Bias in Healthcare Decision-Making

Mar 08, 2022
28 People Read

Disclosure:  

"Who's the more foolish: the fool, or the fool who follows him?"

Obi Wan Kenobi

The Dispatch…

It’s another morning, on shift as a paramedic for a 911 ambulance in a suburban EMS system. My partner and I had just finished our morning rig check, when suddenly tones go off in the ambulance bay. As we get into the cab of the ambulance, my partner contacts dispatch via the radio reporting that Medic 21 is “en route”. Dispatch responds with the following, “Medic 21 you are being dispatched code 3, to an 82-year-old male for a report of Altered Mental Status with a history of stroke. Spouse is on scene and reports that patient is having difficulty speaking and walking.” Because this call was coming from a remote location in our district, the fire department was also dispatched to the scene since they were much closer, which gave them a much quicker response time.

When we arrive on scene it is at single-story private residence. We are greeted by a firefighter who helps my partner bring in the stretcher, as I carried in our cardiac monitor and ALS kit. As I walk in the doors, I see another fire fighter writing on a clip board as they are interviewing the spouse for patient information, another paramedic is taking vital signs and running a 12 lead EKG, and the last fire fighter is starting an IV in the patient’s arm. Overall, a well-oiled team of seasoned ALS professionals. The primary firefighter reports that this will be a “Stroke Alert” based on ataxia and expressive aphasia, noticed by the spouse at 0700 when the patient awoke. During such a call, time is critical as medical interventions such as tPA have a narrow therapeutic window to administer once at the hospital. Thus, my partner and I quickly loaded the patient onto the stretcher and into the ambulance. We then go en route code 3 to the nearest stroke center which is 20 minutes away. Prior to our departure one of the fire fighters provides a list of the patient’s meds and history for me to include as part of my patient care report.

En route…

Alone and in the back of the rig with the patient for the first time, I notice just how pale and lethargic the patient really appears. The interview and physical exam are complicated by the fact that this patient is altered and severely hard of hearing. As I continue my care, I notice that the patient is in atrial fibrillation with a rapid ventricular response (RVR) at a rate of 132, and the blood pressure is 104/88, while the blood sugar and remainder of vital signs all fall within normal parameters. I begin to scan his med list and notice that he is takes Warfarin for his chronic afib. While there was no report of any recent falls, I do a quick “head-to-toe” just to be sure. I look for signs of recent trauma, especially a head injury, however on exam I can find no obvious signs of trauma. As I continue with my neuro exam, I notice that he has good extraocular movements and pupils are round and reactive at approximately 4mm bilaterally. As previously mentioned, the patient’s altered mental state and hearing loss challenge my ability to perform a patient assessment, however I do observe that while he is generally weak, he is still able to spontaneous move all 4 extremities. Facial features appear to be symmetrical without any obvious drooping or loss of tone. I forcibly pinch the patients toes and he responds with a loud “Ow”, as he retracts his leg from the painful stimuli. While he responds with few words, his speech generally seems clear and without noticeable slurring. Now 5 minutes from the hospital I give my ambulance report, indicating I have a code 3 stroke alert. The receiving hospital acknowledges my EMS report and instructs us to bring the patient directly to CT for evaluation and possible imaging.

However, as I continue my exam, I am more perplexed about the patient’s presentation and current condition. What do you think?

Arrival…

Upon arrival to the receiving hospital, my partner and I unload the patient and quickly move them through the emergency department and to the CT scanner. Once there I provide the team with a brief update on the patient and a copy of my run report. As we quickly unload the patient onto the CT scanner, we move out of the way and take the stretcher back to the ambulance, allowing the team to continue their care and evaluation of the patient.

Afterwards, while my partner was cleaning and stocking the rig, I completed my chart in the EMS lounge. However, in doing so I notice that I had forgotten to get a signature from the patients primary nurse for the transfer of care, so I went back to patient's room to see if the nurse was available to sign for the patient. Once I found the patient’s nurse, I asked how our stroke patient was doing? He said, “Actually, his blood work came back, and he was extremely anemic, we are currently transfusing him with blood." As it turned out the patient was super-therapeutic on is Warfarin levels and had been chronically bleeding from his lower GI.” I remember thinking, “Wait, what?”

Reflection…

So, what did you think about this call? Have you ever been so convinced that you knew what was wrong and how to fix it (anchoring bias), only to find out that you had unwittingly convinced yourself of your initial diagnosis (confirmational bias)? How do errors in thinking seep into our decision making? Is there anything that can be done to mitigate this from happening?

No doubt if you are reading this as a healthcare professional, you are probably already keenly aware of the risks of anchoring bias in decision making. Yet, even seasoned, and experienced clinicians will fall into this trap from time to time.

So, what is anchoring bias? Anchoring bias is a form of cognitive bias that causes us to rely on the first piece of information we are given about a topic ("Why we tend.." n.d.). Anchoring not only impacts the choices we make, but is often supported by way of confirmation bias, whereby we stop looking for objective evidence and instead settle for what we believe to be true. In the previous scenario I was initially biased by the dispatch information that was suggestive of a possible stroke, this was further reinforced when the group of first responders on scene decided that the patient was to be a stroke alert. At that point, I downplayed my own exam findings of generalized weakness and pallor and accepted stroke as the only possible medical diagnosis. Once I stopped looking for alternate explanations, I subconsciously sought out information that best supported a diagnosis of stroke. If anchoring and confirmation bias are largely subconscious is there anything we can do to minimize their effect?

Ironically, problems in faulty thinking are often best managed by deliberate thinking with metacognitive exercises. For example, try and delay your final decision until you have had the opportunity to consider other possible alternative solutions. Test your initial hypothesis by actively arguing counterpoints to your decision, all the while acknowledging your own risk for decision making bias. It may also be helpful to elicit input from team members that will challenging the current diagnostic momentum by offering other previously unconsidered solutions ("Why we tend.., " n.d.). Lastly, minimize situations that may potentially erode your ability to reason, think, and make decisions. Situations such as emotional distress, fatigue, illness, and hunger are just a few (Tay, Ryan, & Ryan, 2016). While it is difficult, if not impossible to completely avoid decision making bias, these metacognitive strategies can help mitigate some of the common pitfalls associated with anchoring and cognitive bias. Next time, before "anchoring" onto a single diagnosis, start by asking, what biases may be affecting your decision making?

Test your knowledge

 

The Decision Lab. (n.d.). Why we tend to rely heavily upon the first piece of information we receive. https://thedecisionlab.com/biases/anchoring-bias

Tay, Ryan, P., & Ryan, C. A. (2016). Systems 1 and 2 thinking processes and cognitive reflection testing in medical students. Canadian Medical Education Journal, 7(2), e97–103. https://doi.org/10.36834/cmej.36777

Disclosure: