Anchoring Bias: Is It Real?

Publication
Article
Oncology FellowsMarch 2017

You may rely too heavily on the information provided by another physician, which is (un)popularly known as “anchoring bias.”

Tania Jain, MBBS

Tania Jain, MBBS

Tania Jain, MBBS

Case 1

During a typical afternoon, you have a cramped schedule with 8 to 9 patients to see during a 3-hour clinic. Your pager buzzes many times, in between these clinic encounters about another patient receiving chemotherapy under your care. The stage is set you to make a decision for your patient without thinking too much. Alternatively, you may rely too heavily on the information provided by another physician, which is (un)popularly known as “anchoring bias.”An elderly woman was evaluated for new-onset strabismus. A magnetic resonance imaging (MRI) scan of her brain revealed a petrous-caval area lesion, which was reported to be a meningioma based on its radiographic appearance. She also reported worsening back pain. An MRI of her spine revealed a L4 lytic lesion. A biopsy of the vertebral lesion revealed findings of a plasmacytoma—an extramedullary version of monoclonal plasma cell neoplasm that is not related to meningioma in the oncology pedigree.

Initially, I was inclined to accept the radiological diagnosis as it was. After all, it was a busy day like any other and my decisions had to be made in the blink of the eye. However, I was unable to eliminate the ever-nagging question in my head: “What if...?”

After deliberating with the patient about the pros and cons, we decided to obtain a biopsy of the intracranial mass. Ironically, this biopsy revealed a plasmacytoma. This was important. It changed our management plan from focal radiation to the hip and spine for the solitary plasmacytoma to, instead, systemic therapy for multiple plasmacytomas and systemic disease. The long-term benefits have yet to be determined, but we hope and believe that our lastminute dramatic intervention allowed us to make the right decision for the patient.

Case 2

At first, I didn’t think there was a reason to share this case. However, I was later inspired to share my story after attending a morbidity and mortality case conference during which the case mentioned below (case 2) was discussed.A middle-aged man was evaluated for a metastatic melanoma located in the lung. The man’s brain scan revealed a lesion that was, incidentally, deemed to be a meningioma based on his radiology report. But was it? This time, a lack of time to consider the man’s options for diagnosis got the better of the clinician’s judgment.

The decision was made to start systemic therapy for melanoma while following the brain lesion, as would be done for a meningioma. As we know, intracranial metastasis from melanoma warrants radiation or surgical intervention. Nevertheless, treatment for melanoma was initiated with immunotherapy, a treatment consideration given former President Jimmy Carter’s case. As we have learned, the median time to treatment response is roughly 3 months with immunotherapy.1

Case 3

Recognizing Anchoring Bias

As oncologists, we don’t get many shots at being right. After a couple of weeks, the patient developed a change in mental status, which turned out to be secondary to increasing size of the brain lesion and worsening edema surrounding it. Upon surgical resection and microscopic evaluation at this time, the brain lesion was found to be metastatic melanoma, which was then managed with postoperative radiation in addition to systemic therapy.After initially being signed out of the emergency department by his physician as having meningitis, as suggested by a high fever, tachycardia, headaches, and tremors, a patient was admitted to the medicine floor for further care. An additional elaborate patient history, obtained by a medical student on the hospital floor, revealed that the patient had a history of hyperthyroidism. The patient also mentioned that he had not taken his medication for the condition in a while. Fortunately, the medical student was neither shy nor slow to speak up. Based on the results of additional workup, a hyperthyroid crisis was diagnosed, which could have become potentially fatal if not recognized in time and managed appropriately.Anchoring bias refers to a cognitive tendency to lock on to a clinical diagnosis too early, often by making decisions based on information provided by another physician. It’s the tendency to fit the symptom complex into the diagnosis box and anchor onto a diagnosis already provided to us.

Historically, anchoring bias was first conceptualized by professors Amos Tversky, PhD, and Daniel Kahneman, PhD. Anchoring bias is an issue not only recognized by physicians in healthcare, but also in other professions and even in our daily lives. Over the years, psychologists have tried to put together various theories describing the mechanism of this occurrence.2 Regardless of how it happens, it can potentially be catastrophic, although rarely so.

We all fall prey to it. After all, it’s convenient! Having said that, we can’t undermine the external factors that play a role in the execution of clinical errors (fatigue, stress, the distraction of a buzzing pager, supporting services such as laboratory data, system pressures to expedite discharges, etc). In a world where medicine is practiced in a more commercial manner, the ticking clock with every patient encounter only reinforces the possibility for errors in judgment.

Addressing Anchoring Bias

It also is important to understand, as a trainee, that believing information already cooked and served to us can hamper critical thinking and kill the spirit of academic inquiry. It adversely affects what we try to seek, every moment of our professional lives—quality of patient care. It leads to clinical error, that “never-forgiven mistake” in medicine. Although anchoring bias may bring a transient sense of reward for being able to explain the story, it offers no long-term glory.I have described several examples of how this fast and frugal approach could have dire and loathed consequences. It would be unjust and incomplete of me to end this article without sharing a few tips for addressing anchoring bias that I have learned through personal experience.

Tip 1: The first step to solving a problem is realizing that a problem does, in fact, exist. Although most physicians acknowledge that errors occur, they may not recall a diagnostic error of their own. Being cognizant of the fact that we are human and that pursuing a deeper understanding never hurts; it is a general tendency we are all expected to have been inculcated with. When juggling multiple medical complaints, which may or may not be relevant to a patient’s cancer or acute medical issue, it is imperative to separate out all relevant information.

Tip 2: When reviewing a case, I suggest taking some time to develop a mental list of things it “could be” and consider each possibility with an open mind. Following from the decree from French National Convention, which promoted “great responsibility follows inseparably from great power,” the importance of differential diagnosis can also be considered inseparable from the actual final diagnosis.

Tip 3: I also suggest seeking insight from colleagues. Team members, whether junior or senior, can all play an integral role when making decisions about diagnosis and treatment. It’s not only helpful to us (as trainees), but encouraging as well, to keep an open mind when running through different variables. Nevertheless, there are, and will always be, unmodifiable factors that will influence our decisions. Therefore, we should accept them, identify them, and shout out to a colleague when appropriate.

References

  1. Robert C, Ribas A, Wolchok JD, et al. Anti-programmeddeath-receptor-1 treatment with pembrolizumab in ipilimumabrefractory advanced melanoma: a randomised dose-comparison cohort of a phase 1 trial. Lancet. 2014;384(9948):1109-1117. doi: 10.1016/S0140-6736(14)60958-2.
  2. Furnham A, Boo HC. A literature review of anchoring bias. J Socio-Economics. 2011;40(1):35-42.
Related Videos
Corey Cutler, MD, MPH, and Hana Safah, MD, experts on GvHD
Shivaani Kummar, MBBS, FACP, Margaret and Lester DeArmond Endowed Chair of Cancer Research, Professor and Division Head, Division of Hematology/Medical Oncology, Oregon Health & Science University School of Medicine; co-director, Center for Experimental Therapeutics, co-deputy director, Knight Cancer Institute
Andre Goy, MD
Wenxin (Vincent) Xu, MD,
Guenther Koehne, MD, PhD
Alessandro Villa, DDS, PhD, MPH
Joseph Mikhael, MD
Michael Richardson, MD
Video 1 - "HR+/HER2- Early-Stage Breast Cancer: Background and Risk Stratification "