Chapter 15 — Water and Hydration: How Much You Actually Need (It's Not 8 Glasses), Electrolytes, and the Overhydration Nobody Talks About

The Hook: She did everything she was told

Priya Achterberg ran her first marathon at thirty-four, in five hours and twelve minutes, on a warm day, and she crossed the line feeling considerably worse than she'd expected to feel.

Not exhausted — she'd trained for exhausted. Nauseated, headachy, confused, and puffy. Her fingers were swollen. Her wedding ring wouldn't turn. She sat down in the recovery area and someone in a medical vest asked her some questions and then asked her, with a change in tone, how much she had drunk.

Roughly a cup at every aid station. There were fifteen of them. Plus a bottle before the start, and a bottle in the corral, and more at the finish because she felt bad and drinking seemed like the obvious response.

Somewhere in the region of four and a half litres, over five hours, on a run in which she — being a slower finisher on a warm but not brutal day — had sweated considerably less than that.

She had gained weight during a marathon.

Her blood sodium came back low. Exercise-associated hyponatremia — dilutional low sodium caused by drinking more fluid than she lost. She was treated, observed, and went home the same evening, and she was lucky, because the severe form of this causes seizures, cerebral oedema, and has killed people.

Here is the sentence I keep coming back to, which she said in clinic three weeks later:

"Every single thing I did wrong, I did because someone told me to do it."

Drink before you're thirsty. By the time you're thirsty it's too late. Take fluid at every station. Stay ahead of your losses. She had followed the advice precisely, and the advice was the problem.

Hydration is the topic where confident, universal, unsupported advice has the widest reach in all of nutrition. Everybody knows the number. Nobody can tell you where it came from. The market for hydration products has grown enormously, and — uniquely in this book — the failure mode of following the advice too well can put you in a medical tent.

So this chapter has an unusual shape: a widely-believed number with no evidential basis, a body system that regulates itself extremely well, a genuine and underappreciated risk from overdoing it, and — because this book is not a debunking exercise — one of the greatest public health interventions in medical history hiding inside it.

🏃 Fast Track: §15.4 (how much), §15.6 (how to actually tell), and §15.10 (hyponatremia). Twenty minutes.

🔬 Deep Dive: §15.7 (the dehydration-and-performance evidence, which is weaker than everyone assumes), §15.9 (electrolytes), and §15.10 are where students, coaches, and anyone advising athletes should spend real time.

15.1 What water actually does

Around 50–60% of an adult's body mass is water — higher in infants, lower in older adults and in people with more body fat, since adipose tissue holds less water than lean tissue.

It's doing at least five jobs:

Solvent. Nearly every biochemical reaction in this book happens in water. Glycolysis, the citric acid cycle, enzyme function — all aqueous.

Transport. Blood is mostly water. Nutrients in, wastes out, hormones around.

Thermoregulation. Sweat evaporation is the primary mechanism by which humans dump heat, and it is the reason we can run long distances in warm conditions when most mammals can't.

Structure and lubrication. Cerebrospinal fluid, synovial fluid, the fluid in your eyes, the mucus lining your airways.

Excretion. The kidneys need water to produce urine and clear nitrogenous waste — including the urea from protein metabolism (Chapter 6 §6.10).

Losses, on a typical day: urine (the largest and the adjustable one), skin and respiration (insensible losses, roughly 700–1,000 mL/day even at rest), faeces (small), and sweat (highly variable — from near zero to over 2 L/hour in extreme conditions).

15.2 The system that regulates it

Here's why the advice matters less than you'd think: water balance is regulated with remarkable precision, by a system that has been doing it successfully for your entire life.

Osmoreceptors in the hypothalamus detect changes in plasma osmolality — the concentration of solutes in your blood — and they are sensitive to changes of around 1–2%.

When osmolality rises, two things happen:

1. Antidiuretic hormone (ADH, vasopressin) is released, telling the kidneys to reabsorb more water and produce more concentrated urine.
2. You feel thirsty.

When osmolality falls — you've drunk more than you needed — ADH is suppressed and the kidneys excrete the excess. A healthy adult kidney can excrete water at a rate of roughly 0.8–1 L per hour.

That last number is the one that matters for §15.10, because it defines the ceiling. Drink faster than the kidney can clear, sustained, and blood sodium falls.

💡 Aha moment. Notice the asymmetry, because it explains the whole chapter.

Under-drinking has an alarm. Thirst. It escalates, it becomes uncomfortable, and it is genuinely hard to ignore for long.

Over-drinking has no alarm at all. There is no sensation that tells you to stop. Your kidneys quietly clear the surplus, and if you exceed their rate — which requires effort, but is entirely achievable if you're being conscientious — nothing warns you until sodium has fallen far enough to cause symptoms.

The body defends against the risk it evolved with, and chronic access to unlimited palatable fluid is not that risk. Which means advice that overrides thirst is overriding the only signal in the system.

15.3 Where "eight glasses a day" came from

This is one of the best-documented origin stories in nutrition, and it is a small masterclass in how a recommendation mutates.

The trail leads to a 1945 US Food and Nutrition Board publication, which recommended roughly 1 millilitre of water per calorie consumed — about 2.5 litres a day for a typical adult.

And then it said something else. The passage went on, in effect: most of this quantity is contained in prepared foods.

The first sentence travelled. The second didn't.

So a recommendation for total water intake, explicitly noting that most of it arrives in food, became a recommendation for two and a half litres of drinking water on top of everything else — roughly eight eight-ounce glasses.

In 2002, the physiologist Heinz Valtin published a review in the American Journal of Physiology titled, memorably, "Drink at least eight glasses of water a day. Really? Is there scientific evidence for '8 × 8'?" He went looking for the evidence base and reported that he could not find one — no study supporting the specific recommendation, and considerable evidence that healthy adults are adequately hydrated without it.

That was over two decades ago and the number is still everywhere.

🔬 Claim → Evidence → Verdict

The claim: "You need eight 8-ounce glasses of water a day."

Where it comes from: A 1945 recommendation for total water intake — roughly 1 mL per calorie — accompanied by an explicit note that most of it comes from food. The qualifying sentence was dropped and the number survived.

What the evidence actually shows: No study has established 8 × 8 as a requirement. Valtin's review found no supporting evidence. Water requirements vary enormously with body size, activity, climate, diet, and health status — a single universal number is the wrong shape of recommendation, before you even ask whether the number is right.

📉 Evidence quality: No supporting evidence identified; the origin is a documented misquotation.

Verdict: ❌ Not supported. It isn't harmful for most people — you'll excrete the surplus — and it is a fabricated number that has displaced a perfectly good regulatory system for eighty years.

15.4 So how much do you actually need?

The Adequate Intakes, from the National Academies:

Three things about those numbers that get lost constantly.

1. That's TOTAL water, from all sources. Not water you drink. All beverages — tea, coffee, milk, juice, soft drinks — plus the water in food.

2. Food contributes roughly 20%, and more if you eat a lot of fruit, vegetables, soups and stews. A cucumber is about 96% water. Watermelon, roughly 92%. Soup, obviously.

3. It's an AI, not an RDA — Chapter 13 §13.2. It was derived from observed median intakes in apparently healthy populations, not from a measured requirement. It describes what adequately hydrated people happen to drink, which is a much weaker kind of number.

So the practical translation: roughly 2–3 litres of fluid per day from drinks for many adults, with enormous variation — and you do not need to hit it deliberately, because your thirst is already doing this.

What genuinely increases requirement: heat, humidity, exercise, altitude, fever, vomiting or diarrhoea, breastfeeding, some medications, high protein intake (more urea to excrete), high sodium intake, and low-carbohydrate diets in the first week (glycogen depletion releases bound water — Chapter 4 §4.9).

15.5 Is thirst good enough?

For most healthy adults under ordinary conditions: yes.

This is the position of the physiology, and it's a genuinely underappreciated one. Your osmoreceptors detect a 1–2% change in plasma osmolality. That is a more sensitive instrument than any tracking app, and it has been calibrating itself against your specific body since birth.

🔬 Claim → Evidence → Verdict

The claim: "By the time you feel thirsty, you're already dehydrated. Drink before you're thirsty."

Where it comes from: Technically true and misleadingly framed. Thirst is triggered by a change in osmolality — around 1–2% — so yes, when you feel thirsty, something has already shifted.

What the evidence actually shows: A 1–2% shift in plasma osmolality is not "dehydrated" in any clinically meaningful sense — it's the normal operating range of a feedback system, in the same way that feeling hungry doesn't mean you're malnourished. Thirst is an early signal, not a late one, and it reliably drives adequate intake in healthy adults with access to fluid.

And this framing has a specific cost. It was among the pieces of advice that put Priya in a medical tent, and drink-ahead-of-thirst guidance in endurance events is now widely regarded as having contributed to exercise-associated hyponatremia. Current guidance for endurance events has largely shifted toward drinking to thirst for exactly this reason.

📉 Evidence quality: The physiology is clear; the harms of the alternative advice are documented.

Verdict: 🟠 Probably false as usually deployed. Real exceptions, where thirst genuinely is unreliable — §15.11: older adults, illness, some medications, extreme conditions, and situations where you can't respond to it.

Where thirst genuinely under-delivers:

• Older adults — thirst sensation declines with age, and so does renal concentrating ability. This is a real clinical risk.
• Infants and young children — can't act on it independently.
• During intense exercise — thirst can lag behind sweat losses in high-rate situations, though the evidence that this matters as much as claimed is §15.7's subject.
• Illness — fever, vomiting, diarrhoea, and reduced consciousness.
• When you're busy — Camila Ortiz on a twelve-hour night shift genuinely does forget to drink, and that's a logistics problem rather than a physiological one.

🧩 Productive struggle. Five minutes before reading on. This is the chapter's central problem and it's worth attempting properly.

Priya drank ~4.5 L over five hours. Her kidneys can clear roughly 0.8–1 L/hour. On those numbers, she should have been able to excrete the surplus comfortably.

So why didn't she? Generate at least three reasons.

What I'd say

1. Exercise itself suppresses water excretion. Physical exertion — particularly prolonged, particularly with any degree of physiological stress — raises ADH (vasopressin), which is precisely the hormone that tells the kidneys to retain water. So the clearance ceiling of 0.8–1 L/hour is a resting figure, and during a five-hour endurance event it is substantially lower. The system that would normally protect her was working against her.

2. She was losing sodium while replacing only water. Sweat contains sodium (§15.9). Replacing sweat with plain or low-sodium fluid means the total body sodium falls and the volume it's dissolved in rises. Two mechanisms pushing the same direction, which is why the effect is larger than volume alone predicts.

3. The intake wasn't evenly spread. Fifteen aid stations across five hours means she was drinking in boluses. Instantaneous rates at each station were far above the average.

4. NSAIDs, if she took any. Many runners take ibuprofen before or during long events. NSAIDs impair renal water excretion, lowering the ceiling further. This is why the ⚠️ in §15.10 is specific.

5. And the reason the arithmetic looked survivable: she was slow. Five hours means fifteen opportunities to drink and a low sweat rate relative to a faster runner in the same conditions. The people most at risk are the ones with the most time and the least loss — which is exactly backwards from how everyone imagines this risk is distributed.

If you got the ADH point, you've understood the section. It is the piece that makes this a genuine hazard rather than a theoretical one, and almost no hydration advice mentions it.

15.6 How to actually tell

Three practical indicators, in order of usefulness.

1. Urine colour. The most useful everyday marker. Pale straw to light yellow = adequate. Dark yellow or amber = concentrated, drink something. Completely clear, consistently, may mean you're drinking more than you need.

⚠️ Confounders: riboflavin (B2) turns urine bright yellow — so anyone on a B-complex or multivitamin gets an uninterpretable reading. Beetroot, some medications, and some foods also colour it. And first-morning urine is normally more concentrated.

2. Thirst. As above. In healthy adults, listen to it — and note that it is also triggered by dry mouth from talking or mouth-breathing, which isn't the same thing.

3. Morning body weight. For athletes and in hot conditions, a stable morning weight (measured at the same time, after voiding) indicates stable fluid balance. A sudden drop of 1–2% overnight suggests under-replacement from the previous day.

What isn't useful: counting glasses. Hydration apps. Any single universal target.

And a fourth marker worth knowing, for clinicians and for anyone caring for an older adult: in a clinical setting, hydration status is assessed by a combination of things — skin turgor, mucous membranes, blood pressure lying and standing, heart rate, urine output, and blood tests including sodium, urea, creatinine and osmolality. No single one of these is reliable alone, and several perform poorly in older adults specifically — skin turgor, for instance, is confounded by the loss of skin elasticity that comes with age.

Which is worth knowing for a reason beyond trivia: it means "am I dehydrated?" is genuinely harder to answer than the hydration market implies. If a combination of clinical signs and blood tests is what it takes in a hospital, a colour chart on a bottle is not measuring your hydration status. It's measuring the concentration of your urine, which is one input among several and is exactly what a working kidney is supposed to adjust.

15.7 Does mild dehydration actually impair you?

This section requires care, because the standard claims are stronger than the evidence.

Performance

The classic claim: losing more than 2% of body mass in fluid impairs endurance performance.

Where it came from: laboratory studies, many of them in controlled conditions where participants were dehydrated in advance and then tested — often unblinded, often without airflow, and often without the ability to drink to thirst.

What's complicated it: more recent work, including studies using blinded intravenous or nasogastric fluid administration to separate the sensation of drinking from the fluid itself, has generally found smaller effects than the classic figure implies — and field studies have repeatedly found that the fastest finishers in endurance events are frequently the most dehydrated, having prioritized speed over drinking.

The honest position: substantial dehydration impairs performance, that isn't disputed. Whether the threshold is as low as 2%, and whether drinking to thirst is inadequate for most athletes in most conditions, is genuinely contested — with a serious argument that thirst-driven drinking is sufficient and that prescribed drinking schedules carry the hyponatremia risk in §15.10 without delivering proportionate benefit.

Cognition and mood

The claim: mild dehydration measurably impairs concentration, mood, and cognitive performance.

The evidence: studies exist, they're generally small, frequently unblinded (participants know whether they've been given water), heterogeneous in method, and inconsistent in which cognitive domains are affected. Effects, where found, are often modest.

🔬 Claim → Evidence → Verdict

The claim: "Even mild dehydration significantly impairs cognitive performance and mood — a 1–2% loss makes you measurably worse at thinking."

Where it comes from: Real studies finding real associations, plus a mechanistically plausible story, plus enormous amplification by companies selling water and electrolyte products.

What the evidence actually shows: The literature is small, heterogeneous, largely unblinded, and inconsistent about which domains are affected and by how much. Blinding is the core problem — a participant who has been deprived of water and knows it is not a clean comparison, and expectancy effects on subjective mood measures are exactly what you'd expect. Severe dehydration unambiguously impairs cognition; the mild end is much less clear.

📉 Evidence quality: Numerous small studies, methodologically limited, inconsistent.

Verdict: 🟡 Unclear / it depends. Probably something there at the more substantial end; considerably overstated for the 1–2% range, and heavily promoted by parties with an interest. If you feel foggy and haven't drunk anything for six hours, have a drink — the intervention costs nothing. Just don't build a monitoring regime on it.

15.8 Measuring your own sweat rate

If you train, this is worth doing once, because it converts a guess into a number and because individual sweat rates vary by an order of magnitude.

The method:

1. Weigh yourself (minimal clothing, after voiding) immediately before exercise.
2. Exercise for a measured time, in conditions representative of what you actually do.
3. Record all fluid consumed during the session, in litres.
4. Weigh yourself again immediately after, towelled dry, minimal clothing.

Sweat loss (L) = (weight before − weight after) + fluid consumed
Sweat rate (L/hr) = sweat loss ÷ hours

Worked example. Devi, 60-minute run, warm conditions:

Before:              55.0 kg
After:               54.2 kg
Weight change:        0.8 kg
Fluid drunk:          0.5 L
Sweat loss:      0.8 + 0.5 = 1.3 L
Sweat rate:               1.3 L/hr

(1 kg of body mass change ≈ 1 L of fluid, which is close enough for this purpose.)

Typical rates run from about 0.5 to 2.0 L/hour, with extremes beyond that in heat. The variation between individuals is large enough that population advice is nearly useless — which is the honest argument for measuring it once rather than following a schedule.

Repeat in different conditions — cool, hot, humid — because the number changes substantially.

15.9 Electrolytes: when they matter

Sweat isn't just water. It contains sodium primarily, plus smaller amounts of potassium, chloride, calcium and magnesium.

Sweat sodium concentration varies enormously between individuals — commonly cited across a range of roughly 200 to 2,000 mg per litre. Some people are genuinely "salty sweaters," visible as white residue on clothing and skin.

When electrolyte replacement is genuinely warranted

When it isn't

A one-hour gym session. A walk. A commute. A normal day. For these, water is fine and the sodium in your food is more than adequate — recall from Chapter 14 that most people consume 3,000–4,000 mg of sodium a day without trying.

And the one that genuinely deserves ✅

🔬 Claim → Evidence → Verdict

The claim: "Oral rehydration solution is effective treatment for dehydration from diarrhoeal illness."

Where it comes from: A specific physiological discovery — sodium-glucose cotransport. The SGLT1 transporter in the small intestine moves sodium and glucose across the gut wall together, and water follows osmotically. Crucially, this transporter continues to function during many diarrhoeal illnesses when other absorptive mechanisms are impaired.

What the evidence actually shows: The discovery led directly to oral rehydration solution — a precisely balanced mixture of glucose, sodium, and other electrolytes in water — which allows rehydration by mouth in situations that previously required intravenous fluids.

The consequence is difficult to overstate. Diarrhoeal disease was, and in places remains, a leading cause of childhood death. ORS is cheap, requires no clinical training to administer, needs no refrigeration, and works. Its introduction and global dissemination is credited with saving an enormous number of lives — it has been described in The Lancet as potentially the most important medical advance of the twentieth century.

📉 Evidence quality: ✅ Rung 6–8. Trials, decades of programmatic implementation, and a clear mechanism.

Verdict: ✅ Well supported. ⚠️ And practically: the WHO/UNICEF-formula ORS sachets sold in pharmacies are the correct product for genuine fluid-and-electrolyte loss from illness — not a sports drink, which has a different sugar and sodium profile designed for a different purpose. Keep a box. It costs very little and it is the single most useful thing in this chapter for a household with children or older adults.

🔍 Why this works. The sodium-glucose cotransport mechanism is worth understanding properly, because it is one of the most elegant things in this book and because it explains why the ratio in ORS matters rather than just the ingredients.

The intestinal cell membrane has a transporter called SGLT1, and it has a specific and unusual property: it will not move sodium alone, and it will not move glucose alone. It moves them together, coupled. One binds, then the other, and the pair crosses.

Once sodium and glucose have crossed into the cell and then into the bloodstream, they raise the osmotic concentration on the far side of the gut wall — and water follows them across passively, down the osmotic gradient. You are not pumping water. You are pumping solute and letting water chase it.

Now the part that made it a medical revolution. In cholera and many other diarrhoeal illnesses, the toxin drives massive secretion of fluid into the gut lumen — but SGLT1 keeps working. The absorptive pathway that depends on coupled sodium and glucose is largely spared while other mechanisms fail.

So you can give fluid by mouth, into a gut that is actively pouring fluid out, and still achieve net absorption — provided you supply sodium and glucose in roughly the right proportions. Too little glucose and the transporter idles. Too much and you draw water the wrong way, worsening the diarrhoea, which is exactly why a sugary drink is not a substitute and why the WHO formula is a formula rather than a suggestion.

A sachet of salt and sugar, dissolved in clean water, in the right ratio, administered by anyone, anywhere, with no equipment. That is what a well-understood mechanism buys you, and it is worth remembering the next time someone tells you that mechanistic biology never delivers.

And the sports drink question

🔬 Claim → Evidence → Verdict

The claim: "Electrolyte drinks are better than water for hydration. You should use them routinely."

Where it comes from: True physiology applied outside its context. Electrolyte replacement genuinely matters in prolonged exercise and in illness. Sodium does aid fluid retention.

What the evidence actually shows: For prolonged endurance exercise, heat, and high sweat rates, there's a real case (§15.9's table). For ordinary daily hydration and short exercise, the case evaporates — you're already consuming ample sodium (Chapter 14), and most commercial products deliver modest electrolytes alongside substantial sugar. A typical sports drink is largely a dilute sugar solution with some sodium; a typical "electrolyte water" delivers electrolyte quantities small relative to what's in a meal.

📉 Evidence quality: Good for the endurance and clinical contexts; absent for routine use.

Verdict: 🟡 Unclear / it depends — and the "depends" is almost entirely duration, intensity, and heat. 🟢 Genuinely useful beyond ~60–90 minutes of hard exercise, in heat, or for salty sweaters. ❌ For sitting at a desk.

🧾 Cost check. Electrolyte sachets and tablets typically run $0.60–$2.00 per serving, and a daily habit costs $220–$730 a year. For comparison: a WHO-formula ORS sachet costs pennies, and for prolonged exercise, a homemade solution of water, a pinch of salt, and some carbohydrate does most of the same job. The premium is for flavour, convenience, and branding, which are real things — just not physiological ones.

15.10 Hyponatremia: the risk nobody sells against

Priya's section, and the most important safety content in this chapter.

Hyponatremia is a low blood sodium concentration. In the exercise context — exercise-associated hyponatremia (EAH) — it is usually dilutional: caused by drinking more fluid than is lost, so that the sodium in the body is diluted across a larger volume.

Why it's dangerous: sodium determines the osmotic gradient across cell membranes. When blood sodium falls, water moves into cells. In most tissues there's room. In the brain, enclosed in a rigid skull, there isn't — producing cerebral oedema, which causes confusion, seizures, and in severe cases death.

Symptoms, which overlap confusingly with dehydration and with normal post-race feeling:

⚠️ The diagnostic trap that makes this lethal: the mild symptoms look like dehydration, and the instinctive response — drinking more — makes hyponatremia worse. If someone finishes an endurance event confused, nauseated, and puffy, do not give them more fluid until sodium status is known. This is a medical situation.

The single most useful distinguishing sign is body weight. Someone who has gained weight during an endurance event has drunk more than they've lost, and dehydration is not the problem.

📊 Diagram (described). Picture a single brain cell, sitting in the fluid that surrounds it, and think of the cell membrane as a wall that water can cross freely but sodium mostly cannot.

Normally, the sodium concentration inside and outside is balanced such that water has no net reason to move. The cell holds its shape.

Now dilute the fluid outside by adding litres of water without sodium. The concentration outside falls below the concentration inside — and water, obeying osmosis, moves into the cell. The cell swells.

In most of the body this is tolerable, because tissues have somewhere to expand into. Muscle can swell. The gut can swell. Fingers swell — which is why Priya's ring wouldn't turn, and why puffiness is a warning sign rather than a cosmetic detail.

The brain cannot. It is enclosed in a rigid box. Swelling brain tissue has nowhere to go, so pressure rises inside the skull — producing headache, then confusion and nausea, then seizures, then in severe cases herniation and death.

Two things follow, and they're the whole safety argument.

First: the early symptoms are unremarkable. Headache, nausea, feeling unwell after a marathon. A person can be several hours into a dangerous process while presenting as someone who ran too far, which is why it gets missed.

Second, and this is what makes it lethal: the intuitive treatment is the mechanism. Someone feeling terrible after an endurance event is offered fluid, because that is what you do. Every cup makes the gradient steeper and drives more water into the cells.

The scale in the medical tent settles it in ten seconds. If they weigh more than they did at the start, the problem is not that they are short of water.

Risk factors — and the pattern is instructive because it inverts expectations:

• Slower finishing times — more time and more aid stations
• Smaller body size — a given fluid excess is a larger proportion of body water
• Female sex — partly body size, partly other factors
• Longer event duration
• High fluid availability — abundant aid stations
• NSAIDs — impair renal water excretion
• Being conscientious about drinking

Read that list again. This is not a risk that falls on people who are careless. It falls disproportionately on smaller, slower, first-time participants who have been told to drink a lot and have done so diligently. Priya ticked six of seven.

Deaths from EAH have been documented in marathons and other endurance events, and it was the recognition of these cases that drove the shift in guidance.

🍽️ On your plate. Current guidance from endurance medicine bodies has moved substantially toward drinking to thirst rather than to a schedule.

Practically, for an endurance event: know your sweat rate (§15.8) · drink to thirst rather than to a plan · use sodium-containing fluid for long or hot events · ⚠️ avoid NSAIDs around endurance events · weigh yourself before and after — gaining weight is the warning sign · and if you feel unwell and puffy at the finish, say so, and don't drink more until someone has assessed you.

15.11 When the calculus is different

⚠️ When to see a professional. That last row matters more than anything else in this table. Fluid restriction is prescribed in several serious conditions, and general "drink more water" advice is dangerous in heart failure, advanced kidney disease, and some liver disease. If you have been given a fluid target by a clinician, that target overrides everything in this chapter.

See a doctor for: confusion or drowsiness with vomiting or diarrhoea · inability to keep fluids down · very dark urine with reduced output · dizziness on standing · in infants, reduced wet nappies, lethargy, or a sunken fontanelle · and ⚠️ any endurance athlete who finishes an event confused, nauseated, and puffy.

🔄 Check your understanding. Camila Ortiz works three twelve-hour night shifts a week in an ICU. She says she "never has time to drink" and often goes six or seven hours without fluid. A colleague tells her she's chronically dehydrated and should carry a two-litre bottle and finish it every shift. Assess.

Answer

The colleague is right about the problem and wrong about the prescription.

What's genuinely true: Camila is one of the §15.5 exceptions — not because her thirst is unreliable, but because she can't act on it. A nurse mid-shift has a functioning signal and no opportunity to respond, which is a logistics problem wearing a physiology costume (the same shape as Chapter 11's fiber problem for her). Six or seven hours without fluid, three times a week, repeatedly, is worth addressing.

What's wrong with the prescription: a fixed two-litre target overrides thirst in exactly the way §15.5 warns against. It replaces a signal she has with a number she doesn't need. And there's a specific practical problem — finishing two litres on a ward means needing to urinate, which is the same access problem in a different direction. A target she can't meet becomes another thing she has failed at.

What I'd actually suggest: a bottle where she already stops — at the desk, at the station, wherever her existing pauses are — so that drinking attaches to an event that already happens rather than requiring a new one. Chapter 33's argument: change the environment, not the resolve. Then drink to thirst when the opportunity exists, rather than to a total.

And check her urine colour rather than her volume, since that's the output measure and it's free.

⚠️ One more thing worth raising with her: night shift plus inadequate fluid plus caffeine plus irregular eating is a combination associated with kidney stones in shift workers. That's a reason to take it seriously — and a better reason than "you're chronically dehydrated," which she has no way to verify and will reasonably ignore.

15.12 The hydration market

Briefly, because the pattern is by now familiar.

Alkaline water. Claim: neutralizes acidity, improves health. Your stomach maintains a pH of roughly 1.5–3.5 (Chapter 3 §3.3). Alkaline water reaching it is neutralized essentially immediately. Blood pH is regulated within an extremely tight range by the lungs and kidneys and is not meaningfully influenced by what you drink. ❌ Not supported.

Hydrogen water. Dissolved molecular hydrogen, claimed as an antioxidant. Some small studies exist; the literature is early, small, and frequently industry-connected. ⚗️ Untested in any meaningful outcome sense.

"Structured" or "hexagonal" water. Not a recognized physical state of water in this context. ❌ Not supported.

Electrolyte powders for general use. §15.9 — 🟡, and mostly ❌ for a desk job.

Hydrogen-rich, oxygenated, or "activated" waters. The oxygen one is worth a moment: you absorb oxygen through your lungs, not your gut, and the quantity dissolvable in water is trivial compared to a single breath.

💡 Aha moment. Water is the one nutrient that is genuinely free from a tap in most of the wealthy world. So the entire market has to be built on differentiation from free — which is why the claims are so exotic. There is no way to sell water on being water. The strangeness of the claims is a direct function of the strength of the competition, which is a public utility.

15.12b Does coffee dehydrate you?

A brief section, because it's the single most common hydration question and the answer is counterintuitive in a useful way.

The claim: caffeine is a diuretic, so caffeinated drinks don't count toward fluid intake — and may even leave you in deficit.

What's true: caffeine does have a mild acute diuretic effect, particularly at higher doses and particularly in people who don't consume it habitually.

What the evidence shows: tolerance develops. In habitual coffee and tea drinkers — which is most people who drink coffee and tea — the diuretic effect is substantially attenuated. And the studies that have measured net fluid balance from caffeinated beverages consistently find that they contribute positively: the volume of fluid in a cup of coffee comfortably exceeds any additional urine output it provokes.

🔬 Claim → Evidence → Verdict

The claim: "Coffee and tea dehydrate you — they don't count toward your fluid intake."

Where it comes from: A real acute pharmacological effect, generalized past its domain and repeated for decades.

What the evidence actually shows: Habitual consumers develop tolerance to the diuretic effect, and hydration-status studies comparing caffeinated beverages to water in regular consumers have generally found no meaningful difference in markers of hydration. The fluid in the drink more than covers any additional loss.

📉 Evidence quality: Multiple controlled studies measuring hydration markers and fluid balance; consistent.

Verdict: 🟠 Probably false. Coffee and tea count toward your fluid intake. (Two caveats worth keeping: very high caffeine doses in non-habitual consumers do increase output more; and tea and coffee with meals inhibit non-heme iron absorption — Chapter 14 §14.2, which is a genuine reason to move them, just not a hydration one.)

15.13 What to actually do

1. Drink to thirst. For healthy adults under ordinary conditions this is sufficient and is what your physiology is for.

2. Check urine colour occasionally — pale straw is the target. ⚠️ Remember riboflavin makes this uninterpretable.

3. Don't count glasses. The number was a misquotation from 1945.

4. Drink more deliberately when the situation demands it — heat, exercise, illness, altitude, and if you're an older adult where thirst is less reliable.

5. If you train: measure your sweat rate once. It converts a guess into a number, and individual variation is enormous.

6. Use electrolytes when duration, heat, or losses justify it — beyond about 60–90 minutes of hard exercise, in heat, or if you're a salty sweater. Not at a desk.

7. ⚠️ Keep WHO-formula ORS sachets in the house. For gastroenteritis, especially with children or older adults. Pennies, and genuinely one of the great medical interventions.

8. ⚠️ In endurance events: drink to thirst, avoid NSAIDs, weigh yourself, and know that gaining weight during the event is the warning sign.

🧾 Cost check. Hydration is the one nutrient where the correct answer is free from a tap, so the annual spend is worth seeing laid out:

The item with the strongest evidence in this chapter costs about eight dollars a year and sits in a cupboard unopened until someone gets gastroenteritis. Everything above it on the spending list has weaker evidence than everything below it — which is Chapter 11's rule, appearing for the fourth time.

What we don't know

The dehydration-and-performance threshold is genuinely contested. Whether the classic 2% figure holds under blinded conditions, and whether drinking to thirst is adequate for most athletes in most conditions, remains an active argument among sports scientists — with real implications, since the prescribed-drinking approach carries the §15.10 risk.

And the mild-dehydration-and-cognition literature is too methodologically weak to settle, primarily because of the blinding problem. We may not get a clean answer, because you cannot easily blind someone to whether they've had a drink.

Spaced Review

Answer before reading on.

1. (Chapter 13) The 8 × 8 rule and the "most people don't get enough vitamin X" claim share a structural flaw. What is it?

Both are artefacts of a number being detached from its qualifying context. The 1945 recommendation said 1 mL/calorie and that most of it comes from food — the second clause was dropped. The "deficiency" claim compares intakes to the RDA, a figure explicitly set to exceed what 97–98% of people need. In both cases the number is real and the framing around it was removed.

2. (Chapter 3) Why is alkaline water's central claim implausible?

The stomach maintains a pH of roughly 1.5–3.5 (§3.3), which neutralizes alkaline water essentially on arrival — and blood pH is regulated within a very tight range by the lungs and kidneys, not by intake. You cannot drink your way to a different blood pH, and if you could, it would be a medical emergency.

3. (Chapter 14) Sweat sodium losses vary between individuals by roughly an order of magnitude. Why does that make population hydration advice particularly weak?

Because the quantity being replaced differs tenfold between two people doing the same session — so any single prescription is substantially wrong for most of the people following it. It's the same argument as Chapter 14's mineral absorption variability: the between-person spread exceeds the effect the advice is trying to produce.

Project Checkpoint: Your Hydration Self-Check

Component fifteen. Fifteen minutes, and unusually, the likely output is "you're fine."

Step 1 — Urine colour, for three days. Note it at two or three points each day. Pale straw to light yellow is the target. ⚠️ If you take a B-complex or multivitamin, riboflavin will make this uninterpretable — note that and rely on thirst instead.

Mostly pale / mostly dark / uninterpretable: __

Step 2 — Thirst check. For three days, note when you feel thirsty and whether you act on it. Are you ignoring it, or are you drinking without it? Both are informative.

Step 3 — The situational audit. Do any of these apply to you?

• [ ] I'm over 65 (thirst is less reliable)
• [ ] I work in heat, or exercise in heat
• [ ] I train for more than 60–90 minutes at a time
• [ ] I'm pregnant or breastfeeding
• [ ] I take diuretics, lithium, or an SGLT2 inhibitor
• [ ] ⚠️ I have been given a fluid restriction by a clinician
• [ ] I have a job where I genuinely can't stop to drink (Camila)

Any box ticked = your answer is not "drink to thirst and forget it." The last box overrides everything else in this chapter.

Step 4 — If you train, measure your sweat rate once. §15.8. Weigh before, record fluid, weigh after. One session, one number, and it's yours rather than a population's.

My sweat rate: ______ L/hr, in ______ conditions

Step 5 — The household check. ⚠️ Do you have WHO-formula ORS sachets in the cupboard? If you have children, or care for an older adult, this is the single most useful item in this chapter and it costs almost nothing.

Step 6 — Cost the market. If you buy electrolyte products, bottled water, or hydration supplements — total the annual spend. Then ask which of §15.9's justifying conditions actually applies to you.

Non-tracking alternative. Skip Steps 1 and 2 entirely. Do Steps 3, 5 and 6 — the situational audit, the ORS check, and the cost. Those three contain nearly all the actionable value, and for most readers the honest conclusion of this checkpoint is that their hydration was never the problem.

Next checkpoint (Chapter 16): your supplement audit — every bottle you own, evidence grade, annual cost, keep or cut. This is the one that saves people the most money.**

Chapter Summary

Water balance is regulated with precision. Osmoreceptors detect 1–2% changes in plasma osmolality → ADH plus thirst. Healthy kidneys clear roughly 0.8–1 L/hour.

Under-drinking has an alarm — thirst. Over-drinking has none. Advice that overrides thirst is overriding the only signal in the system.

"Eight glasses" traces to a 1945 recommendation of ~1 mL/calorie — which explicitly noted that most of it comes from food. That clause was dropped. Valtin's 2002 review found no evidence base.

The AI is ~3.7 L (men) / ~2.7 L (women) of TOTAL water — all beverages plus food, which contributes roughly 20%. And it's an AI, derived from observed intakes, not a measured requirement.

How to tell: urine colour (pale straw; ⚠️ riboflavin confounds it) · thirst · morning weight for athletes. Not counting glasses.

Sweat rate: (weight before − weight after) + fluid consumed. Typical 0.5–2.0 L/hr, varying by an order of magnitude between people — which is why population advice is so weak here.

⚠️ Hyponatremia — dilutional low sodium, causing cerebral oedema. Risk factors invert expectations: slower finishers, smaller body size, female, longer events, abundant aid stations, NSAIDs, and being conscientious about drinking. Mild symptoms mimic dehydration, and the instinctive response makes it worse. ⚠️ Weight gain during an event is the distinguishing sign.

This chapter's verdicts:

Cost, for the fourth time in this book: tap water ~$0.50/year · **ORS ~$8/year, and it has the strongest evidence here · bottled water $500–$1,100 · daily electrolytes $220–$730 · premium "functional" waters up to $2,000. The spending runs precisely inverse to the evidence.**

The one thing to remember: "Every single thing I did wrong, I did because someone told me to do it." In this chapter, unusually, the danger is in following the advice too well.

What's Next

Chapter 16 closes Part III with the $50 billion industry that Parts I through III have been circling.

DSHEA 1994 and why supplements are regulated more like food than like drugs — and what that means for whether the bottle contains what the label says. Third-party certification. The short list that genuinely works: creatine, caffeine, vitamin D in the deficient, B12, folic acid, iron in the iron-deficient. The long list that doesn't. And the shorter, more important list of what's actively dangerous.

And Walt's cabinet, finally, in full — including the emergency room visit where 3,000 micrograms of biotin very nearly cost him a correct diagnosis of a cardiac event.