“So, I’m getting routine labs on her.”

Wait, what?

Statements like these often make the hair on the back of my neck stand up. One of our residents was seeing a healthy young woman for her “annual physical,” seen just a few days after a routine postpartum visit with her OB/GYN after the uneventful birth of her third child two months ago.

The resident was coming for his afternoon continuity clinic from an inpatient rotation, where he is on the cardiac step-down unit taking care of a large list of patients.

There, everyone is sick, everyone needs labs every eight hours (troponins or PT/PTT), continuous telemetry monitoring, serial EKGs, and they’re all teetering on the end edge of instability.

Then he comes to see us for his afternoon outpatient practice session, and has to change the whole way he has been thinking, change his whole mindset about how he sees patients, and there are some lessons to be learned here.

He had ordered a CBC, complete metabolic profile, and a lipid panel. Nothing dramatic or crazy, and perhaps not even likely to be problematic. But every test comes at a cost. Often multiple costs, known and unknown, expected and unexpected.

We had a discussion about her recent hospitalization for the delivery, and reviewed all the blood tests she had during her admission (never knew you needed so many blood tests done just to have a baby!). She had previously been anemic, but was not so during this admission, even after the birth of the baby.

Digging back through her chart, we found a ridiculously low cholesterol reading she had done at our hospital about eight years ago, most of it good HDL cholesterol.

And what about this complete metabolic profile he ordered? Was there any reason to suspect that she has suddenly developed liver disease, or kidney disease, or has an anion gap acidosis?

No, she felt fine, everything was normal, and probably would be so when the labs came back.

A lesson I learned long ago was not to check cholesterol in breast-feeding women, since they often become cholesterol factories, helping the newborn baby myelinate their developing nervous systems, and why in the world would you want to tell a healthy young woman that her cholesterol is high and that she might need medications for this?

There’s almost no cholesterol level in almost no woman in her early 20s that I would ever consider treating, let alone give anything more than a little recommendation that as the years go by she should watch what she eats, make some good healthy diet changes. But not while she’s helping her baby grow and develop.

I recall as an intern rounding in the medical ICU, terrified of missing something, checking everything, fretting every detail. One morning on rounds, as I reviewed the morning labs with the team, I reported that a patient’s magnesium level was normal.

The attending asked me what the patient’s magnesium was the day before. And the day before that. And before that.

When I told him they had all been normal, he asked me if I knew of anything we were doing to the patient that would perturb her magnesium, and why we needed to keep checking it.

Because she was in the ICU? Because we own a lab? Just because?

We stopped checking her magnesium, and she was fine.

Studies have told us what to do, guidelines are published, expert recommendations appear in print, and now our electronic health record prompts us what to do and what not to do.

But ultimately, we’ve got to use our brains. We’ve got to learn to think about doing what’s right, what’s best for each patient.

As the Cost of Care and Slow Medicine movements gradually work their brilliance and insight into the ways we practice medicine, we need to continuously remind ourselves that we don’t need to get labs, just because we own a lab, just because the patient has blood, just because we have tourniquets lying around and 28 gauge needles that need to be stuck somewhere.

It’s always easier to get a CT scan when the patient has a headache, it’s always easier to say let’s get some labs and see what we’ll see, and then we’ll know more about what’s going on.

But sometimes the best treatment is some good old talking to, some common sense advice, and a tincture of time.

To this day, patients often come to see us from a doctor who has been taking care of them for many years, and are shocked when they don’t get an enormous panel of labs, urinalysis, EKGs, and chest x-ray, done at every routine visit.

If we are going to truly be patient-centered, then we need to truly act in the best interest of our patients, not giving in to every request for a test, or a medicine, or a treatment that they think they need right now, sometimes with the caveat “I’m not a doctor but … ”

Perhaps someday artificial intelligence, or a really smart electronic medical record, or whatever thinking systems that our IT friends come up with to link to our patient care activities, will be smart enough to tell us what needs to be done or not done, without defaulting to a system that tests for everything, just because.

For now, we have got to continue to do no harm, and to keep reinforcing in the next generation of providers that every test we do, from a historical question to a physical exam maneuver to a lab test to a consult, comes at a cost, and we need to be thoughtful and watchful of those costs, to always do what’s best for our patients.

Routine or otherwise.

Fred N. Pelzman is an associate professor of medicine, New York Presbyterian Hospital and associate director, Weill Cornell Internal Medicine Associates, New York City, NY. He blogs at MedPage Today’s Building the Patient-Centered Medical Home.

Image credit: Shutterstock.com