Drinking eight cups or two litres of water a day is longstanding advice. But is there any scientific basis for it, asks Dr Chris van Tulleken.
You know those ads that remind us that even a small drop in hydration levels can massively affect performance so you need to keep hydrated with whatever brand of isotonic super drink they're selling?
They seem pretty scientific don't they? Man in white coat, athlete with electrodes attached and so on. And it's not a hard sell because drinking feels right - you're hot and sweating so surely replacing that fluid must be beneficial.
Well earlier this year sports scientists in Australia did an extraordinary experiment that had never been done before (British Journal of Sports Medicine, September 2013, Current hydration guidelines are erroneous: dehydration does not impair exercise performance in the heat, Wall BA).
This group wanted to find out what happened to performance after dehydration. So they took a group of cyclists and exercised them until they lost 3% of their total body weight in sweat.
Then their performance was assessed after rehydration with either 1) nothing, 2) enough water to bring them back to 2% dehydration or 3) after full rehydration.
So far nothing unusual, but the difference between this and almost every other study that's ever been done on hydration was that the cyclists were blind to how much water they got. The fluid was given intravenously without them knowing the volume.
This is vital because we all, and especially athletes, have such an intimate psychological relationship with water consumption.
Remarkably, there was no performance difference between those that were fully rehydrated and those that got nothing. This study was part of a growing movement to "drink to thirst" which hopes to persuade athletes not to over hydrate with the potentially fatal consequence of diluting your sodium level, causing hyponatraemia.
Perhaps the result shouldn't be so surprising. Humans evolved doing intense exercise in extreme heat and dryness. We are able to tolerate losses in water relatively well whereas even slight over hydration can be far more dangerous. In simple terms, being too watery is as bad for you as being too concentrated.
But what about the rest of us who aren't cycling around the desert in Western Australia?
There is a very well accepted idea that we should drink about eight cups of water per day (two to three litres) in addition to our food and other drinks.
We are awash with positive messages about the healing properties of water and how it will improve everything from our brains to our bowels. And we know that without it we will die in days.
It's a short leap of logic to think that if a lack of water is bad for for you then hydration must be good - purifying, cleansing water washing through your organs must be beneficial, detoxifying. It surely improves your skin, helps you think, reduces your risk of kidney stones and turns your urine a lovely light, straw/champagne colour rather than the fetid orange syrup you produce at the end of a long day where you haven't had time to drink.
So I've looked through the literature and I found a review article saying all of this and more. It was written by a group of respected physicians from American and French hospitals and it clearly supported the widely held belief that you should drink two to three litres of water a day.
It said that people with a high urine output have a lower rate of kidney stone disease, that the flushing action of the water may reduce the risk of a urinary tract infection (especially in women after sex). Perhaps most importantly, they referenced a surprising study which showed that paradoxically an increased intake of water increased the risk of bladder cancer. But only tap water. And there's the clue.
A footnote at the end of the article explained that what you thought was a scientific article in a scientific journal is in fact a supplement, sponsored by a major mineral water manufacturer. All of the authors received honoraria from this company, which also provided medical writing assistance. So this isn't research, it's marketing.
And this is one of the reasons we're even discussing this - because increasingly drinking water doesn't just come out of our taps for free. It's sold to us by the same clever people that sell us yoghurts with bacteria in them that probably don't do us much good, something I look at separately in the television series I've been making. And these companies pretty consistently recommend two to three litres of water per day.
So where did that number come from and is there any reason to think it correct?
Well the grain of truth is this - people in temperate climates who are not doing sustained physical exercise do need around six to eight cups per day but that can be contained in food, alcohol or caffeinated beverages.
Yes, beer and coffee do not dehydrate you to any noticeable extent (there's a nice paper where some medical students got to drink quite a lot of beer and had their urine studied - British Medical Journal (Clin Res Ed), December 1982, Acute biochemical responses to moderate beer drinking, Gill GV).
There is no evidence that adding the eight cups of water to everything else you drink will do you any good and it could do you harm (American Journal of Physiological - Regulatory, Integrative and Comparative Physiology, November 2002, Drink at least eight glasses of water a day. Really? Is there scientific evidence for "8x8"? Valtin H).
But the great thing is that just like a top-level athlete you don't need to worry about exactly what that total daily requirement is because your body will sort it all out for you.
If you drink too much you pee it out. If you drink too little you get thirsty and pee less. It's all exquisitely well-controlled in the same way that your intake of oxygen is well-controlled.
Saying that you should drink more water than your body asks for is like saying that you should consciously breathe more often than you feel like because if a little oxygen is good for you then more must be better.
Like most things in life there's a Goldilocks amount - not too little and not too much. With this in mind, next week I'll deal with the health benefits of porridge and how to avoid being eaten by bears.