Decision Fatigue Is Intuitive—But Time of Day Is Not Proof
Decision fatigue in healthcare is a claim about what happens to professionals as they make one demanding medical judgment after another. The question is narrower than whether a clinician can feel worn down: can cumulative effort be shown to alter later judgments, and how can that effect be separated from ordinary changes in cases, staffing, and time of day? This distinction gives readers a practical way to judge the evidence without dismissing either the experience of fatigue or the need for stronger causal proof.
The subject is sequential medical judgment
Healthcare professionals rarely make one decision in isolation. A nurse answering a triage call may listen to a patient’s description, ask follow-up questions, interpret symptoms, weigh possible risks, and decide how urgently the patient needs in-person care. A physician may move from one diagnostic, prescribing, or treatment decision to the next. Each judgment is bounded by guidelines and professional training, but it still requires selecting and processing information under uncertainty.
That sequence is what makes decision fatigue a healthcare question rather than a general statement about mood. The proposed problem is not simply that a person becomes uncomfortable, sleepy, or less cheerful after work. It is that the cumulative mental burden of effortful decisions changes the content or quality of subsequent medical judgments. A professional might seek less information, rely more heavily on a familiar rule, choose a simpler option, or become more conservative when deciding what should happen next.
The kinds of decisions matter. A change in diagnostic questioning is not identical to a change in antibiotic prescribing; a shift in cancer screening or a therapeutic recommendation may have different causes and consequences. The 2025 systematic review treats all of these as relevant parts of the healthcare literature, but that breadth is also a warning. If studies use the same label for different outcomes, a pattern that appears repeatedly may still fail to identify one common mechanism. The subject remains healthcare professionals making real decisions, not an abstract score called “performance.”
The stakes are obvious, but the claim is easy to overstate. A later decision can differ from an earlier one for many reasons: the patients may not be comparable, the service may be receiving a different mix of calls, or the organization may operate differently at night. Even when a professional is genuinely tired, that does not by itself show that fatigue caused a particular medical choice. Three statements must be kept apart:
- Fatigue can be experienced. Sustained mental work may feel effortful, boring, frustrating, or exhausting.
- Decisions can vary with time or position in a sequence. A later appointment or call may produce a different observed outcome.
- Cumulative decision effort causes the later change. The change follows from the mental burden of earlier decisions rather than from other features that travel with the clock or the queue.
Only the third is the decision-fatigue hypothesis in its causal form. The first may be true without the second, and the second may be true without the third. Keeping these claims separate is not an exercise in skepticism for its own sake. It is how a patient-relevant explanation becomes testable. It also protects practical reasoning: an organization may have good reasons to improve rest, staffing, or workload even if a particular behavior has not been shown to result from decision fatigue.
Why the story feels convincing
The intuitive story has a reasonable starting point. Medical judgment often involves retrieving and interpreting information, weighing trade-offs, and imagining possible outcomes. If those activities become more costly after prolonged effort, a decision-maker might be less willing to gather one more piece of information or to reconsider a familiar option. In that account, a shortcut is not necessarily careless. It is a way of reducing the mental work required to reach a decision.
One formal version of this idea treats information as both useful and costly. More information can improve a decision when the situation is uncertain, but acquiring and processing it takes effort. If the cost rises, a decision-maker may rationally use less information and lean more on prior beliefs. In healthcare, that could mean moving toward a professional’s usual judgment for a particular kind of patient or toward an option that feels safer when a severe condition cannot be ruled out without further inquiry.
This account gives decision fatigue a shape that can be examined. It does not predict that every decision will become worse, that every tired professional will make the same error, or that later choices will simply be “bad.” It predicts particular shifts in behavior under particular conditions. In a telephone-triage setting, for example, researchers can ask whether choices become more aligned with a nurse’s personal default for a type of call, or whether urgency ratings rise after greater cumulative effort.
Those two predictions are meaningful because they describe different kinds of simplification. Moving toward a personal default suggests greater reliance on a familiar prior for a particular patient type. Raising an urgency rating could reflect a safer or more conservative choice when additional information feels costly. Neither outcome is automatically harmful, and neither is automatically proof of fatigue. Their value is that they can be specified before looking at the result and then compared with a situation in which cumulative effort is plausibly different.
A professional’s training, guidelines, or decision-support system could also change the pathway. A structured checklist may make information acquisition less costly; experience may make difficult judgments more efficient; or a high-stakes case may motivate more effort rather than less. These possibilities do not rescue every positive or negative result. They show why the same broad hypothesis may have different observable consequences across organizations and specialties. A general theory must survive those differences by stating what is expected and where the expectation applies.
That specificity matters because a broad claim can explain almost anything after the fact. If a later decision is more conservative, someone might call it fatigue. If it is less conservative, they might call it disengagement or risk tolerance. If the decision takes longer, fatigue might be blamed for slowing information processing; if it takes less time, fatigue might be blamed for premature closure. Without a pre-specified outcome and direction, the explanation can move to fit the result rather than being tested by it.
The mechanism is also not the same as a settled fact about human cognition. The healthcare review and the field study treat the effort-to-fatigue-to-simplification pathway as a theoretical account. It may be plausible, but plausibility does not determine its size, timing, or generality. A mechanism that appears in one kind of work may be weaker in another, buffered by training or decision support, or mixed with sleepiness, motivation, boredom, and the changing difficulty of cases.
Time of day is a clue, not a verdict
A clock makes a tempting proxy for cumulative effort. Morning workers may have handled fewer cases than colleagues nearing the end of a shift. A call taken late in the day may follow hours of work, while a call taken at the start of an afternoon shift may not. If the outcomes differ, it is natural to tell a fatigue story.
But clock time carries more than workload. Patients do not arrive in a constant stream with identical needs. Other clinics may be open in the morning and closed at night. The reasons people call, the urgency they report, the staff available, and the procedures used by an organization can all change across the day. Call order is not an independent measure if it is tied to time of day. A later-day pattern can therefore be genuine and still leave its cause unresolved.
A large healthcare field study makes this problem visible. In its descriptive data, the proportion of calls assigned the highest urgency was 14.8% around 08:00, 19.3% around 18:00, and 21.1% around midnight. That pattern looks compatible with a tendency to choose a more conservative or urgent option as work proceeds. The researchers did not treat it as confirmatory evidence, because the same pattern could reflect changes in the calls themselves or in the surrounding healthcare system.
There is a useful difference between asking whether the pattern matters and asking what produced it. If a nighttime service receives more severe calls, higher urgency is operationally important even if fatigue is not the cause. If a clinic’s opening hours redirect less urgent patients elsewhere, the remaining calls may look more serious. If only certain nurses work nights, their training, experience, or usual case mix may be associated with the outcome. A descriptive graph can reveal a real scheduling disparity while remaining unable to identify which of these pathways is responsible.
The better question is not “Do outcomes change over the day?” It is “Can high and low cumulative effort be compared while the other relevant conditions are made as similar as possible?” That may require comparing workers in overlapping shifts during the same clock window, or comparing decisions immediately before and after a break. It also requires naming the expected behavioral signature in advance. A stronger design does not make every alternative explanation disappear, but it makes the fatigue explanation carry more of the burden.
This is why a credible test needs more than a large sample. More observations can estimate a confounded pattern very precisely; they do not automatically turn the pattern into a causal effect. Researchers need a reason that high- and low-effort observations differ in cumulative workload while remaining similar in patient flow and time context. They also need to decide in advance which outcomes count as support, which effects would be too small to matter, and what would count as evidence against the proposed direction.
This is the standard that will guide the discussion that follows. A broad systematic review can tell us how often healthcare studies report patterns consistent with decision fatigue, while a focused registered field test can ask whether two specified behavioral changes appear when cumulative workload is contrasted more carefully. Those forms of evidence answer different questions. Neither should be reduced to a slogan about whether decision fatigue is “real.”
The central lesson is therefore modest but consequential: time of day can tell us when a pattern occurs; it cannot, by itself, tell us why. To understand what healthcare professionals do under repeated effort, we first need a clear mechanism, then a comparison capable of challenging the mechanism as well as supporting it. That is the difference between taking an intuitive explanation seriously and treating it as already proven.