The human brain is a complex organ with the wonderful power of enabling man to find reasons for continuing to believe whatever it is that he wants to believe.– Voltaire
It’s half five in the morning and the sun is just beginning to rise amongst the trees. The occasional car door is heard as the unlucky few make their way to work ahead of rush hour traffic. Otherwise it is silent. A new day has just begun.
But for me the day is ending, and not soon enough. I am in the last hour of my twelve-hour night shift in the Emergency Department and I am exhausted. We are one doctor down, leaving just myself and a senior colleague to run the entire department. We have had two trauma calls enter resus which has left me alone to deal with the majors and minors patients while my registrar is in resus. The ambulances never stop as line after line flashes on the computer screen in the staff-only area, each one representing another ambulance which is making its way towards our hospital: acute asthma attack – expected arrival time 5 minutes, chest pain – 6 minutes, right-sided weakness ? stroke – 10 minutes, abdominal pain – 12 minutes. The list goes on and on and on as I manage these patients with the help of a team of nurses for whom this is just another day at work. My skills as a doctor go as far as making sure people aren’t going to die imminently, deciding whether they need to come under the medical or surgical team and then rapidly referring them to the appropriate team with a quickfire handover down the phone to a doctor who sounds as tired as I feel.
By 5am things have died down; the most unwell patients tend to come in the early hours of the morning and now all that’s left are a few sprained ankles, the occasional drunk sleeping off her embarrassment and a young man who has come into the department in handcuffs with two policemen. My registrar has finally managed to find five minutes for herself to eat her first meal of the shift while I pick up the clinical documents of the man in custody. This will be my final patient of the shift and I can feel my eyelids drooping. I try to concentrate as I skim the notes: twenty-two, male, involved in a road traffic accident, found to be drunk, just needs a quick check over before being sent to the police station. His observations are stable and there is nothing worrying in the history: no head injury, no loss of consciousness, no seizures. This should be a quick one.
I manage a brief assessment of the man while the two policeman watch and aside from some bruising over his shins he’s good to go. I sign the paperwork and declare him ‘medically fit for discharge,’ then go and lie down in the staff room for an hour.
One hour later my registrar and I are sitting in a room filled with fresh-faced doctors who will be on shift today. We are going through each patient who is currently in the department, making sure any unwell patients are known to the day consultant and the registrar. I sit in silence and offer only a few sentences of information when it is my turn to handover my patients; most are waiting to go home or are now stable and waiting to go to the wards. We are almost done and I can already feel the soft weight of the duvet on my body when one of the receptionists walks in and says there is a phone-call for me.
Maybe I can just go after this call, I think to myself. Handover is almost over anyway. I pick up the phone behind the glass booth of the receptionists station, the screens having been built after the constant abuse the receptionists faced, dealing with the public who wanted to see a doctor ‘immediately!”
“Hello, I’m one of the junior doctors, how can I help?” I say down the phone.
A polite, male voice travels back to me.
“Good morning, sorry to disturb you. I am the Custody Sergeant and wanted to speak to you about Mr Smith; he is currently in custody and I understand you assessed him a few hours ago and discharged him from your department?”
I think back to the patients I have seen today (or yesterday). Even though he is the last patient I saw it takes me a while to remember who he was; the night was a blur and my thoughts are jumbled with the image of my bed.
“Yes, the intoxicated male involved in the road traffic accident?”
“Yes, that’s correct. I wondered if you had x-rayed his right shoulder? You see, it is very swollen and he has a lot of difficulty moving it. I think he may have his right humerus.”
I close my eyes in horror. I’ve missed a broken shoulder, or maybe a broken clavicle. I already know it’s my fault – I didn’t do a thorough clinical assessment and I know why. I was tired, hungry, fed up and just wanted to get some sleep. I had ruled out the red flags and that was that. I don’t say much as the Custody Sergeant informs me he will be sending the patient back to A&E to be reassessed and apologise down the phone before hanging up. Like a robot, I make my way back to the Consultant who will be covering the department today and inform him of my ‘failed discharge’ who will be returning.
We all make mistakes. We are all human. But why do we make these mistakes – and why do we keep making them? Much research has been done into the external factors that affect our decision-making, particularly in the clinical environment: bullying, lack of senior support, fatigue, hunger. But there are also internal factors that affect our decisions; the way we think, the paths we go down to reach our answer and the assumptions we make to come to the best decision as quickly as possible. In other words, we use heuristics.
Heuristics are strategies or short-cuts that help us come to a decision without requiring all of the information. This can be particularly helpful when working in an environment which requires fast decisions and uncertain knowledge. Think back to a recent decision that you made. It can be any decision – it doesn’t have to be life-changing. How did you make it? Did you sit down and gather all the information, weigh the pros and cons of both and then reach a valid conclusion? Most likely not. Such a strategy would be time-consuming and onerous. Instead, we concentrate on our current knowledge and our experiences. We make assumptions, leap to conclusions and follow patterns – and often with good results.
When I made this mistake and missed a possible broken shoulder in the Emergency Department I went home and reflected on what had happened; I knew I had been hungry, tired and overworked. As a result I had conducted only a brief examination and I knew it was not right. But that wasn’t all. Prior to this I had assessed many intoxicated patients who had been involved in road-traffic accidents and most of them had been completely fine. I had been very thorough in my assessments of them and rarely was there anything amiss; more often than not they would return to the police station with some paracetamol and ibuprofen. I was using ‘availability heuristics:’ a mental shortcut which I used to apply my past experience of intoxicated patients involved in car accidents to the guy sitting in front of me; because I had never come across any serious injuries, I had assumed this guy wouldn’t have any either. Of course these mental shortcuts can be incredibly helpful; clinicians use them all the time to come to a diagnosis and management plan. But patients are also complex beings and we need to be aware of the drawbacks when we use these mental shortcuts to come to an answer that will ultimately affect people’s lives.
heuristics, medicine
Medicine as a career choice attracts the scientific and logical. Those that rely on facts and a certain way of seeing the world. When students ask me what is the one thing I wish I had known about medicine before becoming a doctor, I tell them that there is more uncertainty in medicine than I had expected. You are very rarely 100% certain about a diagnosis or investigation. You are not dealing with chemicals and titrations but human beings; equations are not always accurate, A does not always lead to B. Patients do not always present to hospital with the list of symptoms written in textbooks and people’s anxieties and relationships frame what they choose to tell you and what they don’t. The most common phrase you will say as a doctor is “I don’t know.” This is why it is important to be aware of how our brains work when it comes to making mental shortcuts and why we should constantly be questioning ourselves to make sure there is nothing amiss.
The word heuristics is used synonymously with the word ‘bias’ and has thus taken on a tainted reputation. But heuristics can be useful in medicine. Heuristics, which comes from the greek word ‘heurisko’ meaning discover, merely refers to a certain rule or method that is used in problem-solving. Having all of the information is not enough; you need to know – both through expertise and learning – which information is important and which you can ignore. Sometimes too much information can turn into a barrier as you become overwhelmed by detail. You need to know which information is applicable to the individual sitting in front of you – what is the point in prescribing a statin to someone who isn’t going to take it?
There are some heuristics that are more well-known than others. Occam’s Razor, that one should choose the most simple explanation, is perhaps one of the most well-known and is used in clinical decision making, as is the infamous “when you hear hoofbeats, think horses not zebras,” coined by Dr Theodore Woodward, which teaches medical students to think of common diagnoses rather than rarer ones as these are statistically more likely. Another common heuristic is extrapolation; when we take results from studies and apply them to our own practice. For example, antihypertensive medications shown to reduce blood pressure in middle-aged men may not necessarily have the same effect on elderly women, but clinicians will often extrapolate such data to do so. This is not a bad thing – we do not have enough data on many of the treatments that we offer to know how well they work, but we need to be aware of this uncertainty around such treatments to ensure our patients are well-informed.
There are of course examples of when heuristics can lead to biases and these are just as important for clinicians to be aware of. Common examples include confusing correlation with causation e.g. just because people who are on antihypertensive medications live longer does not mean the two are linked and knowledge bias in which people tend to have greater confidence in their own knowledge than is actually warranted.
It is not surprising that A&E is where these heuristics tend to be used most. With limited time and information, these mental shortcuts can be very useful. During my four months working in the Emergency Department I found myself improving as a doctor not only through the knowledge that I learnt but also the confidence that I gained in my decision-making skills. A&E taught me how to think like a doctor and to appreciate the uncertainty that surrounds medicine. Rather than wallow on diagnoses that I may have gotten wrong or patients who I was unsure about, I would reflect on the information that I had at the time, my assessment of the patient and how I used these tools to come to my decision. This was a powerful lesson; indeed, research shows that recent clinical errors are made not as a result of lack of knowledge but errors in the way information is collected, integrated and applied to the clinical context. I would walk away from every patient that I saw and ask myself: what was my initial impression of them? What do I think is wrong with them and why? How confident am I in my diagnosis? What are the alternative diagnoses and how likely are they? These reflections were quick and simple but they gave me the chance to break down my thinking patterns so that while I could reap their benefits, I was also conscious of their pitfalls.
mistakes, medicine
Medical schools teach their future doctors about the science of the human body and signs and symptoms of common diseases. Alongside this vital knowledge, students also need to appreciate the career they are going into; one that deals with uncertainty. A good doctor is one who is able to identify mistakes in their thinking and learn from them. We do this already in our hospitals through Mortality & Morbidity Meetings, Quality Improvement Projects and Simulation, but perhaps we need to pass this teaching on to our future doctors before they even reach the hospital, so that when they do start their clinical placements they come prepared and not overwhelmed.
- Hussain, A. Oestreicher, J. 2018. Clinical decision-making: heuristics and cognitive biases for the ophthalmologist. Survey of Ophthalmology. 63: 119-124
- Marewski, J. N., Gigerenzer G. 2012. Heuristic decision making in medicine. Dialogues in Clinical Medicine. 14: 77-89
- McDonald, C., J. 1996. Medical Heuristics: The Silent Adjudicators of Clinical Practice. Annals of Internal Medicine. 124: 56-62
- O’Sullivan, E., Schofield, S. J. 2008. Cognitive bias in clinical medicine. The Journal of the Royal College of Physicians of Edinburgh. 48: 225-232
- Wegwarth O., Gaissmaier W., Gigerenzer, G. 2009. Smart strategies for doctors and doctors-in-training: heuristics in medicine. Medical Education. 43: 721-728
