We are all roughly two-thirds water. It makes up about three-quarters of our lean body mass, about 10% of our fat, and in terms of the amount of time we could live without consuming it, water is the most essential of our nutrients. But exactly how much water should we be taking in each day?
You might expect science to have provided a reliable answer to such a question - water is a fundamental constituent of life, after all.
But with the highly evolved ability of the human body to regulate water so exquisitely - and with lots of individual variability in the optimum intake - there is no definitive answer for the amount of water one person or another should get each day.
In fact, the best guidance is simply to follow the natural call of the body when more fluid is needed: just follow your thirst.
Yet the question of amount persists, and there is misinformation in abundance. Vested interests come up with highly questionable ideas about how we should all be drinking more water. Even well-respected sources cite daily intake amounts that lack good scientific evidence to support them.1,2
Contents of this article:
Here are some key points about daily water consumption. More detail and supporting information is in the article.
- We must consume foods and fluids as the source of almost all of the water in our bodies
- Humans cannot live without fluid intake for more than a few days
- Lots of myths about optimal daily water intake are widely circulated, and even get repeated by official sources
- Some guidance about water intake, such as to counter dehydration during endurance exercise, would be dangerous if taken without caution against overconsumption and hyponatremia
- The intuitively plausible claim that 2% dehydration from exercise adversely affects performance may be debunked
- Concerning normal daily consumption, the often-cited recommendation that we should drink 8 glasses containing 8 ounces of water a day is not based on evidence
- No set volume can be provided reliably because a recommended amount cannot match the wide variety of individual fluid needs and differences in water loss
- There are times when we do need to be wary of the risk of dehydration, such as during illness
- The healthy body is masterful at maintaining an exquisitely well-tuned water balance against the amount we consume and lose - and the best way it lets us know when we need more, or this need has been quenched, is through thirst.
How much water is in the human body?6,7
The principal chemical making up the human body is water (H2O), roughly comprising two-thirds of the body because humans show considerable variability in body composition.
The average young man has a percentage water composition anywhere between 50-70% of their body weight.
A similar range is seen between early and later years - infants have 75% of their body weight accounted for by water, whereas the proportion in older people is just 55%.
Variability in the overall water composition of the body is mostly due to differences explained by age, sex and aerobic fitness. These affect the ratio between:
- Lean body mass, which is about 73% water, and
- Fat body mass, which is about 10% water.
How is water regulated by the body?6-8
Without water, there can be no life at all, let alone human life. From the earliest origins of life on earth to continued survival today, organisms have adapted to avoid dehydration.
Healthy humans can survive only a matter of days without water intake, and water loss through illness that does not get replaced can quickly prove dangerous in vulnerable people such as the very young or old.
We have evolved fantastically effective physiological mechanisms for maintaining our bodies' fluid homeostasis (water balance). The two main mechanisms for maintaining water balance are:
- Thirst - this tells us when we need to take in more fluid
- Urine output - the kidneys regulate the surplus, or deficit, of the water we consume by either emptying it into the urinary bladder or holding onto it in the blood plasma.
The kidneys also regulate the balance of electrolytes such as sodium and potassium in the body fluids. The kidneys receive hormonal signals to conserve or release water into the urine following the brain's detection of alterations in the concentration of the solutes in the blood (changes in the plasma osmolality), via tiny changes in cell size according to the amount of water inside versus outside cells.
The brain's response to plasma osmolality is also partly responsible for the sensation of thirst that drives us to replace lost water.
Water is lost by other means aside from the action of the kidneys. Total water loss from the body is accounted for by the following means, each of which shows a range of variability:
- Urinary - water regulation by the kidneys, with excess fluid excreted into the urine being estimated at 500-1,000 mL per day
- Respiratory - water lost from the lungs as we breathe out, about 250-350 mL per day
- Fecal - excreted in our solid waste, estimated at 100-200 mL per day
- Other "insensible water loss" - other unnoticed water loss, especiallyvia evaporation from the skin, which increases with sweating. Sedentary loss is somewhere between 1,300-3,450 mL a day but can range from 1,550-6,730 mL a day with sweating due to physical activity.
The '8 by 8' water mantra: 8 times 8oz glasses a day1-5
How much water is enough? How much is too much?
The start of the millennium saw the widespread repetition of the idea that we needed to drink at least eight glasses of water a day. However, the origins of this mantra, and scientific support for it, remain elusive, even though it is still widely cited.
Even the UK's National Health Service (NHS), through its NHS Choices website, once gave the recommendation to drink up to eight glasses of fluid a day - although it has since changed its recommendation to "plenty of water" for "quenching your thirst at any time."
The one-size-fits-all mantras of daily water intake do seem to be retreating a little, however, or are at least including the idea that other fluids aside from water can contribute to the "8 by 8" recommendation.
Other questions are currently refocusing the ideas about optimum hydration, too. Are marathon runners drinking way too much water, for example? And is there any real role in the fight against obesity to be played by drinking plain water as a substitute for sugary drinks?
How much is too much? Water intake and hyponatremia1,2,9-13
One of the major questions about water intake concerns exercise, and this is the area saturated by marketing messages from manufacturers of bottled beverages.
All shapes and sizes take part in marathons, and drinking too much water over the course of a race can lead to blood plasma becoming dangerously dilute.
Within the world of guidelines for normal daily intake in healthy adults, there are also the schools of thought on how much to drink during exercise, but such guidance is controversial.
Messages about water intake during exercise could even be proving dangerous in the case of high-endurance exercise. A study of runners taking part in the 2002 Boston Marathon, for example, estimated that almost 2,000 of the participants would have had some degree of hyponatremia (abnormally low levels of sodium in the blood).
The study also found that some 90 finishers in that event might have had critical hyponatremia. These abnormally and potentially dangerously low sodium levels in the blood plasma at the end of the race were put down to excessive fluid consumption, as evidenced by weight gain while running.
Hyponatremia is a real danger whether too much fluid is taken in during exercise, or even as a result of the "eight glasses a day" guidance for regular intake.
A critique of evidence behind such messages - written by Dr. Heinz Valtin, emeritus professor of physiology and neurobiology at Dartmouth Medical School in Hanover, NH, and published in the American Journal of Physiology in 2002 - concluded:
"Not only is there no scientific evidence that we need to drink that much, but the [8 by 8] recommendation could be harmful, both in precipitating potentially dangerous hyponatremia and exposure to pollutants, and also in making many people feel guilty for not drinking enough."
In the field of sports science, there is a well-established belief that a certain level of dehydration during exercise is bad for sports performance, but even that message has been called into question by recent research.
Guidance from the American College of Sports Medicine, published in 2007, recommends an individual estimation of the fluid replacement needed by people taking part in exercise, to avoid dehydration. But controversy lies in the question of exactly how much dehydration is detrimental to performance.
It may be a myth that 2% dehydration during exercise has any adverse effect on performance.
The ACSM is following advice in the position statement that bodyweight should not drop by more than the threshold of 2% as a result of dehydration during exercise:
"The goal of drinking during exercise is to prevent excessive (>2% body weight loss from water deficit) dehydration and excessive changes in electrolyte balance to avert compromised performance."
Other respected bodies following this recommendation have included the International Olympic Committee and two North American associations of dietitians.
This 2% idea is based on a belief that exercise performance is affected by this level of dehydration. But one carefully controlled study published in the British Journal of Sports Medicine in 2013 dismisses the premise of this guidance.
The study concludes that hypohydration of up 3% has "no effect" on real-world sports performance (and the experiment was a cycling time-trial over 25 kilometers in hot conditions of 33 °C and 40% relative humidity).
The research involved well-trained participants being blinded as to their hydration status. This was so that there could be no influence on their performance measures due to a placebo effect from drinking more fluid. Instead, they were hydrated at various levels via an intravenous drip, which also ironed out individual gut effects on water uptake.
Performance, physiologic and perceptual variables were the same between the groups tested.
This was, the authors remind us, because of "the body's rapid defense of its plasma and blood volume following dehydration" - in other words, that highly evolved, exquisite capacity for the healthy body to tightly regulate water balance.
A study involving more than 18,300 adults in the US taking part in the National Health and Nutrition Examination Survey 2005-2012 published results in March 2016 showing that increased total water consumption correlated with lowered total energy intake.
This study, published in Diabetologia in March 2015, concluded that "reducing consumption of sweet beverages...and promoting drinking water and unsweetened tea or coffee as alternatives may help curb the escalating diabetes epidemic."
"Water preloading before main meals" was tested as a strategy for weight loss in a small study of people with obesity, published in August 2015.
A study in JAMA Pediatrics in January 2016 concluded that providing access to drinking water in schools could be a low-cost way of delivering a small reduction in children's bodyweight.
How are human water needs calculated?6,7,14
The amount of water we need to take in to maintain a normal balance in our bodies is decided by how much water we use and lose that must be replaced.
The body responds to any deficit or excess that results from an imbalance between these two of even just a couple hundred milliliters. Over the course of 24 hours, healthy resting adults regulate their water balance to within around 0.2% of body weight.
But measuring the amount of water used or lost by the body is difficult to do, and there is a great range of differences in the amounts that have been measured across groups of people taking part in studies.
People also range greatly in the amounts of water they use and lose according to various conditions and activities - sweat loss, for example, increases with heat and activity.
And only one study has been conducted to measure the average loss of water in the feces.
Guidance on recommended intakes, as given by the non-profit organization the Institute of Medicine among others, has relied only on survey results of the average amounts consumed from all dietary sources.
But this is not a scientific measure of how much water we need, just a measure of what people generally get. It does not reflect how well hydrated, or over- or under-hydrated, people are against their daily intakes.
Rather, official guidance simply assumes that population surveys of fluid intake produce a figure that matches or exceeds our needs. The problem is that we could be greatly exceeding our needs.
Recommended daily water intake6,7,14-18
Food provides about a quarter of our daily water intake.
The "adequate intakes" recommended for total water from all sources each day (for adults between 19-30 years of age) are:
- 3.7 liters (or about 130 fl oz) for men
- 2.7 liters (about 95 fl oz) for women.
These dietary reference intakes, however, are based only on survey results of the average amounts that are consumed by people, on the assumption that these amounts must be about right for optimal hydration.
But the amounts measured for people in the temperate climate of the US, for example, with plenty of access to water, may be too high, and intakes do vary greatly according to activity, environmental conditions (including clothing) and even social activities.
The recommended amounts are of limited value for another reason - total intake figures fail to give a breakdown of how much can be divided up between different kinds of food and beverage.
Consequently, the common sense guidance that has gained ground goes along the lines of this summary from the an official source, the US National Library of Medicine:
"If you drink fluids when you feel thirsty and have beverages with meals, you should get enough water to keep you hydrated."
Dietary sources of water intake
Intake from plain water can, of course, come from the tap or bottles, but water is available from other beverages and solid foods, too.
Again, the estimates for the proportions of fluid we obtain from fluids versus foods are based on surveys of average diets, but the individual variety is huge.
One survey of an adult population sample in the US in the late 1970s found that total water intake was made up of 28% from foods, with the same proportion coming from drinking water and the remaining 44% from other drinks. Other surveys give a lower figure of about 20% of water intake coming from foodstuffs, with 80% obtained from fluids.
The water content figures for different foods and fluids are much more consistent than the proportions of consumption. Water content ranges along the spectrum of the following examples:
Coffee is popularly thought of as dehydrating, whereas scientific evidence is to the contrary.
- 90-99% water: fat-free milk, juicy fruits such as strawberries, vegetables such as lettuce, celery and spinach
- 80-89%: fruit juice, yogurt, fruits such as apples, pears and oranges, vegetables such as carrots and cooked broccoli
- 70-79%: bananas, avocados, baked potatoes, cottage and ricotta cheeses
- 60-69%: pasta, beans and peas, fish such as salmon, chicken breasts, ice cream
- 30-39%: bread, bagels, cheddar cheese
- 1-9%: nuts, chocolate cookies, crackers, cereals
- 0%: oils, sugars.
Water content does not directly translate into hydrating 'power,' however. Milk, for example, is more effective as a source of hydrating fluid than plain water, even though it obviously has a lower water content.
Caffeinated drinks are widely mistaken as being poor at hydrating us because of a belief that they have a diuretic effect on our water balance.
A number of studies to properly test the effect of caffeinated fluids on hydration have shown that tea and even coffee are in fact good sources of water and do not dehydrate us. "Advising people to disregard caffeinated beverages as part of the daily fluid intake is not substantiated," says one study, while another concludes that there is "no evidence of dehydration with moderate daily coffee intake."
Maintaining safe hydration during exercise10
The messages of sports drinks makers - and even of respected bodies such as the American College of Sports Medicine in their 2007 guidance - could have people serious about exercise worrying about drinking lots of fluids to beat dehydration during physical activity.
The truth, however, is that there is a relatively simple way to ensure that fluid replacement strategies are sufficient and not over-the-top: just follow thirst.
The study about the runners in the 2002 Boston Marathon, for example, produces the following conclusion in the paper published in the New England Journal of Medicine:
"Because runners vary considerably in size and in rates of perspiration, general recommendations regarding specific volumes of fluids and frequencies of intake are probably unsafe and have been superseded by recommendations favoring thirst or individual perspiration rates as a primary guide."
Measuring the amount of fluid lost as sweat during exercise - as a guide for how much fluid to take under similar conditions - is simple: take a reading of body weight before and after exercise. The difference in grams is the same as the amount of fluid lost in milliliters.
Likewise, this is also a reliable measure of whether too much fluid is being taken in during exercise, because a gain in weight signals too much consumption versus the amount being lost.
This individualized approach would avoid hyponatremia caused by taking in too much water during endurance running. The development of dangerously low blood plasma osmolality varies between individuals and so it is not safe to recommend a fixed volume of fluid intake for everyone.
As the study authors explain, runners weighing themselves before and after training is best because, for example:
"Smaller runners may drink larger volumes of fluids in proportion to their size than larger runners.
"Conversely, in proportion to their size, larger runners may lose less free water than smaller runners through evaporation (by means of sweat), as a result of a lower ratio of surface area to volume."
In this study, reported by researchers in a letter to the journal Nature in January 2016, wearable tech was able to detect a peak in sodium levels in sweat excreted during prolonged exercise without water intake - thereby signaling dehydration.