Increased fluid intake is the most consistently supported measure in kidney stone prevention — a 2026 systematic review of 31 studies lists it first. The familiar target is urine output above about two litres a day. What almost nobody mentions is that this cornerstone recommendation rests on a single randomised controlled trial.
The target is urine, not water
This is the most useful correction on the page, and it changes what the advice means.
The guidance is about how much urine you produce, not how much water you drink. Those are different numbers, and the gap between them is everything you lose to sweat, breath and climate. Someone working outdoors in Arizona and someone in an air-conditioned office in Oregon can drink identically and produce very different volumes.
Which is why nobody can honestly give you a number of glasses. What they can give you is a target output — and a way to measure whether you are hitting it.
Your 24-hour urine collection measures volume directly. It is the first line on the report. If you have ever had one, you already know your answer; if you have not, no estimate from a website is a substitute for it.
What the evidence actually is
The 2026 systematic review in Annals of Internal Medicine — 31 studies, 26 of them randomised trials — lists increased water intake among the measures that may reduce recurrence in adults with calcium stones. It rates that conclusion, like almost all its conclusions, as low strength of evidence.
A 2017 review of fluids in stone disease is more specific about why. It describes maintaining urine volume above two litres a day as the long-standing cornerstone prescription, and then notes that this recommendation relies on only one randomised controlled trial, with the remaining support coming from epidemiological studies and basic science.
That is a striking thing for a measure recommended to essentially every stone former on earth.
Why it is still the first thing to do
Being under-trialled is not the same as being wrong, and three things make this the strongest card in the deck anyway.
The mechanism is unambiguous. Every stone begins as a crystal forming out of solution. Dilute the solution and you reduce the tendency to crystallise. This is not a subtle biological argument.
The epidemiology is consistent and points the same way as the single trial.
And it is measurable. Unlike almost every other change in stone prevention, you can find out whether it worked: collect, change, collect again, compare the volume line.
The same 2017 review notes that the more open question is not whether to drink more but what to drink — where the evidence is thinner still, and where the answers differ by beverage rather than by volume.
Where the evidence is genuinely unsettled
Caffeine. A 2018 review found the picture contradictory: caffeine increases urine output, which helps, but may nudge stone risk indices the other way, while three large cohorts suggested a protective association. Coffee is not obviously in either column.
Thresholds. A 2024 analysis of 24-hour urinary chemistries points out that earlier work often assumed either a straight-line relationship or an arbitrary cut-off, and that the true hierarchy of effects among urinary factors is less settled than reference ranges imply. Two litres is a useful landmark, not a cliff edge.
Where more is not better
Drinking substantially more fluid is not universally safe. People with heart failure or advanced kidney disease may have fluid restrictions, and in those situations a general instruction to drink more is actively wrong.
Very large intakes can also lower blood sodium. This is uncommon, but it is a real mechanism rather than a theoretical one, and it is why a target should come from someone who knows your medical history.
What to do with this
Ask your clinician what urine output they want from you, not how many glasses. It is the number the evidence is about, the number your test measures, and the number you can actually check.
If you have had a 24-hour collection, the volume figure on it tells you where you are starting from. If you have not, that is the more useful next step than any amount of guessing at glasses.