Case Objectives

• Describe the just culture approach to investigating errors in health care.
• Analyze system contributions to errors in care.
• Identify best sources of information for designing a response to an error.
• Distinguish accountability for failure and learning from failure.

The Case

An infant was born prematurely at 30 weeks weighing only 1.8 kg. In the neonatal intensive care unit, he was started on total parenteral nutrition (TPN) with Premasol amino acid solution at 3 g/kg/d and dextrose 12.5%, 5 mg/kg/min. After being maintained using those solutions for the first 2 days after delivery, the care team added lipids on day 3. This was ordered as lipid emulsion 20% at a rate of 0.19 mL/hr.

The neonatal intensive care unit had frequent orders for this treatment and kept a stock of lipid emulsion on site. This practice avoided the delay between ordering, sending the order to the pharmacy, and waiting for the pharmacy to dispense the new TPN solution.

Within 4 hours of beginning the lipid emulsion administration through the TPN line using a smart pump, the infant's condition worsened. He showed signs of respiratory distress, pulmonary hypertension, coagulopathy, and liver failure. Soon after, the infant suffered a cardiac arrest and died.

As the symptoms displayed by this premature infant suggested lipid overload, the dose and rate of administration of the lipid formulation were assessed. Assessment revealed that the pump was set to deliver 19.0 mL/hr. In the process of calculating the dose with the concentration of lipid emulsion available on the unit, the RN had erroneously set the pump to deliver 100 times the ordered dose of 0.19 mL/hr. Upon discovery of the error, the nurse involved was fired by the hospital and her license was revoked. The sequence of events and underlying reasons for the error were not investigated further.

The Commentary

This case is severe and shocking, as is the outcome: the death of an infant. In the aftermath of such an event, health care providers and organizations must search for the most appropriate response. Here are a series of questions and answers designed to help in choosing the best and most useful response.

Is the just culture framework helpful for how an organization should respond?

Yes. Firing the nurse and having her license revoked, while doing nothing to investigate the system issues surrounding the error (e.g., technology and interface ergonomics, lipid administration for neonates, dose calculation routines, clinical response to signs of trouble), could be seen as deeply unjust. Unjust not only for the nurse, but also for the family of the deceased: what confidence can they have that this won't happen again; that another patient will not suffer from a similar mishap in the future? A just culture balances accountability with learning.(1) What we see here is only (one very narrow form of) accountability, and no learning. A lost opportunity, a wasted death.

How can investigators determine whether this is an error, or at-risk or reckless behavior?

Determining whether this is an honest error or at-risk/reckless behavior is a very difficult judgment, and the outcome depends more on who is involved in making it than on the behavior itself. So my advice: First, ask the nurse. Then ask other nurses when you explain the case to them. Finally, look at how often (and under what circumstances) this or something similar has happened, either at your own facility or elsewhere. Whether it is an honest error or something more sinister is not easily determined by looking at the act and its circumstances alone—it is a judgment call on the part of the people assessing somebody's performance after the fact, much more than it is a ready-formed or inherent feature of that performance. What can look reckless from one perspective (e.g., the hospital's lawyer or administrator or doctor) can look quite honest, normal, and understandable from another (e.g., the nurses who work in the organization's messy environment every day). So be sure to involve multiple viewpoints when determining whether this is an honest error or something worse, and fairly balance them. Don't just take anybody's word for it. In general, your concern should be why good nurses make mistakes, not why bad nurses do so, because very few nurses come to work to do a bad job.

How can an investigator determine the system contributions?

Independent of what you believe about the error, study the system and the organization that helped bring forth this error or behavior. Do three things to determine why it made sense for the nurse to do what he or she did. First, figure out what multiple (and often conflicting) goals influenced the performance. How many other patients were there at the time; how long had the nurse been on shift? Were there time pressures, organizational or managerial expectations, or protocol or procedures that would get in the way of getting the job done? Second, determine the (clinical) knowledge of the person involved: did he or she have sufficient training for the task at hand; was there a mismatch in the mental model about the technology, the patient, the drug, the procedure? Third, what was the nurse focused on or looking at during preparation and administration; was his or her attention directed to the task or distributed across multiple tasks?

What is the recommended response in case of an error or at-risk/reckless behavior?

First, ask the nurse. What does the person involved believe should be done? This is one of the best starting points for a just culture. People closest to the mishap often feel responsible and eager to help with suggestions for improvement, and their immediate experience can provide the best evidence base for useful intervention that you're ever going to get. In any case, be sure to involve peers in any judgments of what should be done. That is the only way to sustain a just culture, one in which remaining colleagues can feel free to report their own mistakes without fear of undue consequences.

How do hospitals manage these kinds of errors in general?

Not very impressively. The simultaneous belief in individual strength and brittleness is pervasive in health care: Safety lies in the hands through which care ultimately flows to the patient, not in the system that surrounds those hands. When things go well, health care culture tends to celebrate "good doctoring" (2,3): acts by competent people who succeeded despite the organization and its complexity. When things go wrong, health care culture often zeroes in on people at the "sharp end" who, for once, failed to hold that complex, pressurized patchwork together—rather than inquiring about the systemic sources behind the production of all that complexity. The response of the hospital in this case is, sadly, all too typical. It may have managed the organization's risk, but it is probably seen as very unjust by all other major stakeholders: the nurse involved, her colleagues, and even the patient's family.

This, in the end, is how a hospital can balance accountability and learning. Accountability is not just holding somebody responsible by meting out punishment. It is about telling stories of what happened. And by telling such stories, by sharing them among colleagues, similar unnecessary deaths can be prevented. It's accountability and learning at the same time.

Take-Home Points

• A just culture balances learning from failure with accountability for failure.
• In this lipid overdose case, the response can easily be seen as unjust: only (one narrow form of) accountability, and no learning.
• The response, however, is probably typical: hospitals often want to manage their own (liability) risk first, with concern for their just culture coming later (if at all). And health care culture generally overestimates the role of the individual actor in bringing about clinical success or failure, downplaying the contribution of the system.
• Be sure to involve multiple viewpoints in your determination of whether a given error represents an honest mistake or something worse. Fairly balance these viewpoints, don't just take anybody's word for it.
• Accountability doesn't have to be all about meting out punishment, it can be about having people tell their accounts, their stories, from which others in the hospital can learn and improve.

Sidney W.A. Dekker, PhD Professor and Director

Leonardo da Vinci Laboratory for Complexity and Systems Thinking

Department of System Safety

Lund University, Sweden

Faculty Disclosure: Dr. Dekker 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, the commentary does not include information regarding investigational or off-label use of pharmaceutical products or medical devices.

References

1. Dekker SWA. Just Culture: Balancing Safety and Accountability. Farnham, UK: Ashgate Publishing Co.; 2007. ISBN: 0754672670.

2. Gawande A. Complications: A Surgeon's Notes on an Imperfect Science. New York: Picador; 2003. ISBN: 0312421702.

3. Pellegrino ED. Prevention of medical error: Where professional and organizational ethics meet. In: Sharpe VA. Accountability: Patient Safety and Policy Reform. Washington, DC: Georgetown University Press; 2004:83-98. ISBN: 158901023X.