Friday, September 25, 2026 · 4 min read
The Eight Glasses Rule Came From A Sentence Nobody Finished Reading
Where the number came from, why thirst is usually enough, and the two situations where drinking more water genuinely helps.
|
Eight glasses a day. Everybody knows it. Nobody knows where it came from, which is odd for a rule this famous. A physiologist went looking for the source about 25 years ago. What he found is almost funny. Let me walk you through it… The Missing Sentence Heinz Valtin spent his career at Dartmouth Medical School studying how the body handles water. In 2002 he published a review asking a simple question. Where is the evidence for eight glasses a day? He couldn't find any. His paper ran in the American Journal of Physiology. He did find a likely origin. In 1945, an American nutrition board suggested about 2.5 litres of water a day for adults. Then came the next line, which said most of that water is already in the food you eat. Somewhere along the way, that second sentence fell off. The number survived. The context didn't. Worth saying plainly that this is Valtin's reconstruction of how it happened, not a documented paper trail.
What The Official Numbers Actually Say The National Academies set figures for this in 2004. Roughly 3.7 litres of total water a day for men, and 2.7 for women. Those look alarming until you read the fine print. Total water means everything. Tea, coffee, milk, soup, and the water inside your food. About a fifth of it typically arrives on a plate rather than in a glass. Fruit and vegetables run 80 to 95 percent water. So nobody is asking you to drink 3.7 litres. That has never been the recommendation. Thirst Does The Job For a healthy adult, thirst is a good guide. Your body tracks the concentration of your blood closely and tells you when it drifts. There's one group where this gets less reliable, and it matters. Peter Phillips and colleagues deprived healthy older men of water for 24 hours, alongside younger men. The older group ended up more concentrated by every blood measure. They also drank less afterward and reported feeling less thirsty. That study ran in 1984 in the New England Journal of Medicine. So if you're older, or caring for someone who is, thirst alone tells you less. Build in a glass with each meal. You Can Drink Too Much This part gets skipped in wellness writing, and it shouldn't. Christopher Almond and colleagues tested runners at the 2002 Boston Marathon. Of 488 runners with usable blood samples, 13 percent had sodium levels below normal. A small number, around half a percent, were in the dangerous range. The main risk factor was gaining weight during the race, which means drinking more than you sweat out. Their findings ran in 2005 in the New England Journal of Medicine. It can get worse than that. In 2007, a woman died after a radio station water-drinking contest in California. Sports medicine guidance has shifted because of cases like these. An international expert panel now advises athletes to drink to thirst, rather than as much as they can manage. Your Coffee Counts Here's a myth worth retiring. Coffee does not dehydrate you. Sophie Killer and colleagues at the University of Birmingham took 50 regular coffee drinkers. Each man drank coffee for three days, then plain water for three days. Total body water and every hydration marker came out the same. Published in 2014 in the journal PLOS ONE. Caffeine does nudge your kidneys, but the fluid in the cup more than covers it. Tea counts too. Alcohol is the real exception. It switches off the hormone that tells your kidneys to hold onto water, which is why a big night leaves you parched. When Drinking More Genuinely Helps I've spent this whole issue debunking. Here's the other side, and it's solid. Loris Borghi and colleagues in Italy followed 199 people who had already passed a kidney stone. Half were told to drink enough to produce more than 2 litres of urine daily. Over five years, 12 percent of the high-water group had another stone. In the comparison group it was 27 percent. Published in 1996 in the Journal of Urology. Then urinary infections. Thomas Hooton's team studied premenopausal women who kept getting them and who drank little fluid. Adding 1.5 litres of water a day cut their infections by roughly half over a year. That trial ran in 2018 in JAMA Internal Medicine. Read those two carefully. Both studied people with a specific problem who were drinking too little to begin with. Neither one shows that a well-hydrated person benefits from more. Three Things Worth Knowing
Clean water is the second of the seven principles, and the one most people already think they've mastered. The rule they're following came from half a sentence written in 1945. The body they're following it with has a perfectly good system built in. Drink when you're thirsty. Have some soup. Stay well, |