Hang out with doctors, whether in person or online, and you will hear some of them, even very successful ones, say that they would actively discourage their own kids from going into medicine. They say that healthcare has become too corporate, fragmented, and impersonal. That documentation requirements have exploded. They say they’re expected to see too many patients, and that doctors have too little say in decisions that used to belong to them.
I imagine a pre-med student coming across one of these threads. She’s not a doctor yet; just someone still deciding if this is the life she wants. And I ask myself, “What would she think when she sees a group of experienced doctors basically saying they would tell their kids to run?”
Then I think about the structural changes medicine has seen, even in the time I have practiced.
For eight years, I worked in a town of about 9,000 people. Lake Superior lay to the north, and forests and small farms surrounded the area. The town had only one hospital and one psychiatric ward. The closest other inpatient psychiatric ward was 70 miles away.
In other words, I was a rural psychiatrist. If one of my outpatients needed hospital care, I treated them. I did morning hospital rounds. If a patient was admitted to a medical ward, I’d visit them there, sometimes just to check in. In the afternoon, I saw outpatients at my clinic. If a hospital patient was discharged, I’d see them in my office a few days later.
This was nothing like the big academic centers in the city where I trained and studied medicine.
In that rural setting, continuity of care was simply part of the job. I saw for myself how much it helped. I knew my patients and their families. I remembered which treatments hadn’t worked for them three years ago, and I knew why. When I saw my patients in the office, I could tell who was simply having a tough week versus whose illness had gotten worse.
After eight years, I moved back to the city. Even now, I think about that time whenever I see someone write, as they did on LinkedIn last week, that reducing variability in care improves quality, and that higher quality brings lower costs.
That idea sounds easy to agree with. But it’s also where the seeds of burnout in medicine are sown.
There are real reasons behind the push to reduce variation. Reducing variation prevents mistakes that harm patients. It can help control costs that are truly out of hand. It can also help make better use of one of healthcare’s most expensive and limited resources: a clinician’s time. All three reasons are valid. But none of them are the main problem I’m going to focus on in this series.
Not everything that can be counted counts, and not everything that counts can be counted.
— William Bruce Cameron, Informal Sociology (1963)
Not All Variation Is a Problem
When people talk about reducing variation, they often imply that all variation is worth preventing. The idea is that if you cut down on this variation, quality goes up and waste goes down.
Some variation in healthcare is truly unnecessary and preventable, and removing it makes sense.
If one nurse checks for medication allergies and another does not, that is not just a difference in style. It is a mistake waiting to happen. That should be standardized.
Standardization also helps when everyone follows the same steps to check what’s in an IV bag and who will get it, or to confirm and mark which limb needs surgery. There are plenty of examples like this.
But over time, “unnecessary” and “preventable” started to mean almost anything that isn’t exactly the same every time. This could be the extra ten minutes a doctor spends with a patient who needs to talk before they can listen, or a diagnosis that only makes sense because you’ve seen this person many times, not just once. It could even be how knowing a patient and her family leads to better care.
Just because a variation can be prevented doesn’t mean it isn’t needed.
Two Kinds of Variation, and Only One Is the Problem
Standardization can add real value in healthcare, but it can also take value away. The first is easy to spot. The second is much harder to notice, and I think it’s behind much of the frustration those doctors I mentioned above express, even when they say it’s about the paperwork.
The first kind is variation worth preventing. This includes missing a step that would embarrass you if a colleague noticed, or something less obvious, like not giving a treatment that evidence clearly shows works just because no one made sure it happened. There’s no good alternative to either. This kind of variation should be standardized out. My next post will focus on this.
The second kind doesn’t have an official name in the research, so I’m calling it load-bearing variation.
You can’t always tell if a wall is load-bearing just by looking at it. You only find out when someone tries to take it out, thinking it will be easy, and then the ceiling collapses. The information was always there, maybe in a blueprint or in someone’s mind, but no one checked before making the change.
Load-bearing variation can show up in different ways. Sometimes it’s a patient’s own values that help find the right answer. Other times, it’s a doctor’s knowledge of a patient, built up over months or years, that no protocol could create and no patient would know to ask for.
Both of these are often labeled “unnecessary” by those who want to reduce variation. But they aren’t. I think this confusion, repeated across millions of appointments each year, is a big reason why so many good doctors are telling their kids to choose another career.
What is essential is invisible to the eye.
— Antoine de Saint-Exupéry, The Little Prince
Where This Series Is Headed
In the coming weeks, I want to break this argument down step by step:
- Where standardization is truly needed, and why doctors should accept it in those cases without complaint.
- The load-bearing variation that comes from what a patient wants.
- The load-bearing variation that comes from what a doctor has learned about the patient over time.
And what all of this really costs — a workforce that’s already stretched thin, and a patient in the exam room who doesn’t realize what’s been lost.
I’m calling this series The Standardization Tax, because every time you standardize something, you don’t just save money; you also lose something. The key question is: Who pays for what’s lost?
If that pre-med student is still reading by the fifth post, I want her to get an honest answer — not the panicked version from the group chats, but the real one.
The One Practice
Before the next post, try this: think of one process at your workplace that has been standardized in the last few years.
Ask yourself: What kind of variation did that change remove? Was all of the lost variation truly unnecessary and preventable? What meaningful thing did the change cost?
You don’t need an answer yet. Just keep the question in mind.
This post is Part 1 of a four-part series in which I explore where standardization helps in medicine, where it costs us something important that cannot be measured, and how patients and doctors pay the price that I call the Standardization Tax. The next post in the series will be published in one week.
