Sorry, you need to enable JavaScript to visit this website.
Skip to main content

Diffusion of Responsibility Leads to Danger

Thomas J. Balcezak, MD, MPH, and Ohm Deshpande, MD | October 1, 2018
View more articles from the same authors.

The Case

A 70-year-old man was sent to the emergency department (ED) from a nursing facility due to decreased oral intake, fevers, confusion, and falling urine output. On initial evaluation, he was found to be somnolent but arousable on deep stimulation. Laboratory test results revealed acute-on-chronic renal failure, with a serum creatinine of 12.0 mg/dL and potassium of 7.3 mmol/L. The patient's electrocardiogram showed tall T waves, potentially a precursor of a dangerous arrhythmia due to the hyperkalemia.

The emergency physician initiated medical treatment for the hyperkalemia and consulted the on-call intensivist and nephrologist. The intensivist and nephrologist agreed that the patient required urgent hemodialysis to correct his metabolic abnormalities. The nephrologist asked the intensivist to serially monitor the patient's electrolytes and volume status, continue medical treatment of hyperkalemia, and place a hemodialysis catheter. The nephrologist planned to start dialysis as soon as possible.

Unfortunately, the intensive care unit (ICU) was full and the patient was forced to "board" in the ED. The intensivist was busy with other patients but came down to the ED and placed the dialysis catheter after a few hours. Repeat laboratory test results again demonstrated severe hyperkalemia with persistent electrocardiogram changes. The patient remained confused, likely due to uremia. The emergency physician spoke with the intensivist, who reassured her that the nephrologist had evaluated the patient and that dialysis would be initiated as soon as possible.

On arrival to the ICU approximately 5 hours after the initial labs, the patient was now hypotensive and essentially unarousable. The ICU nurse placed the patient on a cardiac monitor and immediately noticed a sine wave—a dangerous arrhythmia characteristic of severe hyperkalemia. The intensivist came to the bedside promptly, but the patient rapidly went into cardiac arrest. He was intubated and given urgent treatment for hyperkalemia, with restoration of sinus rhythm. At that point, the nephrologist arrived at the bedside and was surprised to find that the patient's hemodialysis had not been started. It turned out that the dialysis nurse had been told to start dialysis after the patient was physically in the ICU and was unaware of the urgency of the situation.

Fortunately, the patient was resuscitated quickly. Hemodialysis was urgently started. The patient did not experience any neurologic consequences from the cardiac arrest and was able to be extubated and transferred to the ward within a few days. He was eventually discharged on hospital day 7 to continue outpatient hemodialysis.

The Commentary

Commentary by Thomas J. Balcezak, MD, MPH, and Ohm Deshpande, MD

Dr. Frederick Southwick's narrative about a series of misdiagnoses and errors that befell his wife when she was admitted to his hospital in the early 1990s, "Who was caring for Mary?" focused on inattention by physicians and a general absence of diligent care among his academic colleagues as the root cause.(1) His update 15 years later revised his conclusions to focus on how defective system design prevented his colleagues from making the appropriate diagnoses, and, when recognized, providing timely and appropriate care.(2) On reading this case vignette, it is easy to focus on individual failures and dismiss the physicians involved as inattentive or disorganized. The reality is that the patient harm was directly related to defects in the design of the system of care.

The clinical vignette clearly illustrates how patients are particularly susceptible to defects in care due to miscommunication, an absence of coordination, and a diffusion of responsibility among frontline caregivers. Emergency departments (EDs), especially at tertiary and quaternary medical centers, are routinely stressed by dramatic shifts in patient acuity, volume, and the need for multiple clinical specialties functioning in parallel in an unstructured environment. The inherent risk posed by overcrowded EDs necessitates having a proactive, systems-based approach grounded in a culture of safety to ensure that the highest value care is reliably provided to fragile patients seeking emergency care.

Overcrowded EDs, particularly when accompanied by boarding admitted patients in the ED until inpatient beds are available, have the potential to cause real harm to patients. Emergency departments are designed to rapidly triage, evaluate, stabilize, treat, and admit or discharge patients. They are not well equipped to provide interventions after formulating a disposition.

In one study of patients requiring intensive care unit (ICU) beds, boarding of patients in the ED for more than 6 hours after admission was associated with a median hospital length of stay of 7 days, versus 6 days for patients whose admission was not delayed. Mortality in the ICU was higher at 10.7% for delayed patients, versus 8.4% for nondelayed patients. Overall in-hospital mortality was 17.4% for delayed patients, versus 12.9% for nondelayed patients.(3) Similar increases in mortality and length of stay are seen for surgical ICU patients delayed in the postanesthesia care unit prior to admission (4), and for inpatients whose transfers to the ICUs are delayed.(5) While specific numbers may differ across the literature, the association between ED boarding and poorer outcomes and prolonged length of stay is consistently observed. Academic centers perform particularly poorly at ensuring coordinated processes in situations where multiple services are involved.(6)

When viewed through a systems lens, it is evident that two issues resulted in this patient's suboptimal clinical course. First, the ED was forced by overcrowding to provide patient care outside its routine operational parameters, creating stress on the system. Second, there was no clear structure to systematically share, assign, and communicate responsibility for the various interventions that were needed to appropriately care for the patient.

High reliability principles are increasingly being applied in health care in order to identify and implement solutions to complex systems issues. High reliability organizations maintain an institutional culture preoccupied with rooting out system failures, integrating lessons into workflows, and developing operational resiliency in the face of a myriad of stressors.(7) In addition to a robust event reporting and safety infrastructure, high reliability is achieved through the so-called CHAMP behaviors: communicating effectively, practicing effective handoffs, ensuring accountability at an institutional and individual level, practicing and accepting mentorship, and practicing a questioning attitude regarding all operations. With regards to the case at hand, the solution lies in designing processes to reduce inappropriate utilization of the ED to minimize times when overcrowding results in boarding of admitted patients and having a robust plan to ensure safe care when boarding is unavoidable.

Low-acuity patients should be seen and managed in fast-track and urgent care areas, and patients within the ED must be triaged to well-defined sections of the ED to best meet their needs. The use of satellite EDs away from the main hospital may also be useful, as they can triage, manage, and admit patients directly to inpatient units, completely bypassing the main hospital ED. In addition, creating a well-resourced transfer center is extremely effective in coordinating hospital-to-hospital transfers by arranging physician-to-physician discussions to ensure patients are routed to the appropriate level of care, again bypassing the need for evaluation in an ED. Lastly, the use of a so-called command center that colocates key operational disciplines and provides real-time operational data to rapidly identify and eliminate bottleneck delays is uniquely effective in smoothing patient flow and reducing ED overcrowding. In combination, these workflows can effectively match patient needs to ED resources, such that inappropriate utilization of the ED is minimized. At our hospital, the implementation of our Capacity Coordination Center has led to improvement in ED throughput metrics, earlier discharge times, and has reduced the stress and confusion associated with ED and hospital overcrowding.

Despite these mechanisms, patient acuity and volume remains high at busy academic medical centers across the country, and admitted patients are frequently boarded in the ED for more than 4 hours. Therefore, it is essential to maintain a simple framework that clearly defines clinical responsibility for the care of these patients. The emergency physicians and main inpatient services must be engaged in order to facilitate a reliably equitable division of labor. For example, at our hospital, our emergency and hospitalist services have agreed that the ED physician will be responsible for ensuring appropriate care is provided for admitted patients remaining in the ED up to the 4-hour mark. After 4 hours, our bed management department notifies a hospitalist that care has been transitioned to the hospitalist service. This transition of care is clearly communicated to the hospitalists via a secure electronic communications platform, and our bed management staff update the responsible provider in the electronic health record.

Geographically localizing hospitalists to manage ED boarders can also be useful in minimizing the risk of competing priorities between patients in inpatient beds and in the ED. Published reports indicate that geographic localization can reduce the perception of wasted time, increase ease of communication, enhance a sense of teamwork, and positively impact throughput metrics.(8) In addition to the primary providers, the deployment of pool nurses to the ED when inpatients are boarding there substantially enhances the quality and safety of care until the patient is transferred to an inpatient bed. In totality, a comprehensive situational awareness of patient flow, multiple layers of patient oversight, clear definition and communication of primary clinical responsibility, and standardization of workflows grounded in high reliability principles will create a resilient system where potential errors are caught before the patient can be harmed.

If applied to the situation illustrated in the vignette, these systems would have decreased overcrowding in the ED by reducing unnecessary delays in the ED, inpatient floors, and ICUs, such that the patient would have been admitted earlier and received more timely care. Moreover, a clear system regarding primary responsibility for boarded patients would have increased the likelihood of rapid placement of a hemodialysis catheter and initiation of hemodialysis.

In summary, high patient acuity and volume can lead to delays in safe transitions from the ED to a setting where definitive therapy can be provided, and the transition period is error-prone if primary responsibilities and support mechanisms are not standardized. High reliability principles are essential in designing and implementing a resilient, patient-centered workflow to mitigate patient harm during that tenuous time. The general solution is to implement processes that promote stewardship of scarce ED resources and have transparent, multilayered processes to manage and support patients when boarding of admitted patients in the ED is unavoidable.

Take-Home Points

  • The high reliability behaviors of communicating clearly, practicing effective handoffs, promoting accountability, providing and accepting mentorship, and practicing a questioning attitude are essential attributes of a safe, patient-centered health care environment across the continuum of care.
  • A proactive approach to matching clinical need to level of care prior to patient arrival at the ED is required to reduce stress on the ED. Satellite EDs, easy routing of patients to fast-track areas, centralized transfer and bed management centers, and access to a wide spectrum of patient flow data to provide accurate situational awareness are effective tools to maximize appropriate ED utilization and reduce overcrowding.
  • When high acuity and patient volume requires that admitted patients be held in the ED for prolonged periods of time, a clear, simple, and coordinated workflow to mitigate diffusion of clinical responsibility is essential to avoid patient harm.

Thomas J. Balcezak, MD, MPH
Chief Medical Officer
Yale New Haven Hospital & Yale New Haven Health
New Haven, CT

Ohm Deshpande, MD
Executive Director, Clinical Operations
Yale New Haven Health
New Haven, CT


1. Southwick F. Who was caring for Mary? Ann Intern Med. 1993;118:146-148. [go to PubMed]

2. Southwick FS, Spear SJ. Commentary: "Who was caring for Mary?" revisited: a call for all academic physicians caring for patients to focus on systems and quality improvement. Acad Med. 2009;84:1648-1650. [go to PubMed]

3. Chalfin DB, Trzeciak S, Likourezos A, et al. Impact of delayed transfer of critically ill patients from the emergency department to the intensive care unit. Crit Care Med. 2007;35:1477-1483. [go to PubMed]

4. Bing-Hua YU. Delayed admission to intensive care unit for critically surgical patients is associated with increased mortality. Am J Surgery. 2014;208:268-274. [go to PubMed]

5. Young MP, Gooder VJ, McBride K, James B, Fisher ES. Inpatient transfers to the intensive care unit: delays are associated with increased mortality and morbidity. J Gen Intern Med. 2003;18:77-83. [go to PubMed]

6. Longo DR, Hewett JE, Ge B, Schubert S. The long road to patient safety: a status report on the patient safety systems. JAMA. 2005;294:2858-2865. [go to PubMed]

7. Weick KE, Sutcliffe KM. Managing the Unexpected: Assuring High Performance in an Age of Complexity. San Francisco, CA: Jossey-Bass; 2001. ISBN: 9780787956271.

8. Bryson C, Boynton G, Stepczynski A, et al. Geographical assignment of hospitalists in an urban teaching hospital: feasibility and impact on efficiency and provider satisfaction. Hosp Pract (1995). 2017;45:135-142 [go to PubMed]

This project was funded under contract number 75Q80119C00004 from the Agency for Healthcare Research and Quality (AHRQ), U.S. Department of Health and Human Services. The authors are solely responsible for this report’s contents, findings, and conclusions, which do not necessarily represent the views of AHRQ. Readers should not interpret any statement in this report as an official position of AHRQ or of the U.S. Department of Health and Human Services. None of the authors has any affiliation or financial involvement that conflicts with the material presented in this report. View AHRQ Disclaimers
Related Resources From the Same Author(s)
Related Resources