Moved Too Soon
A 67-year-old man was admitted to a general hospital ward after undergoing a laminectomy. Two hours after arriving, while the patient was still groggy from anesthesia, a nurse entered the room and stated that it was time to administer his clonazepam. As the patient began to take the medicine, his daughter (who happened to be a nurse) stated that she didn't think he should be receiving clonazepam and asked the nurse to double check prior to administration. The nurse returned after checking and asked, "Aren't you Mr. X?" The patient said, "No, I am Mr. J."
It turned out that, due to a bed shortage, Mr. J was to be moved down the hall, and Mr. X, a seizure patient scheduled to be transferred out of the neuro-ICU that afternoon, was to be moved into that room. The room change was made on the hospital's computer before the patients were physically moved. Thus, when the nurse checked the computer, it showed that Mr. X was in that room and due for his clonazepam.
This near miss provides a classic example of patient misidentification, made worse by the fact that a sophisticated computer system expected to prevent such errors likely contributed to its occurrence.
The proper identification of patients during hospitalization continues to challenge hospitals, evidenced by JCAHO making improved identification a patient safety standard in the last several years.(1) A recent analysis of a voluntary near miss and event reporting system shared by 54 neonatal intensive care units in the United States found that 11% of cases involved patient misidentification (2), and wrong patient incidents accounted for 4% of all medication errors.(3) During a recent 3-year period, the Veterans Affairs National Center for Patient Safety received more than 100 patient misidentification root cause analyses, 22% of which involved cases of medication administration.(4)
Tempering Expectations for Information Technology
Increasingly, information technology is seen as a solution to many patient safety issues facing our health care system.(5) While research on the benefits of CPOE and clinical decision support systems give reason to be hopeful (6,7), qualitative studies have found that such systems often precipitate new errors.(8) Not only can the intrinsic design of hardware and software promote error, but so too can the environments within which they are deployed. As this case demonstrates, expectations about the benefits of information technology should be tempered by a recognition that human processes also play a critical role in determining whether a particular technological "solution" will succeed or fail. If human factors are overlooked in planning, information technology may automate a bad practice or accelerate the occurrence of errors.
One in a Series of Errors
In this example, as in most such cases, a series of errors occurred in the process of care which, had it not been for a savvy family member, surely would have resulted in the wrong medication being administered to this patient.
The first and most obvious error was that the patient's location was changed in the computer system before the patient had actually arrived at his new destination. Next, the nurse failed to correctly identify the patient when she began the medication administration process. Upon first entering the room, the nurse did not ask his name, nor examine his wristband in search of two unique identifiers. Upon reentering, the nurse asked a closed-ended question ("Aren't you Mr. X?")—a passive method of identification that is especially problematic when dealing with young children, adults with dementia or delirium, or other language or communication barriers. A better approach would have been to ask an active question such as "What is your name?"(9)
Bar Coding and Radiofrequency Identification
Ubiquitous in both supermarket checkout aisles and in hospital materials management departments, bar coding is finding new applications in patient care.(Figure 1) A bar code medication administration process, in which nurses scan a medication, the patient for whom it is intended, and then themselves, has obvious advantages over one that relies on human vigilance to reconcile data on a medication label and a patient wristband. Bar coding systems offer other advantages, including being able to evaluate the appropriateness of the timing of a given dose and eliminating the documentation required to create a medication administration record. As in the case of CPOE, the number of hospitals that have successfully implemented bar coding into the medication administration process is small in comparison to those that are planning or in the process of implementation. It is now widely believed that Radio Frequency Identification (RFID) will one day replace bar coding as a method to safeguard the medication administration process. The primary advantage of RFID is that the combination of a radio transmitter and receiver eliminates the cumbersome process of scanning. RFID tags can be read at distances up to several feet, while in motion, in any orientation, and through intervening objects.
Could this error have occurred if the hospital had a bar code medication administration system? While evidence supporting the benefits of such systems remains limited (10), bar coding has been advocated by many national organizations and has received further support by a recent FDA decision to require bar codes on most prescription medications within the next two years.(11) Unfortunately, but perhaps unsurprisingly, reports from early adopters have identified a number of unintended effects of bar coding that have led to new kinds of medication errors.(12) Moreover, wristbands are not foolproof.(13) Thus, while bar coding may eliminate many wrong patient or wrong medication errors, the reality is that nurses will continue to play an important role in correctly identifying patients.
Operationalizing Patient Identification Standards
While bar coding could be considered in the intermediate term, several short-term approaches can reduce the occurrence of adverse events from misidentification. JCAHO recommends using at least two patient identifiers when taking blood samples or administering medications or blood products. Education regarding the "5 rights" in medication administration (Right patient, Right time and frequency of administration, Right dose, Right route of administration, Right drug) is a fundamental element of nursing training. At our institution, these lessons are reinforced by highlighting actual cases reported to our own near miss and event reporting system. Measurement and feedback is another effective method of improving performance, and we incorporate direct observation of medication administration into the annual review process for nurses. Moreover, aggregated data are presented and discussed at our monthly Nursing Practice committee meeting.
Whether dealing with a computer system used for tracking patient location, or a simple dry erase board, the hospital should also redesign the current process for initiating and activating location changes, so that these changes are made only when the patient physically arrives in the new location. While not entirely clear from the case presentation, it's possible that the tracking system allowed a patient location change to be made in an already occupied bed. If this was so, the hospital should also work with its software vendor to develop a forcing function in its computer system to prevent such an occurrence. Such a feature (Figure 2) would send a warning to the user and prohibit such actions from being completed.
Peter Lindenauer, MD, MSc Division of Healthcare Quality Baystate Health System Assistant Professor of Medicine Tufts University School of Medicine
1. 2005 National patient safety goals. Joint Commission on Accreditation of Healthcare Organizations Web Site. Available at: [ go to related site ]. Accessed September 16, 2004.
2. Suresh G, Horbar JD, Plsek P, et al. Voluntary anonymous reporting of medical errors for neonatal intensive care. Pediatrics. 2004;113:1609-18.[ go to PubMed ]
3. Bates DW, Boyle DL, Vander Vliet MB, Schneider J, Leape L. Relationship between medication errors and adverse drug events. J Gen Intern Med. 1995;10:199-205.[ go to PubMed ]
4. Mannos D. NCPS patient misidentification study: a summary of root cause analyses. National Center for Patient Safety TIPS. Available at: [ go to related site ]. Accessed September 16, 2004.
5. Kohn LT, Corrigan JM, Donaldson MS. To err is human: building a safer health system. Washington, D.C.: National Academy Press; 2003.[ go to related site ].
6. Dexter PR, Perkins S, Overhage JM, Maharry K, Kohler RB, McDonald CJ. A computerized reminder system to increase the use of preventive care for hospitalized patients. N Engl J Med. 2001;345:965-70.[ go to PubMed ]
7. Bates DW, Leape LL, Cullen DJ, et al. Effect of computerized physician order entry and a team intervention on prevention of serious medication errors. JAMA. 1998;280:1311-6.[ go to PubMed ]
8. Ash JS, Berg M, Coiera E. Some unintended consequences of information technology in health care: the nature of patient care information system-related errors. J Am Med Inform Assoc. 2004;11:104-12.[ go to PubMed ]
9. Campbell D. Listen to the family. AHRQ WebM&M [serial online]. June 2004. Available at: [ go to related commentary ]. Accessed October 14, 2004.
10. Wald H, Shojania KG. Strategies to avoid wrong-site surgery. In: Shojania KG, Duncan BW, McDonald KM, Wachter RM, eds. Making health care safer: a critical analysis of patient safety practices. Rockville, MD: Agency for Healthcare Research and Quality; 2001: 494-500. AHRQ publication 01-E058. Evidence report/technology assessment. no. 43. Available at: [ go to related site ]. Accessed September 16, 2004.
11. Food and Drug Administration. Bar code label requirements for human drug products and blood. Fed Register. 2003;68:12500-12534.
12. Patterson ES, Cook RI, Render ML. Improving patient safety by identifying side effects from introducing bar coding in medication administration. J Am Med Inform Assoc. 2002;9:540-53.[ go to PubMed ]
13. Rosenthal MM. Check the wristband. AHRQ WebM&M [serial online]. July 2003. Available at: [ go to related commentary ]. Accessed September 23, 2004.
Figure 1. A Health Care Bar Coding System in Action.
Figure 2. Forcing Function Prohibiting Transfer to an Occupied Bed.