Does Advice to Drink More Fluids Prevent Recurrent Kidney Stones?

Kenny Lin, MD, MPH
April 13, 2026

Adults who have had one or more kidney stones are typically advised to increase fluid intake. The supporting evidence for this preventive intervention is limited, however. A 2014 Agency for Healthcare Research and Quality Effective Health Care review identified low-quality evidence from two small, randomized trials of people with calcium stones. In these trials, increasing fluid intake to maintain a urine output of more than 2 liters (L) per day over 3 to 5 years reduced the relative risk of symptomatic or radiographic stone recurrence by 45%, with a number needed to treat of 7. No adverse effects were observed. As a result, the American College of Physicians recommended in a clinical practice guideline that people with kidney stones and a daily urine output of less than 2 L increase their fluid intake if not contraindicated for other reasons.

Similarly, Drs. Leonardo Ferreira Fontenelle and Thiago Dias Sarti wrote in a 2019 AFP review article that “the most important lifestyle modification to prevent recurrent kidney stones is to increase fluid intake to 2.5 to 3 L per day to guarantee diuresis of 2 to 2.5 L per day and a urine specific gravity lower than 1.010.”

A trial published in March 2026, the Prevention of Urinary Stones with Hydration (PUSH) study, tested a 2-year multicomponent behavioral intervention to increase fluid intake in 1,658 participants 12 years and older with previous kidney stones from six medical centers in six different US states. The intervention comprised (1) a prescription to increase urine volume to more than 2.5 L per day, (2) a financial incentive of $1.50 per day for the first 6 months for adhering to the fluid prescription (verified by a Bluetooth-enabled smart water bottle), and (3) health coaching and automated text messaging reminders to overcome barriers to adherence. Control participants were also provided a smart water bottle but were not required to use it. The primary outcome was symptomatic stone recurrence.

By the end of the study, a statistically similar percentage of the intervention (19%) and control (20%) groups had either passed a kidney stone or undergone a procedural intervention for a stone. This nondifference occurred despite increased fluid intake in the intervention group; daily average urine volume peaked at 1.8 L at 6 months and gradually declined to less than 1.6 L by 24 months. At 6 and 12 months, intervention participants were more likely to report urinary frequency, urgency, and nocturia. More intervention participants developed asymptomatic hyponatremia (12 vs 2 in the control group; p = 0.018); no one developed hyponatremia requiring hospitalization.

Although the failure of the intervention to achieve the daily urine volume goal likely contributed to the PUSH study’s negative result, it is hard to imagine a different primary care–feasible intervention performing any better. Further, the increases in urinary symptoms and hyponatremia associated with the intervention provided evidence of rare but clinically significant harms. An accompanying editorial reasonably suggested, “If adherence to foundational advice is unattainable even under optimal trial conditions, then … framing [fluid intake] targets more flexibly with individualized goals that are aligned with work or study patterns, beliefs, access to palatable water, thirst cues, and other competing demands might be more successful.”