51

Since I was a kid, I can't remember how many times I've heard that a healthy person needs to drink at least 8 glasses of water a day.

There are definitely plenty of reasons to remain well-hydrated. However is there any scientific evidence of harmful effects to those who don't drink at least the "recommended" 8 glasses of water a day?

Are there any documented and verifiable benefits for those who do consume the 8 glasses a day, as compared to those who do not?

For example, in a blinded study, could one group be told from the other based on objective criteria?

Or is this a myth?

And if so how did it start?

Improve this question
6
  • 1
    If you eat only fresh fruit and vegetables, then you don't need to drink any water at all because you are getting all the liquid you need out of them.
    – Bogdan0x400
    Apr 19, 2011 at 7:42
  • 1
    i never heard of that myth "8 glasses", you just have to replenish all the liquid you lose. If you sport, that needs to be alot more.
    – Terry
    Apr 19, 2011 at 9:21
  • The body needs about 2 liters of water a day (8 glasses) under normal circumstances. Extreme physical effort or extreme environmental circumstances (heat, etc) may mean you need more. However, the body does not care where that water comes from, whether it's coffee, beer, or the juices in fruit, meat or vegetables. Basically, you need to drink when you feel thirsty. Also remember that drinking too much water can kill you, as some athletes have found.
    – herman
    Apr 19, 2011 at 11:08
  • Here's an example of the opposite -- too much water kills a runner in marathon dailymail.co.uk/news/article-511475/…
    – Dan
    Apr 19, 2011 at 12:01
  • 1
    Can't be arsed to scour medical literature at this hour, so posting as comments: many commenters suggesting variations on "you only need to replenish what you lose" seem to overlook the crucial role that water plays in the "waste filtration" system that is your kidney. The more water (to some extent), the less stress is put on the kidneys (by diluting the waste).
    – Dave
    Jul 9, 2011 at 1:19

2 Answers 2

Reset to default
36

TL;DR: drink when you're thirsty. There's no health reason to set an arbitrary goal.

Snopes tackled this issue.

Here's a 2003 CBC story that states:

University of British Columbia nutrition Prof. Susan Barr is part of a joint Canadian and American team of doctors and nutritionists who are looking at how much water people actually need.

Barr said they couldn't find any scientific evidence to support the eight to 10 glass recommendation.

The confusion may have arisen because a typical adult's energy requirements call for two to three litres of fluid – but it doesn't all have to be in the form of glasses of water. All foods and non-alcoholic drinks count toward the goal.

According to Heinz Valtin, a Dartmouth Medical School physician:

The notion may have started in 1945 when the Food and Nutrition Board of the National Research Council recommended approximately “1 milliliter of water for each calorie of food,” which would amount to roughly 2 to 2.5 quarts per day (64 to 80 ounces).

In its next sentence the board stated, “[M]ost of this quantity is contained in prepared foods.” But that last sentence seems to have been missed, so that the recommendation was erroneously interpreted as how much water a person should drink each day.

He goes on to say:

Under some circumstances, significant fluid intake — at least eight 8-ounce glasses — is advisable: for the treatment or prevention of kidney stones, for example, as well as under special circumstances, such as performing strenuous physical activity or enduring hot weather.

However, most people currently are drinking enough water and, in some cases, more than enough. There is potential harm in drinking too much water (Hale, 2010). Water intoxication, a life-threatening condition, can occur when one drinks excessive amounts of water.

Water intoxication occurs when the kidneys are unable to excrete enough water (as urine), which leads to dilution of blood sodium. Mental confusion and death can result.

Improve this answer
5
  • 8
    a tour guide I once had in an arid region, a former military man, put it like this: if you don't need to piss, you need to drink. Of course in a less arid region the body requires less intake of fluids, but the general advise to not let yourself get dehydrated stands :)
    – jwenting
    Apr 19, 2011 at 7:17
  • 7
    Just to clarify, there is a HUGE difference between the 8 glasses of water a day and the volume required to induce hyponatremia and "water intoxication", unless of course you've just run a marathon or you're an infant.
    – Monkey Tuesday
    Apr 19, 2011 at 7:45
  • true. But those 8 glasses on top of what someone already drinks in other liquids might put you over the top. Say your average teen on a drinking binge drinks a gallon or more of beer in a few hours, he's already "predrunk" a liter of booze to "get happy" before starting the binge, and that after drinking his normal 3 liters of cola during the day. Now add 8 glasses, or 2 liters of water to that and you're up to a LOT of liquids sloshing around those intestines. Of course anyone drinking that much alcohol will have more serious problems :)
    – jwenting
    Apr 19, 2011 at 9:28
  • @jwenting: Your argument makes a certain kind of sense, but you are forgetting to allow for kidney function and urination which do a great job at regulating sodium levels. Also, there is a sodium content of about 10mg per bottle of beer, so it will not reduce the sodium levels in the blood the same way that drinking water does.
    – Monkey Tuesday
    Apr 19, 2011 at 19:20
  • 1
    The last sentence, "drink when you're thirsty. There's no health reason to set an arbitrary goal," isn't the logical conclusion from the rest of the answer. Sure, the 8-10 glasses per day is well refuted, but nowhere is it stated that you shouldn't drink if you aren't thirsty. Surely many people's thirst response is not as powerful or fast as the average person.
    – travisbartley
    Aug 8, 2013 at 4:23
-1

Use common sense:

How much water does the human body need? Depends. A 200 pounds man needs more than a 50 pounds child, obviously. If the temperature is high or the person is doing an intensive workout, water intake has to increase proportionally to make up for losses.

Without going any further, you know already that it is impossible to set a fixed amount of water intake that will work for every one in all situations.

From there, you can determine beyond doubt that the "8 glasses" recommendation can not be anything more than a guideline, and not a good one at that.

As many individual factors come into play (including body type, climate, activity, water content of food, etc.), establishing a standard amount of water intake as being good for everyone everywhere is idiotic.

Thirst AND hunger make for a much better rule of thumb. Note that I say "hunger" as well, as empirically, you will often find that a sweet tooth craving can be satisfied with water.

PS: for those who abdicate any right or ability to follow a line of reasoning without having to refer to an authority, here are a list of references:

http://www.mayoclinic.com/health/water/NU00283

http://www.wikihow.com/Drink-More-Water-Every-Day

http://www.guide2.com.au/mind-body/recommended-daily-water-intake/13/69

http://www.bbc.co.uk/health/treatments/healthy_living/nutrition/healthy_water.shtml

...

and of course, the obligatory reference to Wikipedia:

http://en.wikipedia.org/wiki/Drinking_water

They are probably quite relevant to the mater, yet strictly unnecessary to answer the question. But, hey, if you need somebody else's approval to think, go ahead.

Skepticism reduced to reference hunting... Oh, sweet irony.

Skepticism was funded on the observation that truth is not an absolute and that authorities and belief could not be relied on.

And there we are on "skeptics", running Google searches and copying other people's supposed research, because god forbid we would be using our reasoning powers and already available data to come up with even the most simple conclusions.

If that's right, then perhaps I don't belong here.

Improve this answer
5
  • 4
    This answer is not properly referenced. Please add citations to support your claims! :-)
    – Sklivvz
    Apr 19, 2011 at 8:15
  • I've decided to remove the whole discussion about the need for references. If you want to talk about that, please do it on meta.
    – Mad Scientist ♦
    Apr 19, 2011 at 10:42
  • I think you're being a little too cynical. You should check out what's going on hereto see that we do answer questions here by doing our own research, verifying primary sources,contacting those who make the claims, and doing the tedious background work like scouring public records. If you read this whole post, you will definitely not find links to wikipedia or anyone "copying other people's supposed research".Cheers.
    – Monkey Tuesday
    Apr 22, 2011 at 4:53
  • @Monkey Tuesday: It was partly in response to a comment which has since been removed, and partly out of frustration at some of the attitudes I found on here. It's a little overstated, but if it can lead a few readers to question the wisdom of trusting references over logic and observation, it serves a purpose.
    – Sylverdrag
    May 1, 2011 at 17:23
  • 3
    I understand. As skeptics, it's very easy to get frustrated with other people sometimes. Mostly because we are all constantly defending our views and ideas, I just hope you don't become cynical and give up. What I'm saying is, keep at it. It's not that this site gives preference to credible references over logic and observation. It's that it requires all three. Without all three, answers can be correct, but are often not complete. cHEERS
    – Monkey Tuesday
    May 2, 2011 at 0:58

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .