.png)
The Definitive Guide to How Much Water You Should Drink
Somewhere greater than zero and less than 20 liters*
It’s just about the simplest, safest medical advice you can give: Drink more water." You got a headache? Drink more water. Tired? Drink more water. Cold coming on? Drink more water. Tom Brady famously ascribed his QB longevity to water drinking, among some other less… usual activities.
I’m a nephrologist—a kidney doctor. I think about water all the time. I can tell you how your brain recognizes how much water is in your body and precisely how it conveys that information to your kidneys to govern how dilute your urine is. I can explain the extraordinary capacity of the kidney to concentrate urine over a range from 50 mosm/L to 1200 mosm/L—and the physiology that makes it all happen.
But I can’t really tell you how much water you’re meant to drink.
And trust me, I am asked all the time.
There are a number of things I’m confident of when it comes to water. You need to drink some.
Though certain animals, like kangaroo rats, can acquire practically all the water they require from the food they consume, humans are not such creatures. Without water, we perish.
I’m also confident that you can die from drinking too much water. Drinking large quantities of water dilutes the salt in your blood, which interferes with the electrical circuitry in your brain and heart. I really had a patient go on a “water cleanse” and cause herself a seizure.
But, to be fair, presuming your kidneys are operating relatively well, and you’re generally healthy, you’d need to drink roughly 20 liters of water a day to get into fatal jeopardy. The dosage is the poison, as they say.
Ok, so somewhere greater than zero and less than 20 liters of water is the quantity you should be drinking a day. That much I’m sure of.
But the research on where in that range you should target is really fairly limited. You wouldn’t think so if you glanced at the wellness influencers online—with their Stanleys and stringent water consumption regimes. You’d think the evidence that drinking additional water is a benefit is overwhelming.
The respected National Academy of Medicine recommended males drink 13 8-oz cups a day (that’s approximately 3 liters) and women drink 9 8-oz cups a day (a touch more than 2 liters). From what I can determine, this recommendation—like the previous “8 cups of water per day” recommendation—is taken out of thin air.
I’m not proposing that we shouldn’t drink water. Of course, water is vital. I’m simply curious what research there is to genuinely indicate that drinking more water is beneficial.
Fortunately, a team from UCSF has finally done the homework for us. They lay out the real evidence in this report, published in JAMA Network Open.
Source: Hakam et al. JAMA Network Open 2024.
The researchers reviewed the medical literature for randomized controlled studies of water consumption. This is critical—we don’t want testimonials about how clean your skin became after you upped your water consumption. We want freezing cold, clear data. Randomized trials select a group of individuals and, at random, assign some to conduct the intervention—in this instance drinking more water—and others to… well, not. Or in this situation, to keep doing what they’d typically do.
The team analyzed approximately 1500 publications, but only 18 (!) fulfilled the tough requirements to be included in the study, as you can see from this flow chart.
Source: Hakam et al. JAMA Network Open 2024.
This is really the first key finding—there is not enough high-quality research that studies how much water we should drink. Of all, water isn’t some prescription product—so money is likely hard to come by. Can we undertake a trial with Dasani?
In any event, these 18 experiments were all looking at distinct outcomes of interest. Four looked into the influence of drinking more water on weight loss, two on fasting blood glucose, two on headache, two on urinary tract infection, two on kidney stones, and six on various other outcomes. None of the studies looked at energy, or skin tone, or general well-being, though one did assess a quality of life score.
And if I could sum up all these studies in a word, that word would be "meh.”.
For weight reduction—one of the four trials revealed that increasing water consumption had no impact on weight loss. Two trials indicated a benefit, but drinking additional water was paired with a low-calorie diet, so that seems a little like cheating to me. One research randomized people to drink half a liter of water before meals, and that group did experience higher weight reduction than the control group—about a kilogram more over 12 weeks. That’s not quite Ozempic.
For fasting blood glucose, one study showed increasing premeal water lowered glucose levels, whereas the other (which looked solely at increasing water overall) didn’t.
For headaches—and, cards on the table here, I’m a huge believer in water for headaches—one trial revealed nothing, and the other indicated that increasing water consumption by 1.5 liters per day enhanced migraine-related quality of life but didn’t affect the number of headache days per month.
For urinary tract infections—one favorable trial, one negative one.
The best evidence for greater water intake came from the kidney stone studies. Increasing water consumption to reach more than two liters of pee a day was related to a substantial decrease in kidney stone recurrence. I consider this a positive control more or less—you would be hard pressed to find a renal doctor who doesn’t believe that persons with a history of kidney stones should drink more water.
What about that quality of life study? 141 participants were randomized to either consume 1.5 liters of additional water per day or not. And the ratings on the quality of life assessment were no different between the two groups six months later.
Thirsty yet?
So what’s going on here? There are a few options.
First, I need to point out that clinical studies are incredibly challenging. All the studies in this evaluation were quite small, with most recruiting fewer than 100 persons. The impact of more water would need to be quite significant to detect it with so few samples.
But I can’t help but point out that our bodies are really beautifully adapted to control how much water we have in us. As we lose water during the day—from perspiration, from exhalation—our blood gets a small bit more concentrated—the salt content goes higher. Our brains notice it and generate a feeling we call thirst. Thirst is one of the strongest motivations humans have. Animals, including humans, when thirsty, will pick water over food, over drugs, and over sex. It is exceedingly hard to resist, and if we have quick access to water, there is no reason to resist it. We drink when we are thirsty. And it may be enough.
Of course, going beyond thirst is conceivable. We are sapient creatures who can drink more than we wish to. But what we can’t do, if our kidneys operate, is hang onto that water. It goes straight past us. In the case of avoiding kidney stones, this is a positive thing—putting more water in your body leads to more water flowing out—more dilute urine—which means it’s harder for stones to develop. But for all the other stuff? The well-being, the skin tone, and so on? It really doesn’t make much sense. If you drink an additional liter of water, you pee an extra liter of water. Net net? Zero.
Some persons would claim that the additional urine gets rid of excess pollutants or anything like that, but—sorry, kidney doctor Perry here again— that’s not how pee works. The removal of toxins from the circulation occurs considerably upstream of where your urine is diluted or concentrated.
If you drink more, the same poisons come out, but with more water surrounding them. In fact, one of the biggest studies in this JAMA Network Open review investigated whether increasing water intake in adults with chronic renal disease would improve kidney function. It didn’t.
I am left, therefore, with just a little more confidence than when I started. I remain positive you should drink more than zero liters and certain you should drink less than 20 (assuming you’re not losing a lot of water in some other manner, like working in the heat). Beyond that, it seems acceptable to believe in the millions of years of evolution that have made water homeostasis vital to life itself. Give yourself access to water. Drink when you’re thirsty. Drink a little extra if you’d like. But no need to push it. Your kidneys won’t allow you anyhow.
A version of this commentary was initially published on Medscape.com.
*Ok, just a small note here to emphasize that 20 liters of intake is the amount that will kill you, presuming you aren’t losing a ton of water via perspiration or whatever. Even then, replacing perspiration with simply water may lead to electrolyte imbalances and plunge you into a world of problems. Even Tom Brady doesn’t drink this much. I was merely attempting to create an inarguable upper limit. Be reasonable, people.
