Common things are common, except when the diagnosis is rare

During my training at Hennepin County Medical Center (HCMC) in Minneapolis, MN, my mentor would use the following teaching pearl during rounds: “When you hear hoofbeats, think of horses not zebras!” He would also frequently use: “If it looks like a duck, swims like a duck, and quacks like a duck, then it probably is a duck.” What did my mentor, the Sage of HCMC, mean by this?

From Wikipedia: “Zebra is a medical slang term for a surprising  diagnosis. Although rare diseases are, in general, surprising when they are encountered, other [common] diseases can be surprising in a particular person and time, and so “zebra” is the broader concept.”

Therefore, when evaluating a patient, a physician must consider common diagnoses before rare ones because “common things are common.” So when a physician “hears” about symptoms that can be explained by a common diagnosis, the common diagnosis is usually the correct one — not the rare diagnosis.

Wikipedia defines the “duck test” as: a humorous term for a form of inductive reasoning. Inductive reasoning… is a kind of reasoning that allows for the possibility that the conclusion is false even where all of the premises are true.

Clinicians are taught that common things are common. However, sometimes, the rare diagnosis is the right one. So if it quacks like a duck, it probably is a duck — unless it is an American coot which is not a duck.

Occam’s Razor is defined as: the simplest explanation is usually the correct one. In other words, if one diagnosis can explain a constellation of symptoms, the clinician should give the patient one diagnosis, not more. If the clinician increases the number of diagnoses, beyond what is “necessary”, the clinician is making more assumptions than necessary.

In  medicine, parsimony is preference for the least complex explanation for an observation. So the thinking is that if one medical diagnosis can explain all of a person’s signs and symptoms, then it must be the correct diagnosis. (I find that Occam’s Razor and parsimony are used interchangeably in the medical community.)

During an evaluation of a patient, the physician evaluates the signs and symptoms of a particular patient’s problem, tries to put all of the information together, and creates a  differential diagnosis (a list of the possibilities describing what a person’s diagnosis may be). Then, the most likely possibilities determine which tests are ordered and which treatments are prescribed.

So which way of thinking is right here? Do we choose the diagnosis that is rare so that we can have one “label” to explain everything, or do we choose more likely diagnoses that are more common even though that may make the situation more complex?

Let me answer the question with a question. Did you know that a patient is more likely to have many common diagnoses than one uncommon/rare one?

For example, a patient with protein in the urine,  high blood pressure, chronic kidney disease and frequent urination is more likely to have diabetes than  Fabry’s disease. Nephrologists need to know about Fabry’s disease because there is a therapy for this  genetic disorder. However, the patient I describe is more likely to have diabetes in addition to high blood pressure as causes of her  kidney failure than Fabry’s (the zebra possibility in this case). One could argue that if we use Occam’s Razor, the Fabry’s would put everything together and simplify the situation — but it’s not the right diagnosis.

Fabry’s disease is an example of a “must not miss” diagnosis that kidney specialists must know about. However, if we are always “ruling out” the zebras, this way of thinking can become expensive and inappropriate.

Search engines can cause confusion for patients because they surf the web and find these zebra diagnoses. Many times I’ve been asked to order extremely expensive, unnecessary tests because WebMD has an article discussing a zebra diagnosis. I sometimes try to redirect the patient’s passion for garnering information from the internet from searching for zebras to learning more about the diseases they actually have.

So remember that if you hear hoofbeats, you are more likely to see a horse … unless you are at Omaha’s Henry Doorly Zoo’s Hoofstock exhibit — then it may be a zebra!

Michael Aaronson is a nephrologist who blogs at his self-titled blog,
Michael L. Aaronson M.D.

Submit a guest post and be heard on social media’s leading physician voice.


Comments are moderated before they are published. Please read the comment policy.

  • Peter Borden


    But what happens when the patient does have a zebra? In my family, we have had a few experiences where, had we listened to the “what’s common is common” voice, it could have been deadly. Personally, as a sufferer of cluster headaches, I was told for years …. “you have stress headaches, just relax” because that was what fit the duck/zebra test for the doctor I was working with at the time. I suffered for years until I found a neurologist who had experience with cluster headaches and was able to identify the correct ‘duck’.

    So what is the role for a patient or carer in being a strong and educated advocate, who has an incentive to look more deeply at a situation to make sure that the right diagnosis is found?

  • Joan Marie Verba

    I believe diagnosis should be evidence-based, that is, the diagnosis that best fits the facts. Taking the most common diagnosis does not serve people who have the uncommon condition. This results in people who have rare or even uncommon conditions being misdiagnosed again and again, perhaps going from doctor to doctor for years before being properly diagnosed. Doctors also need to inform people of how a diagnosis is reached, and involve them in the process of considering which possibilities may apply

  • Lisa Harrigan

    My Husband IS a Zebra. He is a Rheumatic Fever Survivor. Luckily, the Head of Cardiology finally looked at his file and said “Why are you still alive?” Then gave us a list of things to help keep him that way. One of them, “Avoid the usual heart disease protocols, they don’t apply to you.”
    We have since learned the hard way, that his Normal Blood Pressure should be 150/100, lower than 100/70 and he gets fuzzy headed.
    His cholesterol is low normal, his LDL is low normal, his blood sugar is low normal. He does nothing to maintain this. But his kidneys (due to stones) and liver (don’t know why, probably that rheumatic fever) tend to be shaky.
    Unfortunately, because of how he presents, Doctors keep trying to put him on Cholesterol lowering drugs and diabetes drugs to “protect” him. We keep looking at how badly they will behave with his known kidney and liver problems and ask “What are planning to lower his numbers to, The Floor? Because he will be dead from adverse reactions.”
    I do hate having a Zebra for a husband.

  • Judy Burton

    “…when evaluating a patient, a physician must consider common diagnoses before rare ones because “common things are common.” UNLESS the more rare diagnisis requires emergency action to prevent death!! As in the case of chest pain related to aortic dissection vs “common” cardiac causes. An emergency room physician noted that an aortic dissection came to his mind but that it was “more likely” cardiac related (even though all cardiac function and blood tests were normal). Thus he was denied the simple cat scan or EEG that would have revealed his fast approaching DEATH. He simply lay there with pain meds until he died. SAME with my father. He just lay there undiagnosed until hIs aorta blew. His “common” diagnosis was lower extremity ischemia. In either case, no one had ever heard a zebra’s hoofbeats–so how could they be so sure!!!
    And now the Zebra’s offspring are ridiculed or ignored by Drs who think they MUST BE horses because that’s most common and because THEY HAVE NEVER HEARD THE ZEBRA’S HOOFBEATS EITHER and after all zebra’s ARE uncommon!!