Whenever You Measure, You Define: The Hospital's Unbudgeted Resource

On a normal afternoon round, a registrar is trying to hold the full picture of a deteriorating patient in her head. While she does, the lab system flags a potassium of 3.4 — mildly low, almost certainly irrelevant here. The pharmacy module raises an interaction that belongs to another specialty. An administrative form needs a signature before the screen will let her move on. A nurse rings to ask whether a discharge letter is finally ready. Each interruption is reasonable on its own. Together, in the space of two minutes, they have emptied the one thing she cannot reorder from the pharmacy: her concentration.

We run hospitals on the things we can count. We have figures for bed occupancy, for cash flow, for medication stock down to the last vial. We have almost no visibility into the depletion of what is often the scarcest resource on the ward — the attention of the people we pay to think. And there is a quiet law at work here. Whatever we decline to measure, we implicitly treat as free. Whatever we treat as free, everyone feels entitled to spend.

A commons no one is pricing

Economists have a name for what follows. The tragedy of the commons describes a shared resource that everyone draws on rationally and no one maintains, until it is grazed to the bare ground. In the hospital, the commons is the clinician's eyes. The lab system, the pharmacy module, the compliance form, the well-meaning reminder — each was designed in isolation, by people who never had to account for the cost of the interruption, because no ledger anywhere records it.

So the infrastructure grazes freely. Each new alert seems almost costless to add, and so they accumulate, until the noise floor rises high enough that the one signal that mattered — the patient who is genuinely turning — is lost inside it. A bed without attention is just furniture. A monitor that no one has the spare mind to interpret is just a light in the corner.

The cost of switching

The damage is not merely irritation; it is measurable. The research on knowledge work is consistent: once a complex task is interrupted, it takes a surprisingly long time to return fully to it — in one well-known study, an average of around 23 minutes. On a busy ward, that window of recovery rarely exists before the next interruption arrives.

So when we make a physician click “acknowledge” on a trivial pop-up in the middle of synthesising a difficult case, we tell ourselves we have asked for ten seconds. What we have actually done is scatter the fifteen minutes of unbroken thought the case required. It is an expensive trade. We hire people for their judgement and then design an environment that prevents them from assembling it.

Treating attention as a budget

The usual response is to file this under wellbeing — alarm fatigue as a wellness problem for human resources to soothe. That is the wrong department. It is an operational problem, and it belongs in governance alongside the things we already audit. We scrutinise where the money goes; we could, with no new technology, scrutinise where the interruptions go. How often, in an hour, is a clinician forced to interact with a screen purely to satisfy a process rather than a patient?

And once you are counting, you can set a threshold. An alert ought to earn its place. If a warning fires and is wrong far more often than it is right, it is not protecting anyone; it is spending a scarce asset and returning nothing, and it should be switched off. The discipline is unglamorous and entirely familiar: we apply it to every other resource in the building. We have simply never applied it to the one inside the clinician's head.

The better hospital is not the one with the most sensors beeping. It is the quieter one — the one that uses its systems to suppress the ninety-nine trivial signals so that the hundredth, the one that matters, lands with absolute clarity. We budget our beds carefully. It is time we learned to budget our people's attention with the same seriousness. None of this is easy. But then, little in medicine ever is.