aka To Err is Human 001
You are handed over a patient by a colleague (that’s a ‘handoff’ to you Yanks). Your colleague says that he is sure that this patient, who presented with ‘shortness of breath’, actually had hyperventilation due to anxiety. He says the patient doesn’t meet PERC criteria and his ‘gestalt’ is that the patient doesn’t have PE. However, the patient was being worked up by an intern who sent off a D-dimer. He says he would usually send a patient like this home, but as the d-dimer is pending he suggests the patient should wait for the result.
Questions
Q1. What type of cognitive error or bias is this?
Entrapment
This occurs when a work-up is started but, even though it is later realised to be unnecessary, the work up is continued because of the initial decision made.
Scott Weingart (EMCrit.org) gives a great example in his book (see references):
“If you’ve gone to the movies and half-way through the film it turns out be utter drivel (perhaps it was chosen by your wife?), should you stay to the end and waste even more time?”
Not me, I’m up and out of there….
Q2. What are some other examples of this error or bias?
There is wide potential for entrapment errors to occur in the ED.
Examples include:
- performing serial troponins on a patient and entering a chest pain pathway just because an initial troponin was performed.
- blood tests sent on patients who have sufficiently improved, or further history has made clear that there isn’t a worrying condition. Such an example is the patient with gastroenteritis who was thought to have an acute abdomen at triage.
- performing C-spine radiographs on a patient just because he or she was put in a collar by the triage nurse or paramedics.
Some of these Pat Croskerry would classify as errors due to the following ‘Cognitive Dispositions to Respond ‘(CDR, a morally-neutral alternative to “cognitive bias”):
- Diagnosis momentum
— diagnostic labels are sticky, possibilities can grow into definite diagnoses- Anchoring
— failing to adjust an initial impression as more information comes to light
or- Sunk costs
— unwillingness to let go of a failing diagnosis due the investment of time, energy and ego
Q3. How can the error or bias be corrected or prevented?
Whenever a work-up being performed ask yourself:
- Is the test really needed? Why?
- Will the result of the test alter patient management, disposition or outcome?
If you can’t come up with a good reason, then:
- Document your reasoning in the notes
- Explain the situation to the patient and/or the staff who initiated the work up
- Cancel the test/ work-up
References
- Weingart S, Wyer P. Emergency Medicine Decision Making: Critical Choices in Chaotic Environments. McGraw-Hill, 2006.[Google books preview]
- Crosskerry P, Cosby KS, Schenkel SM, Wears RL. Patient Safety in Emergency Medicine. Lippincott Williams & Wilkins, 2009. [Google books preview]
About Chris Nickson
