How Much Dehydration Does It Take? Part Three: Why the Answer Is Unstable

Abstract

Parts One and Two produced answers that do not sit together comfortably. Severe dehydration clearly impaired cognition; ordinary variation in water intake predicted nothing across two years and nearly two thousand people.

This part covers a 2026 methodological review in Sports Medicine that explains why this literature keeps producing unstable answers — and reports what happened in the small number of studies where participants did not know whether they were dehydrated.

Two things emerge. The field does not agree on what the word dehydration means. And when the experience of thirst is removed, the effect does not simply persist — it depends on how the concealment was done.

1.A Word Without a Definition

Rúben Francisco and Analiza Silva at the University of Lisbon, with Lawrence Armstrong of the Korey Stringer Institute at the University of Connecticut, open by observing that experts continue to debate what dehydration is.

The traditional definition is excessive loss of body water. Modern accounts classify by which fluid compartment is affected, distinguishing hypotonic, isotonic and hypertonic states. Clinical bodies prefer water-loss and salt-loss dehydration; European guidelines use low-intake dehydration; some authors reserve the word for intracellular loss specifically.

There is also a distinction the popular literature never makes at all: dehydration as a process — losing fluid — against hypohydration as a state of established deficit. Studies that measure one and report the other are comparing different things under one heading.

This is not academic tidiness. If four studies use four definitions, a meta-analysis pooling them produces a number that does not describe anything in particular, and disagreement between studies is guaranteed before any data are collected.

2.Underhydration Is Not Dehydration

The review's most useful correction concerns a marker most people trust: dark, concentrated urine as evidence of being dehydrated.

The authors describe it as a common but flawed assumption. People with low daily water intake show higher vasopressin, more concentrated urine and lower urine volumes — with no change in body mass and no change in plasma osmolality. Total body water is not reduced.

Concentrated urine in that situation is the kidney working correctly. It is a hormonal response conserving water, not evidence that water is missing. Stavros Kavouras proposed the term underhydration for this state: low intake and concentrated urine without an actual fluid deficit.

The review cites a controlled study in elite athletes using dilution techniques — the reference method for body water — which found differences in urine specific gravity between low and high intake groups with no differences in extracellular, intracellular or total body water.

This bears directly on Part Two. A cohort in which fifty-six per cent registered as dehydrated while eighty per cent drank adequately is exactly what you would expect if the marker were detecting regulation rather than deficit.

3.The Blinded Studies

Part One ended on a problem that looked intractable: a dehydrated participant knows he is dehydrated, so no oral protocol can separate the physiology from the expectation.

It has been done, by delivering fluid where the participant cannot perceive it — intravenously, or directly into the stomach through a tube. The review devotes a section to these studies, under the heading of nocebo effects.

The results split. Intravenous rehydration studies often show no performance impairment from dehydration at all. Intragastric studies, using body-temperature water so the infusion cannot be felt, consistently report performance declines of eight to eleven per cent.

Same question, both blinded, opposite answers. The review's explanation is that the two methods are not equivalent: intravenous delivery bypasses the gut and uses isotonic saline, which may raise serum osmolality in both arms and mask the hydration signal, while intragastric delivery uses water and preserves the physiological cues — including small oral volumes to activate receptors in the mouth and throat that appear to matter for fluid regulation.

That is a coherent account and it is a hypothesis. What is established is narrower: whether dehydration impairs performance under blinding depends on how the blinding was achieved, and the authors state plainly that this area is still emerging and their recommendations remain preliminary.

The honest reading is that some of the effect in unblinded studies is the experience of thirst rather than the physiology of water loss, and that nobody currently knows the proportion.

4.Everything Else That Varies

The review catalogues the other sources of variation, and the list is longer than a reader of the popular coverage would guess: how the dehydration was induced, how long the protocol ran, what kind of fluid was lost, the environmental conditions, the participants' characteristics, and the assessment methods.

It also names the underrepresentation of female participants as a specific problem. Part One's trial was twelve men, which is typical rather than exceptional.

One number does survive all of this. There is considerable consensus, the authors write, that dehydration impairs endurance performance when fluid losses exceed two per cent of body mass. That threshold is the closest thing this field has to a settled quantity — and it is about endurance, not cognition.

5.Where Three Parts Leave It

Thirty-six hours without water measurably impairs memory, attention and mood, in twelve young men who knew what was being done to them.

Across 1,957 older adults over two years, a calculated marker of hydration status weakly predicted cognitive decline while actual water intake predicted nothing — in a population whose metabolic condition contaminates that marker.

And the field itself reports that it lacks an agreed definition, that its most trusted everyday marker often indicates regulation rather than deficit, and that its blinded studies disagree according to how the blinding was performed.

So the answer to the question this series opened with — at what deficit does cognition change — is that it is not known, and that the research needed to establish it has been specified but not yet done. Which is a more useful thing to be told than a number somebody made up.

What is left standing is worth stating without embarrassment. Drink when you are thirsty; thirst is a well-calibrated signal and the reason concentrated urine exists is that the system is working. Severe fluid deficit is genuinely bad for cognition and for much else. Between those two facts sits a large, confidently occupied space that the evidence does not currently support.

Editorial Comment

MindHeaven® makes no claim relating to hydration, sells no electrolyte or hydration product, and would have nothing to gain from this literature being stronger than it is.

We wrote three parts on it anyway, because hydration occupies a peculiar position: it is universally recommended, costs nothing, and is treated as too obvious to check. Advice that nobody checks is advice that quietly accumulates claims it never earned.

The nocebo section is the part we would most want a reader to carry over into how they read us. A field found a way to remove the participant's knowledge of what condition they were in, and the effect became unstable. There is no supplement whose evidence would survive that test better than dehydration's, and we would rather point at the standard than pretend we already meet it.

One closing health note. Persistent thirst that does not settle with drinking, or a marked change in how much someone is drinking or passing, is worth a doctor's attention rather than a hydration strategy — diabetes and kidney disease both present that way, and both are far more identifiable early than late.