Lost in Transition
- Spotlight Case
Approach to Improving Safety
Setting of Care
- Provide an overview of transitions in continuously operating industries
- Review cognitive error
- Describe the complex dynamics of transitions in emergency care
- Provide suggestions to decrease risk at shift transitions
Case & Commentary: Part 1
A 41-year-old woman came to the emergency department (ED) with mental status changes. She had been diagnosed with a urinary tract infection and started on oral ciprofloxacin 4 days earlier. She had fever, nausea, and vomiting in the days preceding presentation. She did not have headache, focal weakness, or numbness. Past medical history was otherwise unremarkable.
On physical examination, the patient was afebrile, with sinus tachycardia (heart rate 123 beats per minute) and otherwise normal vital signs. Able to follow most commands, she was alert but oriented to person and place only. Neurologic examination was otherwise non-focal. There were no signs of meningeal irritation.
Approximately 40 minutes after the patient arrived, initial laboratory results returned and included white blood cell count 12.7 K/µL with 89% granulocytes, hematocrit 20.2%, glucose 204 mg/dL, blood urea nitrogen 36 mg/dL, serum creatinine 1.4 mg/dL. Urinalysis showed moderate blood. Platelet count was pending at that time.
Sixty minutes after arrival, the patient was admitted to the internal medicine service with a diagnosis of anemia and hematuria in the setting of a urinary infection. The medicine team completed the admission paperwork, with plans to administer empiric broad-spectrum antimicrobial agents and packed red blood cells for the severe anemia. The outgoing ED physician had just completed the shift and signed the patient out to the oncoming colleague as ''admitted,'' with care already transferred to the internal medicine service.
Transitions are necessary in any industry with continuous operations, and health care is no exception. Shift changeover occurs when two or more workers exchange mission-specific information, responsibility, and authority for an operation.(1,2) This moment of care has been characterized as both an opportunity for rescue and a threat to safety.(2,3) Most health care teams now consist of far more than one nurse and one physician—with each additional worker responsible for care, the potential for error or rescue increases. The ability to effectively conclude one level of care and transfer to the next is critical to quality and safety in the ED.
Any discussion of shift changeover involves the complex interplay of two or more individuals working in a broader system. Certain features are common to most ED handoffs. Geography often dictates a starting and ending point (ie, bed 1-10). Interruptions occur frequently. Physicians often perform sign-over separately from nurses and other specialists. The chart or other formal written documentation is rarely used for transitions.
On the other hand, many of the aspects of ED transitions are highly variable.(4) Transition can occur in front of the patient, at a remote setting, or anywhere in between. The hand-off can be predominantly interactional or transactional. That is, the transition may be a two-way dialogue that leads to ''shared sense-making,'' or it can be a one-way purge of information. In all cases, the ongoing demands of the ED necessitate that handoffs be characterized by both brevity and completeness.
Although few EDs have formal systems for transition, one can identify four phases in most ED handoffs.(5) These phases include (i) pre-turnover, (ii) arrival, (iii) meeting period, and (iv) post-turnover. These phases tend to be common in all health care settings with high consequences for failure, like the ED.(6,7)
Examples of effective transition can be found in
reliable organizations such as nuclear powered submarines,
trauma centers, and NASA. Los Angeles–class nuclear
submarines use ''precise, unambiguous, impersonal, and
efficient'' language between the officer on duty and the sonar
technician to navigate safely. Commands, readbacks, and monitoring
help bridge authority
gradients and ensure the crisis-resistant performance vital to
Dedicated local and regional trauma centers also depend on
operational and clinical excellence to achieve
''failure-resistant performance.'' For example, a trauma
nurse asks if anyone has informed the operating room (OR) of a
patient's pending transfer to the OR. Acknowledgment and
action from the team leader follow this inquiry, ensuring that all
the providers have situational
awareness and that there is timely transit to the OR.(9)
These are only a few of the useful strategies. In fact, in an
article on handoff strategies, Patterson describes 21 techniques
used by NASA, nuclear power plants, railroad dispatch centers, and
ambulance dispatch centers.(10)
The techniques include verbal, face-to-face, and interactive
questioning that is coordinated with written summaries prior to
shift change. Additionally, readback, limits on interruptions,
unambiguous transfer of responsibility, and pre-turnover data scans
are all used to ensure joint sense-making at the completion of
Case & Commentary: Part 2
Four hours after arrival, the laboratory called the ED to report a critical lab result, a platelet count of 4,000/mm3 (normal range 150,000-400,000). The critical result was received by the ED unit secretary. It is unclear who this information was passed on to, but neither the ED attending nor the internal medicine service was made aware of this lab result.
Sixteen hours after the patient presented to the ED, the internist noted the abnormal finding when checking the morning lab data. She made a tentative diagnosis of thrombotic thrombocytopenic purpura (TTP). The patient required transfer to the ICU because of progressive deterioration in mental status and was eventually intubated. Hematology consultation was obtained to initiate emergent plasma exchange for treatment of TTP.
The evaluation of mental status change demands the best of the entire health care team. Work-up and decision making vary from patient to patient and fit poorly into clinical care algorithms. Care must be customized. Arriving at the correct diagnosis often involves multiple practitioners; lengthy, detailed, and occasionally invasive work-up; and explicit attention to detail. TTP is a rare (15 patients per million per year) clinical syndrome characterized by low platelets, anemia, fever, mental status change, and acute renal failure. Accurate evaluation, diagnosis, and treatment of TTP are barometers of both the competency of individual providers and their teams, as well as the quality and safety of the system. Transitions in care magnify weaknesses present in individuals, teams, or the broader system.
Communication and Cognition: Evaluating System and Human Actions
Given the rarity of TTP and the often ambiguous and non-specific clinical presentation (eg, fever, anemia), cognitive errors are common and often lead to delays in diagnosis. The disease is fatal in up to 90% of cases without effective treatment, but prompt treatment with plasma exchange can be lifesaving.(11,12) In this case, two major mishaps resulted in late administration of plasma exchange, the communication of serious lab abnormalities, and the recognition of the clinical entity, TTP.
Communication and the System: Failure to Identify Serious Laboratory Abnormality
A combined human and system error resulted in delay in acknowledgment of low platelets. At sign-out, the ''platelet count was pending.'' Two human-transition issues arise. First, most clinicians recognize that a delay in reporting a test often means the result is abnormal (the extra time reflects the lab's protocol to perform further authentication prior to reporting the test result to the caregivers). Thus, knowing that all other CBC results were obtained should raise the suspicion that the unreported platelet count was abnormal. Making this assumption, the outgoing team must make it explicitly clear this test result should be evaluated prior to disposition or further management. Additionally, the outgoing team must ensure that the incoming team has clearly understood this request and the reason for it. The most notable system error is that there simply was no consistent process established to deliver the information to the care team.
A variety of methods are used in the ED and across health care to communicate important information such as critical test results. Direct, face-to-face communication is frequently the most desirable and effective way of assuring flawless transfer of information. However, indirect methods such as text-paging, email alerts, indirect communication (via overhead calls or two-way messaging), color-specific paper charting, or written documents are also used to deliver ancillary test results. Whatever method is chosen (direct communication is the best), it also needs to be coupled with a shared interpretation of what constitutes critical results. Such interpretations cannot be made in a vacuum but require an appreciation of each patient's unique characteristics. For example, a hematocrit of 24% in a patient with coronary disease and no history of anemia constitutes a medical emergency, while the same hematocrit in a young patient with chronic renal insufficiency would barely elicit a yawn.
Cognition and the Human: Failure to Reach the Correct Diagnosis
In this case, errors occurred at several of the
transitional phases: pre-turnover, meeting, and post-turnover. In
the pre-transition phase, the ED doctors were content with the
diagnosis of urinary infection as the cause for altered mental
status and demonstrated anchoring bias as
contradicting evidence was set aside. During the transition itself,
it was unclear who was responsible for follow-up and interpretation
of these results—a systems problem. Finally, after the
turnover, the clinicians became subject to a framing effect, as all
the findings were considered in light of the diagnosis of
''urinary tract infection and anemia,'' rather than
reconsidering the new information as it came in.(13) In this way, the case illustrated Canadian ED
physician and safety expert Pat Croskerry's observation,
''When the diagnosis is made, the thinking
Case & Commentary: Part 3
Despite these interventions, the patient's status continued to deteriorate. The patient died the following day, within 48 hours of presentation to the ED.
TTP requires timely treatment in order to reduce mortality. It shares this time dependency with many other ED diagnoses and situations, many of which have been converted into quality metrics (eg, 10 minutes to EKG for patients with chest pain, or 240 minutes to antibiotics for patients with pneumonia). These time-dependent outcomes invariably rely on many individual processes and interactions; the transition phase is just one of these. As Richard Cook notes, ''Catastrophe requires multiple failures—a single point failure is not enough.''(14)
Communication failures fall into three categories.(15) First, system failures occur when communication channels are used infrequently, are non-functional, or are non-existent. Second, message failure occurs when there is poor or non-existent transfer of information. Third, reception failure occurs with misinterpretation or late arrival of proper information. In one recent study, a written (computerized) sign-out sheet for surgical residents augmented verbal sign-out and was widely adopted.(16) It was able to centralize and organize information and daily work. Another recent analysis describes ''collaborative cross-checking'' as a way to improve resilience of the health care transition.(17) In one illustrative case, a nurse overheard a confusing order at sign-over and questioned the order with the physician. The order was re-evaluated, which prevented a potential catastrophe. That case illustrates that proximity may well be serendipity, and that cross-checking can serve as a valuable rescue mechanism.
In industries with continuous operations such as health care, it is vital that we look for tools to improve resilience at shift transitions. The unique need for customization, the requirement to manage surges in volume, the varied professional expertise, and the tight coupling of different processes to each other make this a challenge. Take-home points for potential improvement include:
- Reduce transitions when they are not necessary.
- Assure clear delineation of authority and responsibility at times of transition. (''Dr. Smith, please check the platelet count; it is currently pending.'')
- When possible, use discrete end-points, simply communicated. (''If the platelet count returns below 30,000, call hematology.'')
- Describe symptoms and ancillary studies to support interpretations, particularly when the work-up is continuing. Use diagnoses only when they are clearly supported. (''This patient has recently been treated for UTI and presents with mental status change, hematuria, and anemia,'' instead of ''This patient has UTI and anemia.'')
- Encourage and accept cross-collaborative feedback and questioning.
- Transition—like all high-stakes technical procedures—should be observed and taught to students before they become transition leaders.
- Limit interruptions. If interruptions are necessary, delay transfer of responsibility.
- Use written and verbal tools to augment transfer knowledge.
Assistant Professor of Emergency Medicine
Northwestern University, The Feinberg School of Medicine
Faculty Disclosure: Dr. Beach has
declared that neither he, nor any immediate member of his family,
has a financial arrangement or other relationship with the
manufacturers of any commercial products discussed in this
continuing medical education activity. In addition, his commentary
does not include information regarding investigational or off-label
use of pharmaceutical products or medical devices.
1. Lardner R. Effective shift handover: a literature review. Health and Safety Executive. June 1996. Offshore technology report-OTO 96 003. Available at: http://www.npsf.org/download/Focus2004Vol7No2.pdf. Accessed February 3, 2006
2. Wears RL, Perry SJ, et al. Shift changes among emergency physicians: best of times, worst of times. Proceedings of the Human Factors and Ergonomics Society 47th Annual Meeting; October 13-17, 2003; Denver, CO.
3. ACEP Patient Safety Task Force. Patient Safety in the Emergency Department. Dallas, TX: American College of Emergency Physicians; 2001.
4. Perry S. Transitions in care: studying safety in emergency department signovers. Focus Patient Safety. 2004;7:1-3. Available at: http://www.npsf.org/download/Focus2004Vol7No2.pdf. Accessed February 3, 2006
5. Matthews AL, Harvey CM, Schuster RJ, Durso FT. Emergency physician to admitting physician handovers: an exploratory study. Proceedings of the Human Factors and Ergonomics Society 46th Annual Meeting; September 29-October 4, 2002; Baltimore, MD.
6. Beach C, Croskerry P, Shapiro M, for the Center for Safety in Emergency Care. Profiles in patient safety: emergency care transitions. Acad Emerg Med. 2003;10:364-367. [ go to PubMed ]
7. Propp DA. Emergency care transitions. Acad Emerg Med. 2003;10:1143; author reply 1143-1144. [ go to PubMed ]
8. Bierly PE, Spender JC. Culture and high reliability organizations: the case of the nuclear submarine. J Manage. Winter 1995.
9. Xiao Y, Moss J. Practices of high reliability teams: observations in trauma resuscitation. Proceedings of the Human Factors and Ergonomics Society 44th Annual Meeting; July 30-August 4, 2000; San Diego, CA.
10. Patterson ES, Roth EM, Woods DD, Chow R, Gomes JO. Handoff strategies in settings with high consequences for failure: lessons for health care operations. Intl J Qual Health Care. 2004;16:125-132. [ go to PubMed ]
11. Rose BD, George JN. Treatment of thrombotic thrombocytopenic purpura-hemolytic uremic syndrome in adults. UpToDate Online 13.3. Available at: http://www.uptodate.com/index.asp. Accessed November 20, 2005.
12. Terrell DR, Williams LA, Vesely SK, Lammle B, Hovinga JA, George JN. The incidence of thrombotic thrombocytopenic purpura-hemolytic uremic syndrome: all patients, idiopathic patients, and patients with severe ADAMTS-13 deficiency. J Thromb Haemost. 2005;3:1432-1436. [ go to PubMed ]
13. Croskerry P. The importance of cognitive errors in diagnosis and strategies to minimize them. Acad Med. 2003;78:775-780. [ go to PubMed ]
14. Cook RI. A brief look at gaps in the continuity of care and how practitioners compensate for them. Cognitive Technologies Laboratory Web site. Available at: http://www.ctlab.org/documents/BriefLookAtGapsRevF.pdf. Accessed December 28, 2005.
15. Reason JT. Managing the Risks of Organizational Accidents. Aldershot, Hampshire, England: Ashgate; 1997:135.
16. Van Eaton EG, Horvath KD, Lober WB, Pellegrini CA. Organizing the transfer of patient care information: the development of a computerized resident sign-out system. Surgery. 2004;136:14-15. [ go to PubMed ]
17. Patterson ES, Woods DD, Cook RI, Render ML.
Collaborative cross-checking to enhance resilience. Proceedings of
the Human Factors and Ergonomic Society 49th Annual Meeting;
September 26-30, 2005; Orlando, FL.