Picture a Tuesday morning. The EHR goes down, imaging software freezes, the medication Nothing dramatic. No cyberattack. No disaster declaration. Just a routine system failure. The kind of thing that happens in hospitals more regularly than anyone publicly acknowledges.

In that moment, something becomes very clear very fast: the clinicians who trained before the screens arrived know what to do. They pull out paper. They rely on protocols they actually internalized rather than ones they navigate through a dropdown menu. They function. Slower, yes. But they keep moving. Some of the others freeze.

Not because they are bad physicians or because they lack intelligence or dedication. It’s because they were never asked to function without the infrastructure. The tools were always there for them. The tools were the plan. And nobody once told them that the plan needed a backup.

How We Got Here

I have spent my career advocating for technology in medicine. I performed the first surgery using Google Glass; pushing telemedicine in Maine before the word had entered common use. I believe in these tools with conviction, and that conviction has not changed.

But conviction about technology and honest assessment of its risks are not mutually exclusive. And the risk that doesn't get discussed enough is this: we have built a generation of clinicians inside an infrastructure so comprehensive and so reliable that the infrastructure itself has become the clinical foundation, rather than the knowledge that was supposed to sit underneath it.

The EHR was designed to support clinical decision-making, but in too many environments it has replaced it. Physicians document for the system, order through the system, and in some cases reason through the system (in ways that were never the original intention). The imaging system was designed to assist diagnosis. So much so that clinicians have slowly lost the confidence to examine a patient without one. The medication dispensing system was designed to reduce error. It has also produced clinicians who have never been required to think through a drug calculation from first principles.

When all of it works, none of this is visible. The system hums and patients get seen. Therefore the gaps stay hidden. Yet the moment the system doesn't work, the gap between the clinician who remembers the old way and the one who never learned it becomes the most consequential thing in the room.

The Tuesday Morning Test

I sometimes think about what I call the Tuesday Morning Test. Not a mass casualty event or some sort of pandemic. Just a regular weekday shift when the power flickers and the backup generator takes forty seconds longer than expected and three critical systems go offline simultaneously.

Who in your hospital knows how to run a trauma bay on paper? Who can calculate a pediatric medication dose without the app? Who can prioritize a ward of patients without the electronic board? Who has actually practiced any of this in the last five years?

For the generation of clinicians who trained entirely inside the digital infrastructure, these are not questions with comfortable answers. And as that generation becomes the majority of the clinical workforce, the institutional knowledge of how to function without the systems is aging out of hospitals the same way other forms of expertise age out: one retirement at a time, without anyone deciding it should go.

What This Is (and Is Not)

I want to be precise about the argument because it is easy to misread.

This is not nostalgia talking. The systems that existed before electronic health records were not better. Paper records alone were a genuine patient safety disaster. Going backward is not the answer and I am not suggesting it.

What I am suggesting is that the baseline has to exist underneath the technology. That a clinician should be able to function (not perfectly, not at full capacity, but functionally) if every screen in the building went dark tomorrow morning.

The solution is not complicated, even if it takes institutional will to implement.

Simulation programs that include system-down scenarios. Training exercises that remove the technology and ask clinicians to manage without it. Deliberate conversations in residency programs about what to do when the infrastructure fails. Senior clinicians brought back into education settings specifically to teach the methods that predate the screens.

None of this is radical. It is the same logic that applies to pilots who train on instrument failure and military units that train for equipment loss. The competence exists until the infrastructure fails, and then what matters is what you built underneath it. The technology in medicine is extraordinary and it is getting better. The clinicians who use it should be extraordinary too. That means being extraordinary with it and without it. Right now, the without-it part is the gap nobody is closing.