The perils of early closure in medicine and management consulting

Happiness in this World has an interesting post from last November (which I happened to see re-posted at KevinMD), which tells the story of a misdiagnosis of one physician by another. The problem: “early closure,” or jumping to conclusions. From the blog:

Early closure, it turns out, is a danger that lies in wait mostly for seasoned clinicians (far more commonly, at least, than for medical students and residents).  Because seasoned clinicians rely more on pattern recognition to make diagnoses and often come to their conclusions rapidly, they’re at far greater risk for leaping toward those conclusions without examining all other relevant possibilities.  Patients often present with a constellation of symptoms that don’t entirely fit the diagnosis they actually have.  Often the discrepancies between these presentations and the textbook descriptions are unimportant—but sometimes those discrepancies exist not because the patient’s body hasn’t read the textbook, but because the diagnosis the doctor makes is the wrong one.  Such misdiagnoses are occasionally unavoidable:  the symptoms with which the patient presents are simply too far afield from the way the medical literature says the disease should present (luckily for us all, this is the exception and not the rule).  At other times, however, these mistakes are made because the physician was simply in a hurry, or tired, or didn’t care enough to think through the evidence in ways he should have, saw a pattern he thought he recognized, and stopped asking the most important question a physician can ever ask:  what else could this be?

I’ve been aware of this tendency in medicine for some time, but this is the first time I’ve thought about how the same problem manifests in other fields and how it’s the more seasoned professionals who fall prey.

It reminds me of my own profession, management consulting. Junior consultants tend to beaver along gathering data and performing analyses, not becoming comfortable with findings and conclusions till everything is in. It’s often joked that their more experienced managers are satisfied with two data points –after all, that’s enough to make a line. And the partners just need one data point –they can assume the slope.

This tendency actually had an impact on my career path at my former firm. One senior partner observed that I walked past him without saying hello once when I was a summer intern. This was apparently brought up during a promotion discussion five years later. Who knows if it was even me? Another senior partner concluded based on an early encounter that I was the quiet, academic type and would be a “farmer” rather than a “hunter.” She may have been surprised when I left to start my own firm!

What I took away from those experiences was the importance of making a good first impression. But now as a “senior partner” myself I also realize the importance of keeping an open mind and not jumping to conclusions. It helps to take a data intensive approach and to work with team members who are not afraid to challenge one another. It’s a good approach when diagnosing business or medical problems, and also a good way to work with people.

September 20, 2011

One thought on “The perils of early closure in medicine and management consulting”

  1. The wisdom of the approach taken by senior experts is that they generate hypotheses quickly and refine them interatively by mulling over how well the hypotheses fit the data. The problems come when one hypothesis fits and others would fit too, or when some possibilities are neglected because they represent new information not familiar to the senior expert.

    In medicine, one way of helping the senior experts as well as the newer clinicians is to use diagnostic software that considers many possibilities, brings in new information, and helps forcus on what further information is most worth collecting.

    Similar approaches would work outside of medicine, and for some of them one could even use types of software that we developed in medicine, re-purposed.

    Imagine if a Google search gave you not only a list of sites but also a set of questions to nnarrow down your search.

Leave a Reply

Your email address will not be published. Required fields are marked *