Approach to Improving Safety
Setting of Care
An infant was born with sluggish respirations. During labor the infant's mother had received meperidine [Demerol, a pain medication], a narcotic with a half-life of 2.5-4.0 hours in adults and 12-39 hours in neonates. The physician started resuscitation and ordered naloxone [an opiate antagonist]. Shortly after administration of the medication, the infant's condition began to deteriorate further.
Prompted by the proximity of the deterioration to the administration of the naloxone the physician checked the packaging of the drug. The syringe had inadvertently been filled with Lanoxin [digoxin, a cardiac medication] instead of naloxone. The packages of both drugs, made by the same manufacturer, were almost identical. ECG revealed bi-directional ventricular tachycardia, consistent with digoxin toxicity.
Approximately 1 hour later the infant died. A
post-mortem digoxin level was 17 ng/ml (therapeutic range 0.8 to 2
What does this tragic case tell us, and more importantly, what can we learn from it? Following symptoms of meperidine toxicity, a doctor prescribed naloxone, an appropriate antidote. However, digoxin, a cardiac glycoside, was administered instead, and the baby died of digoxin toxicity.
On face value, it is almost impossible to imagine how this could have occurred. However, when one realizes that the brand name of digoxin is Lanoxin®, and that in this case both drugs were made by the same manufacturer and presented in similar packaging, it may be slightly easier to understand how the error happened. Both are drugs that would be stocked on a typical neonatal unit, and the doses used are similar. The intravenous dose of naloxone recommended for use in newborn babies is 10 mcg/kg every 2-3 minutes, and a digoxin loading dose is 10 mcg/kg to 17.5 mcg/kg depending on the age of the baby and whether he was full term. If whoever administered the drug confused naloxone with Lanoxin, either because she thought this was the brand name for naloxone, or simply because the name on the packaging was misread, there would be few cues to suggest that the wrong drug had been selected until after it had been administered.
How Often do Errors of This Nature Occur?
Observational studies in hospitals have reported administration errors in 3%-8% of doses in the United Kingdom (UK) (1,2), and in 0.6%-14.6% in the United States (US) (3), excluding wrong time errors. Methods and definitions vary, so it is difficult to compare studies directly, but it is clear that administration errors are not uncommon. Wrong drug errors are typically a smaller proportion of these, occurring in about 0.2% of all doses given.(4) Fortunately, most administration errors do not result in outcomes as tragic as in this case.
Causes of Wrong Drug Administration
Little is known about the true causes of administration errors like this one. However, studies have examined the causes of prescribing errors (5) and other types of medical error (6) using psychological models of human error. In addition to exploring why the error occurred at the level of individuals at the "sharp end," such studies also aim to identify the organizational and environmental factors that make errors possible. The objective of identifying these latter factors is that actions here are likely to have the most impact in preventing future errors. While there have been no in-depth studies of the causes of administration errors, similar packaging, and look-alike or sound-alike names (7) are considered important contributing factors.
There are various approaches to investigation. For example, the Clinical Risk Unit in the UK has developed a protocol for the investigation and analysis of adverse events in clinical practice (8) based on a well-known model of human error.(9) This method requires training to be used effectively, but involves interviewing staff involved to identify care management problems and why each occurred. The investigation aims to identify contributing factors at the level of the institution, organization and management, work environment, team, individuals, task, and the patient.(10) Such an approach would likely be useful in investigating this incident, and would highlight more wide-reaching issues than the simple fact that someone gave the wrong drug.
What Systems Might Prevent This Type of Error?
Defenses could be built in at many stages, depending on the outcome of the investigation. These may include purchasing products from different manufacturers, reviewing how look-alike and sound-alike products are stored, and highlighting drugs that could potentially be confused. Another approach may involve improving systems of communication between prescribers and nursing staff. It has been proposed that all medication orders should include the indication for which the drug is being used, as this may prevent some errors (though probably not in an urgent situation such as this one). In this case, had the physician indicated that the baby was being treated for opiate toxicity, the nurse might have hesitated to draw up a cardiac glycoside.
More global changes are needed to increase the risk assessment carried out when naming medicinal products, to minimize the risks of mix-ups occurring. The Institute for Safe Medication Practices provides proprietary reviews of proposed names, trademarks, packaging, and labeling for pharmaceutical industry clients in an attempt to prevent potentially confusing names. Unfortunately, such reviews are not yet compulsory.
Should This Error be Disclosed to the Parents?
Experts in medical ethics and professional
organizations all endorse disclosure of errors. A recently
published exploration of both patients' and physicians'
attitudes toward medical error disclosure, which used scenarios
with errors specifically involving medication, found that patients
wanted disclosure of all harmful errors and sought information
about what happened, why, and how similar errors will be prevented
in the future.(11)
There was less agreement over whether near misses should be
disclosed. Physicians agreed that harmful errors such as this one
should be disclosed, but the report found that they "chose their
words carefully" when telling patients about errors. Although
physicians disclosed the adverse event, they often avoided stating
that an error had occurred, why it happened, and how it would be
prevented. These findings back up those of other studies (12)
from the viewpoints of both doctors and patients.
- Errors in drug administration occur relatively often.
- Packaging and look-alike/sound-alike drugs are thought to be important contributing factors.
- Risk assess look-alike and sound-alike products and consider how they are stored.
- Improving communication between medical and nursing staff may help to prevent errors.
- Patients and their families want disclosure of errors, along with information on how similar errors will be prevented in the future.
Bryony Dean Franklin, PhD
Director, Academic Pharmacy Unit
Hammersmith Hospitals NHS Trust and The School of Pharmacy, University of London.
1. Dean BS, Allan EL, Barber ND, Barker KN. Comparison of medication errors in an American and a British hospital. Am J Health Syst Pharm. 1995;52:2543-9.[ go to PubMed ]
2. Taxis K, Dean BS, Barber ND. Hospital drug distribution systems in the UK and Germany—a study of medication errors. Pharm World Sci. 1999;21:25-31.[ go to PubMed ]
3. Allan EL, Barker KN. Fundamentals of medication error research. Am J Hosp Pharm. 1990;47:555-571.[ go to PubMed ]
4. Dean BS, Barber ND. The effects of a patients' own drugs scheme on the incidence and severity of medication administration errors. Int J Pharm Pract. 2000;8:209-16.
5. Dean B, Schachter M, Vincent C, Barber N. Causes of prescribing errors in hospital inpatients: a prospective study. Lancet. 2002;359:1373-8.[ go to PubMed ]
6. Stanhope N, Vincent C, Taylor-Adams SE, O'Connor AM, Beard RW. Applying human factors methods to clinical risk management in obstetrics. Br J Obstet Gynaecol. 1997;104:1225-32.[ go to PubMed ]
7. Lambert BL. Predicting look-alike and sound-alike medication errors. Am J Health Syst Pharm. 1997;54:1161-1171.[ go to PubMed ]
8. Vincent C, Taylor-Adams SE, Chapman EJ, Hewett, et al. How to investigate and analyse clinical incidents: clinical risk unit and association of litigation and risk management protocol. BMJ. 2000;320:777-81.[ go to PubMed ]
9. Reason JT. Human Error. New York: Cambridge University Press; 1990.
10. Vincent C. Understanding and responding to adverse events. N Engl J Med. 2003;348:1051-6.[ go to PubMed ]
11. Gallagher T, Waterman AD, Ebers AG, Fraser VJ, Levinson W. Patients' and physicians' attitudes regarding the disclosure of medical errors. JAMA. 2003;289:1001-7.[ go to PubMed ]
12. Hingorani M, Wong T, Vafidis G. Patients' and doctors' attitudes to amount of information given after unintended injuries during treatment: cross-sectional, questionnaire survey. BMJ. 1999;318:640-1.[ go to PubMed ]