It turns out the often quoted “drink at least eight glasses of water a day” dictum has little underpinning scientific evidence. The recommendation was traced back to a 1921 paper, in which the author measured his own pee and sweat and determined we lose about 3% of our body weight in water a day, about eight cups. Consequently, for the longest time, water requirement guidelines for humanity were based on just one person.
There is evidence that not drinking enough may be associated with falls and fractures, heat stroke, heart disease, lung disorders, kidney disease, kidney stones, bladder and colon cancer, urinary tract infections, constipation, dry mouth, cavities, decreased immune function and cataract formation. The problem with many of these studies is that low water intake is associated with several unhealthy behaviors, such as low fruit and vegetable intake, more fast-food, and less shopping at farmers markets. And who drinks lots of water? People who exercise a lot. No wonder they have lower disease rates!
Only large and expensive randomized trials could settle these questions definitively. Given that water cannot be patented, such trials seem unlikely. Who’s going to pay for them? So we’re left with studies that find an association between disease and low water intake. But are people sick because they drink less or are they drinking less because they’re sick? There have been a few large prospective studies in which fluid intake is measured before disease develops. For example, a Harvard study of 48,000 men found that the risk of bladder cancer decreased by 7% for every extra daily cup of fluid one drinks. Therefore, a high intake of water—like eight cups a day—may reduce the risk of bladder cancer by about 50% (eight cups times 7% per cup), potentially saving thousands of lives.
Probably the best evidence we have for a cut off of water intake comes from theAdventist Health Study, in which 20,000 men and women were studied. About one-half were vegetarian, so they were also getting extra water by eating more fruits and vegetables. Those drinking five or more glasses of water a day had about half the risk of dying from heart disease compared to those who drank two or fewer glasses a day. Like the Harvard study, this protection was found after controlling for other factors such as diet and exercise. These data suggest that it was the water itself that was decreasing risk, perhaps by lowering blood viscosity (blood thickness).
Based on all the best evidence to date, authorities from Europe, the U.S. Institute of Medicine, and the World Health Organization recommend between 2.0 and 2.7 liters (8 to 11 cups) of water a day for women, and 2.5 to 3.7 liters (10 to 15 cups) a day for men. This includes water from all sources, not just beverages. We get about a liter from food and the water our body makes. So this translates into a recommendation for women to drink four to seven cups of water a day and men 6 to 11 cups, assuming only moderate physical activity at moderate ambient temperatures.
We can also get water from all the other drinks we consume, including caffeinated drinks, with the exception of stronger alcoholic drinks like wines and spirits. Beer can leave you with more water than you started with, but wine actively dehydrates you. However, in the cancer and heart disease studies I mentioned above, the benefits were only found with increased water consumption, not other beverages.