It was a privilige to meet Nicola Jakeman at the London EM trainee’s conference. She gave a great lecture on why it’s important for EM docs to understand Human Factors and was one of my co-judges for SimWars (she was the nice one….!).
Anyway, Nicola has asked us to help share some cases that reflect Human Factor issues in the ED. These run as cases in the ED, and you should work through them sequentially. At the end we would really like to hear what you think as if you think these are a good idea there are more to come. Please post comments on the case, the answers and……, well whatever you think.
As always the case is a teaching case (i.e. it’s hypothetical for learning purposes).
Good luck and may the Factors be with you……….
[DDET “Case 1. I can’t change my mind……..”] A woman presents to the ED with diarrhoea and vomiting. She also had vaginal bleeding, fainting and severe abdominal pain. The patient tells the doctor that she might have eaten some uncooked chicken the previous evening. She tells the doctor that she is about about 7 weeks pregnant. The patient is seen by a junior emergency physician and then reviewed by a junior gynaecologist. The gynaecologist diagnoses the patient as having gastroenteritis and discharges her.
The patient returns to the ED with increased pain in the early hours the following night following a collapse at home. She is found to be hypotensive and tachycardic. The notes written by the gynaecologist from the previous day are reviewed and the likely diagnosis of gastroenteritis is noted. The patient is seen by a junior emergency physician who is also managing a couple of other patients. There is one other ED doctor on shift and 30 patients in the department The patient is given a large amount of fluid for what is presumed to be dehydration secondary to gastroenteritis, and some morphine. The patient is not reviewed in the ED by a senior emergency physician or gynaecologist The patient is then discharged at 4 hours. At this point she remains tachycardic, although this has improved and her blood pressure is within the normal range.
Other than a B-HCG no other investigations are performed.
She returns a few hours later in extremis. Autopsy later reveals a large haemoperitoneum from a ruptured tubal pregnancy. [/DDET]
[DDET “Let’s stop and think……”] As part of our daily work as Emergency Physicians we are required to make multiple decisions which effect the treatment and care our patients receive. However, the process by which we make decisions is fallible and can lead to misjudgements and errors causing patient harm. Understanding the limitations of our personal decision making processes and being aware of those situations which predispose us to making an error can go some way to supporting safe decision making. Diarrhoea and vomiting is a recognised but unusual presentation of an ectopic pregnancy, and the inexperience of the junior doctors involved in this case probably played a significant role in the outcome. However, even the most senior clinicians make errors. Considering the case above describe factors which may have adversely influenced the decisions made by the Emergency Physician when the patient returned for the 2nd time?[/DDET]
[DDET ” Question 1: What might be the potential ‘red flags’ in this case that the medical and nursing staff should have been concerned about?”] [/DDET]
[DDET “Q1 – the model answer”]
The patient might have been seen at about 2am. The natural circadian rhythm is such that ideally we would be in our deepest sleep at this time. The doctor was probably fatigued. Fatigue predisposes us to:
- •Lack of concentration
- •Periods on inattention
- •Reduced vigilance
- •Reduction in level of alertness
- •Slow actions
- •Alteration in short term memory
- •Loss of critical analysis
These factors conspire to increase the risk of fixation error. This occurs when the practitioner concentrates solely upon a single aspect of a case to the detriment of other more relevant aspects. For example focusing on the diagnosis gastroenteritis, ignoring the fact that the patient has PV bleeding, a positive B-HCG and shock.
When we become fixated on one diagnosis, there is a risk that we look exclusively for evidence to support that diagnosis, ignoring evidence to the contrary, or manipulating it to meet the perceived diagnosis. This is termed confirmation bias.
The ED doctor may have perceived the gynaecologist to have more seniority than him/herself and be reluctant to challenge their diagnosis, despite evidence to the contrary. Cognitive overload may have been a contributing factor. The ED doctor was managing 3 patients simultaneously. She was probably interrupted frequently leading to breaks in his/her trains of thought. There is increasing evidence to suggest that the human brain can only manage 7 or perhaps just 5 things at once. Once pushed beyond this our capacity to make decisions is adversely effected.
Q 2. Thinking about the factors discussed in question one, describe strategies that you could employ under similar circumstances?”
[DDET “Question 2. – the model answer.”]
1. Recognise the critical impact of fatigue and sleep deprivation on decision making
2. Recognise when you are being unduly influenced by a pre-existing diagnosis.
3. Review all the evidence presented to you and acknowledge when something doesn’t fit your working diagnosis and consider alternative diagnoses. At this point you may have recognised that you have got overly attached to a particular diagnosis but can’t think of an alternative. It is important under these circumstances to seek a second opinion.
4. Recognise that interruptions and distractions can have a negative impact on decision making. When dealing with a complex or critical patient defer or deflect less important decisions. [/DDET]
[DDET “Let’s go through the key learning points here.”]
1. Decision making is fallible
2. Fatigue has a critical impact on decision making
3. Recognise those circumstances which would make you prone to becoming fixated on a particular diagnosis despite evidence to the contrary
4. Spend some time reflecting on patient’s you have seen, and consider the decision making processes that you have used to reach a diagnosis. [/DDET]
[DDET “References and further reading”]
1. M. Rall, P. Dieckmann. Crisis Resource Mangement to Improve Patient Safety. Euroanaesthesia. May 2005 107-112
2. Crosskerry, P. Achieving quality in Clinical Decision Making: Cognitive Strategies and Detection of Bias. Academic Emergency Medicine November 2002, Vol. 9, No. 11 1184-1204
3. Confidential Enquiry into Maternal and Child Health (CEMACH) 2007 [/DDET]
Case studies on St.Emlyn’s
We do present hypothetical cases on St.Emlyn’s. These are based on the experience of our team as educationally active emergency physicians. For centuries doctors and nurses have used stories to teach and learn from each other. However, we are careful not to break any patient confidentiality rules. As a result if we present a case then it will always be fictional and not relating to any specific case or patient. For example if we present an (anonymised) X-ray or ECG we will create a clinical history that is compatible with the radiological/ECG findings but which does not relate to a specific time, location, patient or circumstance.
Whilst it may be argued that this detracts from the clinical learning we believe that patient confidentiality is more important in these matters.
We will create time, date, age, sex, details of the patient and their circumstances etc. Our cases are therefore an amalgam of different cases and experiences. Any resemblance to patients treated by us now, in the past or the future is entirely unintentional and accidental. Our cases are presented to help us all reflect and learn, in that way we might become better clinicians for our patient.
Vive la FOAM! (Free Online Medical Education).
11 thoughts on “Human Factors cases with @JakemanNicola.”
These are great cases and more of them would be useful.
In retrospect it’s easy to get the diagnosis right.
I know most places insist on SpR review for 2nd presentation – I’m not convinced this would alter any of the factors present here.
I think a senior second review is a really great idea. Two reasons. Firstly a more senior opinion should be a good one, but also a second pair of eyes is also good.
However, if you ask for a second opinion choose your words carefully. There is a world of difference between the following statements.
1. The woman (Mrs X) in cubicle 3 has abdo pain. I’m sure she does not have appendicitis but I can’t convince her. Can you have a look….
2. The woman (Mrs X) has come in with abdo pain. Could you have a look and tell me what you think.
The former predisposes the second opinion to agree with you, and whilst you might think that’s what you want – you don’t!
Great tip. 2nd version helps avoid the trap of confirmation bias
Great case. Way too easy to anchor to dx from previous visit.
Great case Nickie, thank you for posting. Really good to see cases looking at the wider issues such as human factors not just focusing on the clinical aspect. I think the idea of a senior review for re-attenders is good practice, however, you can see how these human factors may lend itself to a similar outcome no matter what grade the physician is.
As doctors we need to be aware of these issues, in order to adjust our practice to reduce risk and provide an environment which is as safe as it can be. Unfortunately there are times when things beyond our control, such as patient numbers in the dept make this very difficult.
That’s a really tragic case, even if hypothetical, but there are some extremely important learning points. I think the way the information is presented to doctors on the shop floor is an important factor here.
The woman in this case “presented with diarrhoea and vomiting”. That builds up an expectation in our minds and a differential diagnosis (which might not include ruptured tubal pregnancy). When we add to that ‘vaginal bleeding’, our initial reaction may be to think that it doesn’t really fit in with the primary complaint. Perhaps we think it’s incidental – and because it’s not the primary complaint, perhaps we think it’s less of a concern. Then we add ‘light headed’, ‘abdominal pain’ and ‘recently ate some dodgy chicken’ to the primary complaint. These later factors all still fit in with a diagnosis of gastroenteritis. With the benefit of experience (and usually that comes by learning from cases like this), more senior doctors will appreciate the importance of excluding serious things like surgical and gynaecological emergencies first – but it’s often experience that teaches us this, and the doctors in this case didn’t have a lot of that. We add to that the cocktail of human factors and we have a perfect storm, a recipe for disaster.
Now, let’s imagine that we present the findings a little differently. All we’re going to do is re-arrange the facts from the case presented above and see what a difference it makes to our mental processes…
Let’s suggest that the primary complaint was abdominal pain in a young female who’s 7 weeks pregnant. Now just about everyone (including many with no medical training) would scream ‘ectopic’ right away. Add to that ‘vaginal bleeding’ and ‘feeling light-headed’ and we’ll be feeling that we’ve virtually nailed the diagnosis. At that stage, you could tell us that there is also diarrhoea and vomiting and that the patient ate some dodgy chicken and it wouldn’t matter – this patient has an ectopic until absolutely proven otherwise.
I guess there are 2 issues here:
(1) We tend to put special emphasis on what we think is the ‘primary complaint’. When there are multiple complaints, this may be inappropriate. It may also be wrong – we might, for example, wrongly assume that the first thing the patient mentions is the ‘main complaint’
(2) We are reluctant to let go of a provisional diagnosis when further information comes along to challenge it. Often this is inappropriate – but remember it’s a one way process…
If we make a relatively non-serious diagnosis (in this case it was gastroenteritis) and some information comes along to challenge that (in this case vaginal bleeding), we should think again about what we might be missing.
BUT if we suspect a serious disease, we shouldn’t stop suspecting it just because something doesn’t fit in exactly with that diagnosis. (In this instance, it was the D&V and dodgy chicken that perhaps didn’t quite fit. The doctors may have ‘ruled out’ ectopic pregnancy because of that atypical symptom, causing a tragedy. In actual fact, atypical symptoms are common with most diseases – and D&V can occur following a ruptured ectopic pregnancy. I would guess that the docs in this case weren’t aware of that).
Recognising these things goes a long way to ensuring that catastrophes like the one reported here don’t happen to our patients.
Really nice example of how system error and human error occur. Like the format of the way this post pans out.
I am interested in the process already in place at the institution though. We have fairly strict rules that returning patients are physically reviewed by a senior exactly to avoid this kind of error. From experience the risk of re-presenters having something serious is high (although this has prompted me to study this ‘anecdote’ in greater detail). To me knowing this human error occurs means you can create a system solution to protect against the swiss cheese effect.
As a recent discussion with @_nmay demonstrated though if the ‘senior’ is victim of human error then maybe you are no better off… 🙂
Pingback: Human Factors cases with @JakemanNicola. Case 1...
I like Amal Mattu’s hoofbeats and thought processes concept here: we need a specific pattern of thinking in EM.
If a primary care practitioner hears hoofbeats they can look quickly for red flags and then assume it’s a horse (because in primary care common things are common).
If a specialtyologist hears hoofbeats they can do lots of tests to prove zebraness (because that’s what they got FRCspecialtyology for).
When we hear hoofbeats our first thought should be “how do I ensure this isn’t a lion wearing shoes?”
Very interesting case from the personal human factors point of view – I have lots of thoughts about the system issues in the case but I guess those will be addressed in later instalments?
Some thoughts partially misremembered from an Amal Mattu podcast regarding patterns of thinking in different medical specialties:
A primary care physician hearing hoofbeats can check for red flags and diagnose a horse (common things are common in primary care).
A specialtyologist hearing hoofbeats can further investigate to confirm zebraness (otherwise what was the point of the FRCSpecialtyology?).
An EP hearing hoofbeats should first think “might this be a lion wearing shoes?”
Great exploration in the case above of the personal human factors; I have many thoughts about the system errors but I’m guessing they will be explored in future instalments?
Pingback: Den mänskliga faktorn. | akut eftertanke