What is fasting? What happens when you stop eating?

Humanity spent thousands of years trying to make food reliably available. We grew crops, domesticated animals, invented refrigeration, and developed delivery services capable of bringing dinner to the door without anyone putting on shoes. Then we created mobile apps to congratulate ourselves for not eating breakfast.

The joke is that there really is something worth investigating here. When food stops arriving, the human body doesn’t simply run down like a phone with a dying battery. It changes fuel supplies, redirects hormones, adapts its metabolism, and keeps essential organs working with remarkably little incoming energy. Some of those changes can improve health. Others are survival responses, and a few may come at a cost.

The useful question, then, isn’t whether fasting is good or bad. It’s what fasting does, which effects matter, and when deliberately extending the gap between meals makes sense.

1. What exactly counts as fasting?

From skipping breakfast to extended food deprivation

Fasting means voluntarily abstaining from some or all food, or from food and drink, for a period of time. Everyone who sleeps through the night already experiences a version of it. That’s why the first meal of the day is called breakfast, although the average person manages this feat without professional coaching.

The definitions become important once we try to compare research. In 2024, Daniela Koppold and an international panel of fasting researchers published a terminology consensus in Cell Metabolism. They distinguished repeated intermittent fasting periods of up to forty-eight hours from short-term fasting of two to three days and prolonged fasting of four or more days. These are research categories, not magical biological thresholds.

Time-restricted eating confines meals to a daily window. A 16:8 schedule allows food during eight hours and none during the remaining sixteen; other common variations include 14:10, 18:6, and 20:4. One meal a day, or OMAD, compresses most daily intake into a single sitting. Alternate-day fasting switches between ordinary eating and fasting or strongly restricted days. The popular 5:2 strategy restricts energy intake on two days a week rather than every day.

Someone doing a twenty-four-hour dinner-to-dinner fast, a three-day fast, and a two-week fast is using some of the same metabolic machinery. But they are not performing the same intervention, any more than a walk around the block is equivalent to an ultramarathon.

What you consume changes the definition

A water-only fast permits water but no calories. Modified fasting permits a limited amount of food energy, generally no more than one-quarter of normal requirements under the consensus definition. Buchinger-style programs may include small amounts of juice or broth. Fasting-mimicking diets provide food designed to produce selected features of fasting metabolism.

Religious fasting may be defined by timing or prohibited foods rather than total calorie absence. Ramadan involves no food or drink between dawn and sunset, with eating and hydration after dark. Some Christian traditions restrict meat, dairy, or other foods while permitting meals. A strict water fast, a religious fast, and a calorie-control eating window therefore answer different questions.

Dry fasting excludes drinking as well as eating. Daily religious dry fasting, when water is consumed overnight, is not equivalent to attempting several consecutive days without fluids. Multi-day dry fasting presents serious dehydration risk and has no established health advantage that justifies it.

This also explains why the question “Does coffee break my fast?” rarely has a universal answer. It certainly fails the definition of a water-only fast if it contains additives, and black coffee isn’t water. But the relevance of a tiny caloric intake to weight management is different from its relevance to a strict research protocol or religious observance. Define the objective before inventing the rule.

2. Your body has a backup pantry

What happens after the last meal

After eating, nutrients are absorbed into circulation. Insulin rises, glucose and fat are made available for immediate use, and some glucose is stored as glycogen in the liver and muscles. Energy beyond current needs can be deposited in body fat. The body isn’t stockpiling for fun; stored energy is what keeps us alive between meals.

As absorption ends, insulin gradually falls and glucagon becomes relatively more important. Liver glycogen can be broken down to maintain blood glucose. Fat cells release fatty acids for tissues that can burn them directly. These processes are already operating overnight. Fasting allows their relative contributions to change over a longer interval.

Over approximately the first day, liver glycogen becomes substantially reduced in many people. Timing varies with previous meals, carbohydrate intake, exercise, and individual physiology. The body increasingly makes glucose internally through gluconeogenesis, using materials including glycerol from fat, lactate, and certain amino acids. It does not turn glucose production off; some tissues continue to need glucose even during a long fast.

Ketones and the metabolic switch

As fat mobilization increases, the liver manufactures ketone bodies, mainly beta-hydroxybutyrate and acetoacetate. These can supply energy to several tissues, including the brain. That matters because the brain doesn’t directly use long-chain fatty acids efficiently and normally relies heavily on glucose.

If the brain demanded its ordinary glucose supply indefinitely while no food arrived, the body would need to manufacture much of that glucose from amino acids. That would consume essential protein at a dangerous rate. Greater use of ketones reduces the brain’s glucose demand and therefore helps conserve protein. It never makes protein loss disappear altogether.

Rafael de Cabo and Mark Mattson described this transition in a 2019 New England Journal of Medicine review as the “metabolic switch.” It’s a useful name provided we don’t imagine someone pulling a lever marked FAT at exactly hour sixteen. The body uses mixtures of fuels, and fasting alters the proportions gradually.

By roughly two or three days, ketosis becomes more substantial, insulin is very low, and the body is operating differently from the way it does after breakfast. Continue further and changes in protein turnover, hormone signaling, fluid balance, and energy expenditure become increasingly important. Those changes are why this series treats three days, seven days, and longer periods as separate investigations.

Why the scale can be misleading

Early fasting reduces glycogen and the water associated with it. Falling insulin can increase sodium and water loss as well. So rapid scale-weight loss is not proof of equally rapid fat loss. You can weigh less without having burned that entire difference in body fat, which is a slight disappointment to anyone whose bathroom scale has already begun celebrating.

A 2026 meta-analysis by Süleyman Ulupınar and colleagues synthesized forty-nine body-composition studies. It found reductions in fat mass, but also larger standardized reductions in fat-free mass and total body water. Fat-free mass includes water and organs as well as muscle, so it is equally wrong to equate every lost kilogram of lean mass with a kilogram of skeletal muscle.

3. Hunger, hormones, and the alert brain

Hunger doesn’t rise like a fuel gauge

Hunger often arrives around the times a person normally eats, then subsides, then returns. Ghrelin contributes to appetite signaling, while learned timing, sleep, stress, smells, and food cues also influence what we experience. A freshly baked pastry can create a sudden interest in lunch even when breakfast was perfectly adequate.

This doesn’t mean hunger is imaginary or that it ought always be ignored. It means hunger is regulated, not simply a gauge that moves steadily upward until someone refills the tank. Temporarily not eating can make that distinction easier to observe.

Growth hormone is not a free gym membership

Fasting lowers insulin and changes glucagon and stress-hormone signaling. Cortisol can rise; during longer food deprivation thyroid-related and reproductive signaling may decline as the body conserves resources. Growth hormone can rise dramatically. In a classic 1992 experiment, Mark Hartman and colleagues found approximately a fivefold increase in twenty-four-hour growth-hormone production after two days of fasting in nine healthy men.

That sounds like an excellent opportunity to replace the gym with a comfortable chair and an empty dinner plate. Unfortunately, the body is not building muscle from nothing. Dietary amino acids are absent, IGF-1 and other anabolic signals change, and net protein loss continues. The elevated growth hormone is better understood as part of adaptation and fuel mobilization than as a bodybuilding program.

Orexin: why food scarcity can increase wakefulness

Orexin, also called hypocretin, is a neuropeptide system associated with wakefulness, arousal, motivation, and food-seeking. This is useful because an animal whose response to an empty stomach is to become steadily sleepier would be poorly equipped to find dinner.

Akihiro Yamanaka and colleagues demonstrated in mice in 2003 that orexin neurons help regulate the increase in wakefulness and activity associated with food scarcity. A much smaller human study of eight volunteers undertaking daytime fasting around Ramadan found that circulating orexin-A increased during fasting hours. These results support a role for orexin in fasting-related arousal; they do not demonstrate that orexin climbs continuously for every additional day without food.

The prayer connection is intriguing. Someone fasting for religious reasons may direct a wakeful, motivated state toward contemplation rather than food-seeking. That is a plausible interpretation of an experience described by fasting traditions; it is not evidence that orexin makes prayer more effective. Nor is feeling alert proof of doing better on cognitive tests. A 2025 meta-analysis by Christoph Bamberg and David Moreau found no meaningful overall cognitive advantage in fasted adults, with some modest disadvantages during longer intervals. Most included fasts were far shorter than three days.

The brain may feel different without food. That doesn’t mean it automatically becomes a better brain. If hunger alone generated enlightenment, every teenager waiting for dinner would be a sage.

4. What benefits are actually supported?

Weight management and glucose regulation

A major 2025 BMJ review by Zhila Semnani-Azad and colleagues combined ninety-nine randomized trials involving 6,582 adults. Intermittent fasting and continuous calorie restriction both reduced weight compared with unrestricted eating. Alternate-day fasting had a modest average advantage over continuous restriction, about 1.3 kilograms; the other strategies were broadly similar.

So fasting works for weight management, but it doesn’t grant an exemption from energy balance. A person who skips breakfast and compensates with a heroic evening meal may achieve very little fat loss. What fasting can offer is an easy structure: “I finish eating after dinner” may be easier to live with than counting every calorie in every spoonful of sauce.

Some studies also find improvements in fasting glucose, insulin, waist circumference, and blood pressure, especially in people with excess weight or metabolic dysfunction. Earlier eating windows may align better with circadian glucose handling than late-night eating. But lowering insulin during the hours without food is a normal fasting response, not proof that a chronic metabolic disorder has been cured.

Digestive rest and the limits of the claim

No incoming food means less immediate work digesting food. Between meals, the gastrointestinal tract also exhibits fasting-related movement patterns such as the migrating motor complex. These are real mechanisms, not imaginary “gut rest.”

What isn’t established is the idea that simply going without food automatically repairs an injured intestinal lining or resets the microbiome. The gut remains metabolically active. If someone stops eating a food that causes symptoms, the disappearance of symptoms during fasting does not establish that the underlying problem has been treated. The restaurant is closed; that doesn’t prove the plumbing has been fixed.

Practical usefulness is personal

One person may find a sixteen-hour fast effortless and liberating. Another might become preoccupied with food, compensate by overeating, or find the schedule disruptive to work, exercise, and family meals. A method that makes healthy eating simpler is useful. A method that turns dinner with friends into an argument with a stopwatch may not be.

That is the strongest case against prescribing fasting universally: most short-term benefits can often be achieved through ordinary dietary improvement, sensible calorie intake, and physical activity. Fasting is one option, not a nutritional admission requirement.

5. The interesting claims we haven’t proved

Autophagy isn’t a timer

Autophagy is part of the body’s cellular recycling and quality-control machinery. Cells break down unnecessary or damaged components and reuse the materials. It occurs even when we eat. Nutrient deprivation can affect pathways such as mTOR and AMPK, and animal studies offer strong reasons to investigate fasting’s influence.

The internet has turned this into a microscopic cleaning crew that allegedly clocks in precisely twenty-four hours after your last sandwich. Unfortunately, cells don’t carry timecards. We lack reliable human measurements proving that a particular hour triggers a universal burst of tissue-wide “cleanup.” In 2025, Julien Bensalem and colleagues reported intriguing but preliminary changes in autophagic flux from a time-restricted-eating study; the results did not establish a universal threshold.

Inflammation and cholesterol can move the wrong way

Some intermittent-fasting studies report improved inflammatory markers. But Serena Commissati and colleagues found an acute rise in inflammatory and platelet-activation markers during medically supervised water-only fasts averaging around ten days in 2025. This could be transient adaptation, and long-term meaning remains uncertain; nevertheless it defeats the slogan that fasting always decreases inflammation.

Similarly, a 2026 meta-analysis of thirty-two water-only fasting studies by Ayşe Çamli and colleagues found that LDL and total cholesterol tended to rise during fasting, with HDL often declining in longer fasts. The studies varied in quality and design, but the direction is a useful warning against interpreting every laboratory change as good news.

Longevity and disease treatment remain unsettled

Fasting affects pathways studied in aging biology, and some dietary restrictions extend lifespan in laboratory organisms. Human trials have not shown that a specified fasting window reliably adds years to a human life. A mouse can have an excellent response to an intervention without obligingly demonstrating the same result in a person who has a mortgage and a much longer lifespan.

Disease-specific studies are also underway, including autoimmune conditions and cancer-treatment adjuncts. A plausible mechanism or small early trial is not evidence that fasting independently treats those conditions. The sensible conclusion is that fasting changes important biology, and we still need to establish which changes produce lasting clinical benefits.

6. How much fasting is enough?

A moderate overnight eating window may fit naturally into a healthy lifestyle. A three-day fast is a substantially different intervention. A seven-day or twenty-one-day fast changes fluid, mineral, hormone, and protein regulation more profoundly, with increasing concern about medication interactions, dizziness, uric acid, electrolyte disturbance, and refeeding.

People who are pregnant or breastfeeding, growing, malnourished, frail, or dealing with a current or past eating disorder should not casually attempt restrictive fasting. Diabetes medications—particularly insulin and medicines capable of causing hypoglycemia—can create serious dangers without appropriate oversight. Multi-day fasting warrants medical assessment, and prolonged fasting should be medically supervised rather than treated as a home endurance challenge.

The more interesting question isn’t how long a human can endure without food. It’s what someone is trying to accomplish and whether the expected benefit justifies the interruption, inconvenience, and risk.

Fasting is a capability, not a requirement

Humans can run on stored fuel, use ketones to help protect protein, and adapt surprisingly well to missing meals. Knowing this can change our relationship with appetite. Hunger may be important information without being an immediate command. We can notice it, examine it, and sometimes decide not to act on it right away.

But a survival adaptation isn’t automatically a treatment. We survive cold, injury, and sleepless nights too; no one should infer that maximizing those experiences makes us healthy. Fasting deserves the same thoughtfulness.

It can be useful for weight management, meal structure, and understanding our own habits. It may also be interesting psychologically and spiritually. It isn’t mandatory, and longer isn’t automatically better.

Research and references

• Koppold DA et al. (2024), International consensus on fasting terminology, Cell Metabolism: https://pubmed.ncbi.nlm.nih.gov/39059384/
• de Cabo R & Mattson MP (2019), Effects of intermittent fasting on health, aging, and disease, NEJM: https://doi.org/10.1056/NEJMra1905136
• Semnani-Azad Z et al. (2025), 99-trial network meta-analysis, BMJ: https://www.bmj.com/content/389/bmj-2024-082007
• Ulupınar S et al. (2026), Fasting and body composition meta-analysis: https://pubmed.ncbi.nlm.nih.gov/42440276/
• Hartman ML et al. (1992), Growth hormone response to two-day fasting: https://pubmed.ncbi.nlm.nih.gov/1548337/
• Yamanaka A et al. (2003), Orexin neurons and fasting-related arousal: https://pubmed.ncbi.nlm.nih.gov/12797956/
• Diurnal intermittent fasting and human orexin-A (2018): https://pubmed.ncbi.nlm.nih.gov/29387256/
• Bamberg C & Moreau D (2025), Fasting and cognition meta-analysis: https://pubmed.ncbi.nlm.nih.gov/41182703/
• Bensalem J et al. (2025), Human autophagic flux trial: https://pubmed.ncbi.nlm.nih.gov/40345145/
• Commissati S et al. (2025), Prolonged fasting and inflammatory response: https://pubmed.ncbi.nlm.nih.gov/40268190/
• Çamli A et al. (2026), Water-only fasting and blood lipids meta-analysis: https://pubmed.ncbi.nlm.nih.gov/41994097/

Medical note

This series is educational journalism, not personalized medical advice or a protocol for extended fasting. Prolonged fasting can have serious risks, particularly with chronic illness, certain medications, malnutrition, or abnormal electrolytes, and warrants professional medical supervision.