Ergonom. 2011; 41:3609.
  • 16
    Wears RL, Perry SJ, Anders S, Woods DD. Resilience in the emergency department. In: Hollnagel E, Nemeth CP, Dekker SW, eds. Resilience in the Emergency Department. Aldershot, UK: Ashgate, 2008, pp 14358.