Does Drinking Water Help You Lose Weight? A Little. Just Not How They Said.

A little. Probably. In one specific way, and not the way you have been told.

Let us start with the part that is definitely wrong, because it is the part doing all the selling.

The 30% metabolism boost is 24 calories

In 2003 researchers gave fourteen people 500 ml of water and measured them in a metabolic chamber. Metabolic rate rose 30%, starting within ten minutes and peaking around forty [1].

Thirty percent sounds enormous. So here is the base it is a percentage of: the total extra energy burned was about 100 kilojoules. Roughly 24 calories. Less than a boiled sweet.

The paper is upfront about something else, too: about 40% of that came simply from warming the water from room temperature to body temperature [1]. Not metabolism doing anything clever. Physics heating a liquid.

The authors then extrapolated: two litres a day would add around 400 kilojoules — about 96 calories [1]. That extrapolation is where the internet got on board.

Then another lab tried to find it again

Three years later a different group ran the careful version: distilled water, saline, and a sugar solution as a positive control, all in a randomised crossover.

Distilled water produced no increase in resting energy expenditure (p = 0.34). Saline, none (p = 0.33). The sugar solution lit up exactly as it should (p < 0.0001), which tells you the equipment was working fine [2].

The only thing that did anything was water chilled to 3°C — and it managed 4.5% over an hour [2]. Not thirty.

Their conclusion, verbatim: the results “cast doubt on water as a thermogenic agent for the management of obesity.”

So the ice-water trick is real and it is worth about as much as standing up slightly more often.

Later results split along lab lines. That same group, at the University of Fribourg in Switzerland, tried room-temperature water again in 2015 on 27 young adults, this time against a sham drink, the drinking version of a placebo: a rise of under 3%, no different from the sham [6]. The 2003 group, in Berlin, repeated its own test in 16 adults with overweight or obesity and reported a rise again, 24% over an hour [7]. Both measured the energy burned right after a drink, not anyone’s weight.

Now the part that might actually work

Forget metabolism. The plausible mechanism was always dumber than that: water takes up room.

Forty-eight adults aged 55 to 75, all on a reduced-calorie diet for twelve weeks. Half drank 500 ml before each meal. That group lost about 4.4 lb (2 kg) more, a 44% steeper rate of loss [3].

A second trial, 84 adults with obesity from English GP surgeries, same 500 ml dose and same twelve weeks, but a different programme — a 30-minute weight-management consultation and a 10-minute follow-up call — and a comparison group told to imagine their stomach was full before meals: 2.9 lb (1.3 kg) more lost than controls [4].

Two independent trials, both positive, both modest. That is a real signal and we are not going to talk you out of it.

Why this is Preliminary and not Supported

Three things keep it there, and they belong in front of you rather than in a footnote.

The second trial nearly vanished. Its 2.9 lb result cleared the bar researchers use to call a finding unlikely to be chance. Then they accounted for differences between the two groups — age, sex, ethnicity and how deprived their neighbourhood was — and the result slipped to just the wrong side of that bar [4].

It did not become nothing. It became a result you would not build on. The authors’ own word for their finding is “preliminary,” and we are taking them at it.

The mechanism was not confirmed. In the first trial, everyone twice ate an eat-as-much-as-you-like breakfast in the lab, once after 500 ml of water and once without. At the start they ate less after the water (498 calories against 541); by week twelve the gap (480 against 506) had slipped just past the line researchers use for “probably not chance” [3]. The effect showed up on the scale without the explanation showing up in the food diary.

Pooled, it blurs. In 2024 researchers reviewed eight randomised trials and pooled the seven that measured body weight. The average came out at a little under a pound lost — but pooling also tells you the range the true answer probably sits in, and that range ran from about four pounds lost to two pounds gained [5].

A range that wide, straddling zero, means the honest summary is “we cannot tell.” The trials’ results also spread far more than chance would produce: an I-squared of 78%, the share of the spread between them that luck does not explain. On the review’s own chart, five of the seven land on water’s side and two on the other, and the one result whose range clears zero is on the other side: the 303-person diet-drink trial further down this page, where the water group lost 4.2 lb (1.90 kg) less [5]. The reviewers’ own conclusion: water intake “may not significantly impact adiposity.”

One caution about that pool: it is not a clean test of water before meals. Only two of the seven weight trials asked people to drink more water — the English trial above, and one advising teenagers to drink eight cups a day. The 48-person trial was not in the weight pool at all (the review used it only for BMI, a weight-for-height score, and waist size), and the other five swapped water in for another drink. Most of that spread sat among the swap trials, which scored an I-squared of 82%: the diet-drink trial on one side of zero, the other four on the other, each of those four with a range that still includes zero. The two drink-more trials scored 0%, meaning their results differed no more than chance would produce. With two trials this imprecise, one landing a little on each side of zero, that says little about whether they agree. Together they averaged 2.1 lb (0.96 kg) lost, but their range ran from 4.9 lb (2.20 kg) lost to 0.6 lb (0.27 kg) gained, so “no effect” is still on the table there too [5].

Two decent trials say yes. Pooled, even the drink-more trials cannot rule out no effect. That is what unresolved looks like, and Preliminary is the tier for unresolved.

One finding we did not expect

In that same analysis, people who replaced sugar-sweetened drinks with water lost a little weight — 1.8 lb (0.81 kg), but the range ran from 3.7 lb (1.66 kg) lost to 0.07 lb (0.03 kg) gained, so it just missed ruling out no effect. The reviewers call it “marginally insignificant” [5].

But people who replaced artificially sweetened drinks with water were 4.0 lb (1.82 kg) heavier, with a confidence interval nowhere near zero [5].

Before anyone builds a diet-soda campaign out of that: it is two trials, and the reviewers themselves say the result “may likely be due to chance,” given “the predominant weight given to one trial” [5]. That trial, of 303 people who regularly drank diet drinks, found the diet-drink group lost 13.1 lb (5.95 kg) in twelve weeks to the water group’s 9.0 lb (4.09 kg). It was funded in full by the American Beverage Association, the trade group for US beverage companies; its one-year report says the association had no role in running or interpreting it [8][9]. The other trial, also in people who already drank diet drinks, pointed the other way: in 89 women on a 24-week weight-loss diet, the ones switched to water lost 19.4 lb (8.8 kg) on average, against 16.8 lb (7.6 kg) for those who kept their diet drinks [10]. We are reporting it because it is in the data and it cuts against the tidy story, not because we think it settles anything.

What to actually do

Drink roughly two cups of water (500 ml) before each main meal — the English trial had people drink it half an hour before. That is what both trials tested, in adults with overweight or obesity who were also trying to lose weight [3][4]. It costs nothing, and two randomised trials found a modest benefit. One caution: the first trial left out anyone with heart, lung or kidney disease, so if a doctor has told you to limit fluids, ask before adding that much water [3].

Do not expect it to raise your metabolism, because the best attempt to reproduce that finding could not. Do not count the 96 calories — that number is an extrapolation from fourteen people, and the lab that went back to test it could not find the rise, in 2006 or again in 2015 [2][6].

And if a diet soda is what keeps you off something worse, the two trials split: they give you no clear weight reason to swap it for water, or to keep it.

Water is not a fat-loss agent. It is a way of arriving at dinner slightly less hungry. Sold honestly, that is a perfectly good reason to have a glass.