Communication failure in the intensive care unit--learning from a near miss.
The patient was admitted to the neurology service of a tertiary referral hospital with suspected Guillain-Barre syndrome. On day two she was transferred to the intensive care unit (ICU) for closer observation due to concerns regarding her respiratory function. The neurology service remained her primary bed-card unit. In the ICU the patient was commenced on intravenous human immunoglobulin at a dose of 30 g daily, for five days, by instruction of the neurology service to be prescribed by ICU medical staff. Review of her chart by the neurology service on the following morning indicated she had been given 150 g of immunoglobulin over a six-hour period - five times the intended daily dose. The neurology and ICU teams informed the patient, documented the incident in the medical record and completed an incident report, which was subsequently discussed at the weekly ICU morbidity and mortality meeting and at the hospital's quality and safety committee.
Immunoglobulin therapy was ceased and the haematology service was consulted regarding the risk of coagulopathy. The patient's condition did not deteriorate. She was discharged from the ICU to the ward on day five where she continued to improve. She was discharged home on day eight with the expectation of a full recovery. Having made a full recovery, the patient provided full verbal consent for her de-identified case to be discussed in a professional publication.
This case provides the opportunity to learn from a 'near miss' (1). The case demonstrates that, despite our best efforts, a patient can be overdosed on intravenous medication in the ICU of a major Australian teaching hospital. In this case, the error was not noticed for roughly seven hours. It is fortunate, albeit irrelevant, that an overdose of intravenous human immunoglobulin does not usually carry serious consequences.
There is clear evidence demonstrating the relationship between multiple system errors, breakdowns in communication and adverse patient events (2). This case highlights that relationship.
It is clear from the outset that this incident was not one person's fault. Based on discussion with the staff involved, it appears that from the time the patient was admitted to the neurology service until the time it was realised that she had been overdosed on immunoglobulin, there were at least 11 specific failures in communication. Table 1 documents these failures and classifies the communication errors (3).
The analysis in Table 1 indicates a diffusion of responsibility (5) by the treating teams, in that neither team took adequate responsibility for documenting and implementing the intended treatment plan. The key factors that led to the patient being overdosed were breakdowns in written and verbal communication and junior medical staff who did not fully understand their role.
The reasons why the neurology service did not document the entire plan in more detail before their patient was transferred to ICU, or why the ICU team did not question ambiguous notes is not clear. Perhaps the plan seemed 'too simple' to warrant detailed documentation? Perhaps junior staff did not feel that they could ask an obvious question? Perhaps no-one noticed that the plan had not been written down, due to the inevitably busy workload of staff in a tertiary hospital? There is no clear explanation.
What is clear, however, is that junior medical staff from both the neurology service and the ICU were charged with implementing a plan that they seemingly did not fully understand. That plan was, therefore, not clearly communicated to the ICU nursing staff. That successive breakdown in communication and subsequent propagation of error resulted in the patient being given at least three times the intended dose of intravenous immunoglobulin.
Situations like this are common in modern healthcare(6). Indeed, mistakes are an inevitable part of practising medicine in a dynamic and time-pressured environment (7). The important question now is, how can this be prevented from happening again?
In general, the medical profession is poor at ensuring that all members of a treating team know and understand not only the what of a plan but also the how to (8,9). One solution that might help in preventing an event such as this from occurring in the future is the adoption of a more uniform approach to note-taking and medical record keeping, in conjunction with a robust and formal orientation program for junior medical staff (10). Standardised frameworks for documentation of information are built into the daily routine of other industries, most notably the military, aviation industry and emergency services (11). However, the use of such frameworks is an uncommon practice in medicine. Historically, the rigour applied to medical note-taking and verbal handover has been left to the individual practitioner, rather than being embedded as a formalised process within the structures of the workplace (11).
The communication failures outlined in Table 1 can be summarised as follows: the management plan was not clearly documented. That initial lack of clarity was not questioned in the early phases of its implementation. As such, errors made in implementing the ambiguous plan were carried over by multiple personnel.
Communication failures such as these are more common when documentation and clinical handover is poor or when junior staff members do not understand their seniors' intent (12). There are several recognised formats designed to reduce communication breakdown.
SBAR is an acronym that originates in the US Navy submarine service. It stands for situation, background, assessment and recommendation. It was developed to allow easy, efficient and safe communication between junior crew and senior officers during extended day/night operations. It has since been adopted by several health services in the US as a way for junior nursing and medical staff to communicate clinical information to on-call consultants (13,14). In this case, an adapted acronym, SBAP (situation, background, assessment, plan), could be utilised by medical staff to convey clinical information and the medical plan in a written format. This may have resulted in a more clearly documented plan by specifically directing the assessing doctor to write one down.
However, while acronyms can assist trained staff members to minimise error and mitigate risk, they would not have solved this problem. The crux of this communication failure is that junior medical staff members were not adequately trained to make complex decisions or implement unusual plans in an intensive care environment. At the time of this incident, a formal orientation program for residents commencing terms in the ICU did not exist. The first day of a new term served as a form of ad hoc orientation and was, therefore, variable in its quality and comprehensiveness. ICUs are unlike other areas of medical practice. While residents from the medical or surgical ward can transport general principles of patient care into the ICU, the specifics of daily practice and the demands of the ICU patient are, by definition, more intense. As a result the consequences of error are more serious. The likelihood of error occurring is heightened when new staff members are not trained to perform their new job (13).
Since this incident, a formal orientation program is conducted for residents in the first week of a new term and an orientation pack with common information and tasks is provided to them during the orientation program. This orientation includes specific instruction on appropriate prescribing and fluid ordering in the ICU. The overarching theme of this orientation program is one of seeking help and clarification of instructions early.
In summary, this case highlights the need to remedy a significant failure in the lines of communication both within and between treating teams in order to avoid similar events of this nature occurring in the future. It provides a unique opportunity to revisit concepts of patient safety and explore new ways to further increase the quality of healthcare provided by tertiary ICUs. The solutions may be applicable to other clinical environments.
Furthermore, this case demonstrates the role of regular and open clinical audit in ICUs to highlight misunderstandings and communication deficits early. This case may demonstrate the need to adequately orient new clinical staff to the specifics of their new job, particularly in an intensive care environment.
Successive failures in communication both between and within treating teams led to the overdose of this patient. More research needs to be conducted into effective ways of improving written and verbal communication, within an intensive care or other clinical setting, if such failures are to be avoided in the future.
This paper would not have been possible without the ongoing and insightful assistance of Dr Julia Harrison.
(1.) Nashef S. What is a near miss? Lancet 2003; 361:180-181.
(2.) Arora V, Johnson J, Lovinger D, Humphrey HJ, Meltzer DO. Communication failures in patient sign-out and suggestions for improvement: a critical incident analysis. Qual Saf Health Care 2005; 14:401-407.
(3.) Gibson WH, Megaw ED, Young MS, Lowe E. A taxonomy of human communication errors and application to railway track maintenance. Cogn Tech Work 2006; 8:57-66.
(4.) Medical record committee, Southern Health. Accepted abbreviations listing and policy. September 1995.
(5.) Lingard L, Espin S, Rubin B, Whyte S, Colmenares M, Baker GR et al. Getting teams to talk: development and pilot implementation of a checklist to promote interprofessional communication in the OR. Qual Saf Health Care 2005; 14:340-346.
(6.) West E. Organisational sources of safety and danger: sociological contributions to the study of adverse events. Qual Health Care 2000; 9:120-126.
(7.) Tucker AL, Edmonson AC. Why hospitals don't learn from failures: organisational and psychological dynamics that inhibit system change. California Management Review 2003; 45:5572.
(8.) Witman AB, Park DM, Hardin SB. How do patients want physicians to handle mistakes? A survey of medicine patients in an academic setting. Arch Intern Med 1996; 156:2565-2569 .
(9.) Mukherjee S. A precarious exchange. N Engl J Med 2004; 351:1822-1824.
(10.) Lingard L, Espin S, Whyte S, Regehr G, Baker GR, Reznick R et al. Communication failures in the operating room: an observational classification of recurrent types and effects. Qual Saf Health Care 2004; 13:330-334.
(11.) Helmreich RL. On error management: lessons from aviation. BMJ 2000; 320:781-785.
(12.) Leonard M, Graham S, Bonacum D. The human factor: the critical importance of effective teamwork and communication in providing safe care. Qual Saf Health Care 2004; 13(Suppl 1):i85-90.
(13.) Spath PL, ed. Error Reduction in Health Care: A Systems Approach to Improving Patient Safety. Jossey-Bass, California, USA 2000.
(14.) US Medicine. DoD Medical Teamwork, Safety Stressed In New Rollout Effort. From http://www.usmedicine.com/article. cfm?articleID=1598&issueID=101 Accessed December 2007.
S. A. HENDEL *, B. T. FLANAGAN ([dagger])
Southern Health Simulation Centre, Melbourne, Victoria, Australia
* M.B., B.S. (Hons.), Anaesthetics Registrar, Alfred Hospital, Prahran.
([dagger]) M.B., B.S., F.A.N.Z.C.A., Associate Professor in Patient Safety Education,
Monash University and Director of the Southern Health Simulation Centre.
Address for reprints: Dr S. Hendel, Anaesthetics Registrar, Alfred Hospital, PO Box 315, Prahran, Vic. 3181.
Accepted for publication on April 6, 2009.
Table 1 Communication failures Day/ Points of What is wrong with this External error time failure in communication? mode/ communication classification of error (3) Day 1, By late 'IV to describe what was Too little 1500 afternoon on actually a specific 10 g in information/ hours day 1, a 200 ml formulation of content error. diagnosis of intravenous human Guillain-Barre immunoglobulin (marketed as syndrome was Octagam [TM]). Nowhere in made and the the medical record were decision was these unfamiliar and taken to treat non-standardised with "IV Ig" abbreviations explained. (verbatim, from the medical record). Day 2, ICU staff At no stage do either the Too little 900 became involved medical or nursing notes information/ hours in the document the specifics of content error. patient's the the treatment plan. In this specifics of instance, the ICU notes are the assessment, ambiguous (i.e. does this the ICU mean 5 days of 'Intragam' registrar noted or 5 doses given over an "management--5 unspecified timeframe?). X Intragam". Day 2, The nursing The nursing notes are Omission/too 1100 notes from the similarly ambiguous and little hours neurology potentially misleading. By information/ service on day documenting only partial content error. 2 state "... information and omitting For transfer to key information (such as ICU ... To the frequency of commence administration and duration Intragam when of treatment), the medical arrives (for 5 record ceases to be a safe days--30 g)". means of inter-professional communication. Day 2, When it was APP is not a widely Wrong action 1100 decided to accepted acronym on drug right little hours start the charts (4). APP was most information/ patient on likely intended to mean 'as content is intravenous per protocol', as is immunoglobulin sometimes the case in the (in the ICU), ICU, but could mean 'as per the neurology plan'. The use of acronyms resident is acceptable if it can be documented five assured (not assumed) that separate orders all involved in their use for "IVIG 30 g" have a shared understanding on the ward of their meaning. drug chart, all dated on the same date and charted that it be administered "APP". Day 2, The ICU The neurology service was Repeated 1200 resident who the primary bed-card unit error/ hours admitted the but they did not document cognitive patient to the the treatment plan in the elements of ICU directly medical record. The ICU human error. transcribed the team began treatment of the IVIG order to patient according to a plan the ICU drug that was not clearly chart from the documented. The result is a neurology medical a treatment plan resident's that is ambiguous and open record and ward to order on the drug chart. The interpretation. While order was for ambiguity exists for all five separate staff, the purpose of 30 g IVIG documenting a management doses, but all plan is to reduce ambiguity dated on the for those charged with same date. The implementation. administration schedule was documented as "APP". Day 2, The ICU This entry and the Repeated 1200 admission notes following entry demonstrate error/ hours state in the the diffusion of cognitive plan "IV Ig as responsibility (5) between elements of per neurology the ICU and the neurology human error. service". service. The implication of these entries is that both teams thought that the other team was in control of implementing the plan. Day 2, On the See above. See above. 1645 afternoon of hours day 2, the neurologist and the neurology registrar saw the patient. Their documented plan states "ventilatory function stable. Observe for 24/24. If remains stable could return to ward. Continue IV Ig". Day 2, The evening ICU This suggests that the ICU Wrong action 1800 round noted staff considered their role right object/ hours that the as one of monitoring a too little patient was neurology patient who would information/ "stable" and be returned to the cognitive and that the plan neurology ward as soon as content error. was to they were able. "continue IV Ig and discharge to ward tomorrow if patient remains stable". Day 2, The nursing This entry suggests that Wrong action 1930 notes, the nursing staff right object/ hours following the responsible for actually cognitive evening ICU administering the error. round state immunoglobulin did not that "all understand the treatment general care plan. attended. Intragam 30 g X 2 given so far". Day 3, The night ICU This entry suggests that Wrong action 400 round notes the night ICU medical team right object/ hours state that the did not understand the too little patient "is on treatment plan, though it information/ IV Intragam". is unclear if the night the cognitive medical staff observed and content exactly how much error. immunoglobulin the patient had been given. There were seemingly no documented concerns with the progress of the patient's treatment, despite had been given at least twice the intended daily amount. Day 3, The neurology When blood products are Wrong action 900 service administered in this right object/ hours recognised the institution, they must be too little dosing error. recorded on both the drug information/ On closer chart and the fluid balance cognitive and scrutiny, the chart because they affect content error. ICU drug chart volume status. The amount shows 150 g of signed for, as being Intragam as administered on the drug being signed chart should equal the and given. The amount recorded on the fluid balance fluid balance chart. The chart documents disparity between the drug 70 g of chart, the fluid balance Intragam as chart and the blood bank being record meant that it was administered not possible for the and the blood neurology service or the bank records ICU team to be sure of the state that 90 g exact amount of of Octagam [TM] immunoglobulin that had was released to actually been administered. the ICU on day There were no surplus 2. bottles of Octagam [TM], so it is reasonable to assume that the patient received the 90 g released by the blood bank and that errors were made in documenting the dose administered on the ICU drug and fluid charts.
|Printer friendly Cite/link Email Feedback|
|Author:||Hendel, S.A.; Flanagan, B.T.|
|Publication:||Anaesthesia and Intensive Care|
|Date:||Sep 1, 2009|
|Previous Article:||William TG Morton's early ether inhalers: a tale of three inhalers and their inscriptions.|
|Next Article:||The influence of gender and experience on intubation ability and technique: a manikin study.|