Variable · advanced
Water fasting
Water only, no calories from any source, for an extended period. The strictest common protocol, the most searched, and the one where the gap between what is claimed and what has been measured is widest. Extended water-only fasting is studied under medical supervision, and the published cardiometabolic improvements largely regress after refeeding.
Who should not do a Water fast fast
- Anyone taking insulin or a sulfonylurea.
- Anyone who is pregnant or breastfeeding.
- Anyone with a current or past eating disorder. Water fasting is the protocol most often used to disguise restriction, and it carries the highest relapse risk on this site.
- Children and adolescents.
- Anyone underweight or malnourished, or with low baseline phosphate, potassium or magnesium.
- Anyone with kidney disease, liver disease, heart failure, an arrhythmia, or type 1 diabetes.
- Anyone taking diuretics, SSRIs, lithium, or medication requiring food.
- Anyone intending to fast beyond about 48 hours without medical supervision. That is the point at which the published protocols stop resembling anything done alone.
This is not medical advice and no clinician has reviewed it. If any of the above applies to you, speak to a doctor before fasting rather than after.

What is actually happening
Water fasting is defined by what is excluded rather than by duration: no calories from any source, water only. The physiology is the same sequence every fast follows: glycogen depletion, then gluconeogenesis, then a progressive handover to ketones. Two things scale with length in a way they do not over shorter fasts. The first is electrolytes. Sodium is excreted as insulin falls, and because water fasting excludes the broths and electrolyte drinks that other protocols permit, it is the protocol most likely to produce dilutional hyponatraemia: sodium falling while water intake rises. The second is refeeding. The longer and more complete the fast, the more the meal that ends it drives phosphate, potassium and magnesium into cells as insulin rises. Both risks are managed by supervision in the studies that exist, which is the single most important difference between the published protocol and the one described on the open web.
What the evidence shows
- In a study of prolonged water-only fasting followed by refeeding, improvements in markers of cardiometabolic risk were observed, and the paper reports the refeeding phase alongside the fast rather than the fast alone. The changes are substantially attenuated once eating resumes.
- Seven-day complete caloric restriction in humans produced extensive plasma proteome changes, most of which emerged only after roughly the first two to three days.
- Extended water-only fasting is studied under medical supervision with monitoring and structured refeeding. The unsupervised version described across most of the open web is not the protocol that was tested.
- Because water fasting excludes electrolyte-containing drinks by definition, it is the fasting protocol most exposed to dilutional hyponatraemia, a condition whose early symptoms are indistinguishable from ordinary fasting discomfort.
What the evidence does not support
- The evidence does not support water fasting as a treatment for any disease. The published cardiometabolic improvements are short-term marker changes that largely regress after refeeding, and no trial demonstrates a durable clinical outcome.
- The supervised study protocols cannot be used as evidence for unsupervised water fasting. Monitoring and structured refeeding are what make the studied version acceptably safe, and removing them changes the risk profile rather than the effect size.
- No human evidence establishes an autophagy threshold at any duration of water fasting, or that fasting-induced autophagy improves any outcome in people.
- This page gives no electrolyte doses and no duration recommendation, because the honest answer is that the safe version of this protocol involves clinical monitoring rather than a number from a website.
How it is run
- Water only. No broth, no electrolyte drinks with calories, no coffee on the strictest reading of the protocol.
- This is the protocol where the strict definition and safe practice conflict most sharply. Sodium supplementation is what prevents the main risk, and it is also what makes the fast no longer water-only. Published extended fasts handle this with medical supervision, not with a rule from a website.
- Extended water-only fasting is studied in residential, supervised settings with monitoring and a structured refeeding protocol. Doing it unsupervised is not the same intervention.
- Break it with a structured, gradual refeed rather than a normal meal, and treat the refeeding period as part of the protocol.
- Stop for confusion, fainting, chest pain, irregular heartbeat, or persistent vomiting.
What breaks a Water fast fast
The rules differ by protocol. Check any food, drink or supplement against this protocol below.
Sources
- The Effects of Prolonged Water-Only Fasting and Refeeding on Markers of Cardiometabolic Risk · Nutrients, 2022PMID 35334843
- Systemic proteome adaptions to 7-day complete caloric restriction in humans · Nat Metab, 2024PMID 38429390
- Understanding Refeeding Syndrome in Critically Ill Patients: A Narrative Review · Nutrients, 2025PMID 40507135
What it does not show: Concerns critically ill hospital patients. It is cited for the mechanism of refeeding electrolyte shifts, not to suggest that an ordinary fast carries the same risk.
- Exercise-associated hyponatremia · Clin J Am Soc Nephrol, 2007PMID 17699400
What it does not show: Concerns endurance athletes drinking heavily during exercise, not fasting specifically. The mechanism, dilution of blood sodium by excess plain water, is the same one that makes very large water intakes during a long fast dangerous.
- Diagnosis and Management of Sodium Disorders: Hyponatremia and Hypernatremia · Am Fam Physician, 2023PMID 37983699
What it does not show: A clinical review for physicians managing sodium disorders. It is not about fasting and is cited here only for what low blood sodium does and how it presents.
- Influence of glucagon on natriuresis and glucose-induced sodium retention in the fasting obese subject · Eur J Clin Invest, 1977PMID 408142
What it does not show: Studied people with obesity during a supervised fast. It explains why sodium is lost during fasting; it does not tell you how much to replace.
No clinician reviews this site. Sources on this page were verified against PubMed by the author, who is not a doctor or a dietitian. What that means.
Last updated 2026-08-16.
The phases of a fast covers the physiology behind every protocol on this site, hour by hour.