Prolonged fasting: why it is never improvised

Prolonged fasting is not the name of a method: it is the term that groups together every fasting practice that extends over several days with no duration fixed in advance. Very different practices live under that same word — a water-only fast, a fast with no drink at all, a fast carried out in a medical facility, or simply a 72-hour fast that nobody stopped after three days.

What this article argues, its central thesis: what unites these heterogeneous practices is not a shared protocol, it is a shared level of risk. And what that shared level of risk means, in practice, is that none of these practices, whatever form it takes, is an individual decision to make alone in front of a screen. Looking for “the” right way to carry out a prolonged fast means searching for something that does not exist under this name: there is no single method, there is a family of practices that all share the same need for supervision.

This page acts as the synthesis article in our fasting section: it frames, as completely as possible, the contraindications shared across these practices, explains what still sets them apart from each other, and points toward our dedicated articles on the 72-hour fast, water fasting, dry fasting and therapeutic fasting for what is specific to each. It details no protocol that anyone could carry out alone, for any of these forms.

What is prolonged fasting?

Prolonged fasting refers to any absence of food intake that extends over several days with no endpoint decided at the outset. It is this absence of a duration fixed in advance, more than the number of days itself, that sets prolonged fasting apart from named, bounded practices such as the 36-, 48- or 72-hour fast: those announce a known end in their name itself, while prolonged fasting leaves it open, decided along the way according to clinical criteria.

In everyday use, the term most often applies to whatever goes beyond those named tiers, or to practices that, from the start, set no precise duration at all. But treating a single day-count as a firm cutoff would be misleading: what matters is not crossing a particular number, it is understanding that having no end fixed in advance changes the very nature of what serious supervision has to cover, as the following sections explain.

How does it work?

Faced with a prolonged absence of intake, the body follows a path that resembles, in its broad outline, what happens during shorter fasts, but that continues beyond it: glycogen reserves in the liver and muscles are exhausted within the first few days, the body shifts progressively toward stored fat as its main energy source, and then, with no stopping point fixed in advance, this adaptation continues for a length of time that nothing in the name of the practice itself limits.

It is this absence of a built-in limit that sets prolonged fasting apart physiologically: every additional day adds its share of strain on muscle reserves, fatigue and electrolyte risk, with no precise number marking a turning point. It is a cumulative process, not a switch that flips at a given hour, which is precisely why continuous monitoring, capable of tracking that accumulation day by day, matters more than any estimate made in advance.

The forms prolonged fasting can take

Prolonged fasting does not break down into three distinct, separately named variants, despite what such a general term might suggest. It instead covers practices already named elsewhere on this site, once they extend with no duration fixed: a water-only fast, which keeps only water; a dry fast, which removes water too; a therapeutic fast, carried out in a clinical setting with organised monitoring; and, more simply, a 72-hour fast that nobody ended once the planned three days were up.

Each of these forms has its own particular issues, detailed in our dedicated articles: water alone does not carry exactly the same risks as no drink at all, and an organised clinical setting is not comparable to a practice carried out alone at home. But what groups them under the label “prolonged fasting” is not a shared technique: it is the fact that none of them, once the duration is left open, is a decision to make alone. That point, more than the detail of each form, is what the rest of this article develops.

Potential benefits of prolonged fasting

The benefits raised around prolonged fasting echo, amplified by duration, those already attributed to shorter fasts: a reduction in calorie intake over the period, mental clarity reported by some people, or a supposed effect on inflammation. The table below summarises what can reasonably be expected, with caution that grows with duration rather than easing off.

Effect on weight loss

Prolonged fasting produces rapid and often marked weight loss, visible from the first few days. But most of that loss corresponds, at first, to water and the glycogen holding it in the liver and muscles, not to fat: a number moving quickly on the scale says nothing, on its own, about what it is actually made of.

The longer the duration stretches, the larger the share of muscle tissue drawn on in turn, and part of the lost weight returns once ordinary eating resumes. This site does not present any form of prolonged fasting as a long-term weight management method: to set a realistic target, our calorie and basal metabolic rate calculator remains the better-suited tool, built on the balance between intake and expenditure rather than on an open-ended stop to eating.

Effect on metabolism

Faced with an absence of intake that extends with no fixed limit, the body puts energy-sparing mechanisms in place, aimed at protecting its reserves. This should not be read as an acceleration of metabolism, despite what some shortcuts suggest, but as an adaptive response that becomes more pronounced the longer the duration stretches.

This adaptation is all the more marked because prolonged fasting, by definition, has no fixed endpoint: unlike a short fast, where the body effectively “knows” intake will resume soon, a fast whose end is not determined in advance draws more heavily on these preservation mechanisms. Nothing in available data supports the opposite claim, that prolonged fasting produces a durable, favourable effect on resting energy expenditure.

Effect on autophagy

Autophagy, the cellular recycling mechanism often invoked to justify multi-day fasts, remains plausible but unquantified in humans, whatever the exact duration under consideration. What is documented with precision comes mostly from laboratory conditions, in animals or isolated cells, in settings that do not transpose directly to a human body in a state of prolonged fasting.

A recurring argument holds that if autophagy exists, a longer fast must mechanically produce more of it. No reliable data supports this proportionality in humans: it is an appealing line of reasoning, but one that goes beyond what current measurements can actually confirm, and it should never be used to justify extending a fast that is already under way.

Effect on blood sugar

During prolonged fasting, blood sugar generally settles at a low but regulated value, with the body mobilising several mechanisms to prevent a dangerous drop. In someone with no disorder of blood sugar regulation, this stability is not, on its own, a benefit worth pursuing: it simply describes the expected state of a body deprived of intake over a stretching period.

In someone with diabetes or on treatment that affects blood sugar, this same mechanism becomes dangerous, and the risk grows with every additional day, without the absence of a fixed endpoint allowing anyone to predict exactly when the situation might turn. This is territory that is never managed through general reasoning, only through continuous, individualised medical follow-up.

Effect on insulin

A prolonged absence of calorie intake keeps insulin secretion at a low level, consistent with the absence of dietary glucose. This is an expected physiological state, not a therapeutic effect specific to prolonged fasting: any sufficiently long absence of intake produces this same general pattern, which does not intensify in a straight line beyond a certain point.

The absence of a duration fixed in advance, specific to prolonged fasting, weighs directly on anyone whose treatment changes insulin secretion or action: a low state that sets in with no known endpoint is not neutral, and it may call for a treatment adjustment, decided and followed by a professional, before the practice is even considered. This is not a minor detail: it is one of the reasons this tier, whatever specific form it takes, does not lend itself to improvisation.

Effect on cholesterol

Data on the link between a multi-day fast and lipid profile remain limited and sometimes contradictory. Some variations are reported during the fasting period itself, without anything indicating that they persist once ordinary eating resumes.

A temporary effect observed during an absence of intake that extends over several days says nothing about a durable cholesterol benefit: it is a distinction that caution requires keeping in mind, otherwise a transient variation risks being wrongly presented as a merit specific to prolonged fasting.

Effect on muscle mass

This is one of the points where the absence of a fixed duration weighs most heavily. Once glycogen reserves are largely mobilised, the body draws an increasing share of its energy from muscle protein, and that share grows with every additional day, with no natural ceiling stepping in to stop it on its own.

This is one of the main reasons prolonged fasting deserves individual evaluation before being considered, particularly for older adults, people whose BMI is already low, or athletes in a heavy training phase. This point is developed further below, in the section detailing the profiles for whom this open-ended duration weighs most heavily.

What the research says

Data available specifically on “prolonged fasting” as a general category are, by construction, difficult to gather: the term covers durations and forms too different from each other to make up a consistent body of evidence. What does exist mostly concerns named, bounded practices — a 72-hour fast, a therapeutic fast in a clinical facility — rarely open-ended fasts, with no fixed duration, carried out outside any clinical setting.

What can be stated more firmly concerns mainly the risks tied to refeeding after a prolonged absence of intake: this point, documented under the name refeeding syndrome, rests on considerably firmer ground than the benefits sought during the fast itself, whatever specific form of prolonged fasting is under consideration.

The risks of prolonged fasting

The first risk concerns sodium, potassium and magnesium balance, already present in shorter fasts but sharpened with every additional day: dizziness, a drop in blood pressure on standing, trouble concentrating, marked fatigue. With no endpoint fixed in advance, nothing guarantees this imbalance stays contained — catching it before it becomes serious is exactly the function that continuous monitoring is there to serve.

The second risk, the best documented, concerns refeeding. A body that has adapted to a prolonged absence of intake can react strongly when food returns: a rapid rise in insulin in response to reintroduced food drives a large shift of phosphate, potassium and magnesium into cells, along with fluid retention. This is what medicine calls refeeding syndrome, and this risk grows with the length of the fast that precedes it, with no sharp threshold marking when it appears.

A third risk, specific to the absence of a fixed duration, deserves naming on its own: that of continually pushing back the moment of stopping. With no endpoint decided in advance with a professional, someone can be tempted to “see how far they can go,” a line of thinking that has nothing rational behind it physiologically: the risk does not fall with practice, it accumulates with the days.

One last, less often named risk concerns the relationship with food: a fast with no fixed end can reinforce, in people already predisposed to it, restrictive patterns of thinking, and can even blur the perception of hunger and fullness signals over time. This risk grows with duration and with the absence of structure, which makes it a vigilance point in its own right, alongside the biological parameters.

Who should avoid prolonged fasting?

This list applies to every form prolonged fasting can take, without exception: water-only, dry, therapeutic, or a simple extension of a 72-hour fast. Several profiles should never consider it without prior medical advice, and some should avoid it outright:

• minors, whose growth requires nutritional intake that no stretched-out absence of food can respect;
• pregnant or breastfeeding women, for whom deprivation with no end fixed in advance concerns the child’s needs directly;
• anyone with a history of disordered eating, for whom a prolonged, open-ended removal of food can reactivate or worsen an already fragile relationship with food;
• people with diabetes on treatment, whose adjustment usually depends on meal timing that an open-ended absence removes;
• anyone on medication whose efficacy or tolerance varies with meal timing, which a fast with no fixed endpoint disrupts for longer;
• people with kidney or liver impairment, for whom every additional day complicates the management of electrolytes further, both during the fast and at refeeding;
• frail older adults, whose already limited reserves tolerate poorly an absence of intake that stretches with no announced end;
• people whose BMI is already low, for whom an open-ended absence of intake mechanically worsens the risk at refeeding;
• athletes in a heavy training phase, whose high energy needs and intense muscular demands make a prolonged absence of intake particularly taxing on the body.

The uncertainty around how long a prolonged fast might last makes the question of warning thresholds even more necessary for anyone whose BMI is already low or whose relationship with food is fragile. Our page on underweight and undernutrition details the signs that justify a consultation, often well before the question of prolonged fasting arises at all.

How to start prolonged fasting?

Considering prolonged fasting does not begin by choosing a duration: it begins by clarifying, with a health professional, which specific practice is actually being considered. Since the term is an umbrella, the first useful step is often to name what you actually have in mind — water-only, dry, therapeutic, or a simple extension of a shorter fast — before even raising the question of duration. The three levels below do not describe durations to reach: they describe degrees of preparation and supervision.

Beginner

Someone who has never considered a multi-day fast starts with a conversation with their doctor: real motivations, medical history, current treatments, and clarification of the exact form being considered. A shorter fast, already evaluated, generally precedes any thought of a fast whose duration is not fixed in advance.

Intermediate

Someone with prior experience already evaluated medically can, with that same professional, discuss a practice whose duration is not fixed in advance: which further tests would be useful, what contact and monitoring arrangement to put in place, and above all, which precise criteria, agreed together, will determine the moment to stop.

Advanced

A prolonged fast carried out in a setting where a medical team remains reachable or present, with biological tests repeated throughout the practice and not only at its start, matches the level of supervision this article considers reasonable. This is not a personal preparation level to reach alone, but a care arrangement built and maintained with professionals, from the first day to the last.

A sample programme

What a supervised episode of prolonged fasting generally looks like follows several phases, described here to help you understand what supervision covers — never as a schedule to reproduce on your own. What sets prolonged fasting apart from a named, bounded fast is that these phases do not sit on a calendar fixed in advance: it is the monitoring itself that determines, day by day, whether the next phase has grounds to happen at all.

• Assessment phase: a conversation about motivations and the specific form being considered, a clinical and biological check-up, and confirmation that none of the profiles detailed above apply.
• Supervised fasting phase: regular contact with, or the presence of, a medical team, biological tests repeated at close intervals, and stopping criteria reassessed continuously rather than fixed once at the start.
• Supervised refeeding phase: a gradual reintroduction of food under close biological monitoring, precisely because this stage concentrates the refeeding syndrome risk detailed above, and because that risk grows with the length of the fast that preceded it.
• Follow-up phase: confirming that the monitored parameters have returned to normal, over a period decided case by case with the professional, never on a schedule fixed in advance.

This sequence describes a logic of supervision, not a set of steps to copy: each phase assumes an individual assessment that nothing in this description can replace without the support that carries it.

What to drink during prolonged fasting?

The answer depends entirely on the specific form of prolonged fasting involved: a water fast keeps water as the only intake, a dry fast excludes it too, and a therapeutic fast follows whatever framework the facility supervising it sets. This is one of the reasons this page stays general on this point: our dedicated articles on each of these forms detail what is specific to them.

What holds true across every form, however, is that water and electrolyte balance becomes, over an open-ended duration, a monitored parameter rather than a choice left to personal judgement. This is why this page gives no quantity or drinking rhythm to follow, whatever form is being considered.

What to eat before prolonged fasting?

The relevant question is not “what to eat the day before” but “what does a prior assessment actually look at.” Before considering prolonged fasting, in any of its forms, a health professional evaluates several factors that weigh directly on the safety of the practice and on how refeeding will need to be organised afterward.

What to eat after prolonged fasting?

This is where this page deliberately details nothing: no quantity, no recommended food, no reintroduction schedule, for any of the forms of prolonged fasting covered in this article. This choice is simply the logical consequence of everything above: refeeding syndrome is a recognised medical complication, which only individual assessment and biological data can help prevent and manage, never a generic set of steps.

What refeeding engages on the medical side is summarised in the table below: it describes what is monitored and why, not what to eat. That distinction is the core of what this article has argued since its introduction.

Whatever specific form of prolonged fasting was followed, the question of what comes after stays the same: our ideal weight page offers guidance built on a personal, long-term goal, rather than a list of foods meant to apply to every practice grouped under this term.

Common mistakes with prolonged fasting

1. Treating “prolonged fasting” as the name of a single method: it is an umbrella term that covers different practices, each with its own issues.
2. Assuming that stretching a short fast by a few hours is enough to shift into this category without changing the level of supervision or vigilance.
3. Fixing a duration in advance and sticking to it no matter what, instead of following clinical stopping criteria reassessed continuously.
4. Not being able to describe, in a consultation, the specific form being considered: water-only, dry or therapeutic fasting do not carry exactly the same risks.
5. Treating an account of a multi-week fast carried out alone as proof it is feasible, when an imbalance can stay invisible in a story that ends well.
6. Pushing back refeeding indefinitely because “you still feel fine”, confusing a comfortable feeling with the absence of biological risk.
7. Underestimating that refeeding syndrome risk rises with every additional fasting day, with no sharp threshold announcing the shift.
8. Assuming the contraindications only apply to “really long fasts,” not to your own plan, when they apply as soon as no duration is fixed in advance.
9. Reassuring yourself by comparing a multi-day fast with no water to a religious dry fast bounded by daylight hours, two very different realities our dry fasting article distinguishes clearly.
10. Improvising salt or mineral supplements without medical advice, thinking you can self-correct an electrolyte risk that needs an individualised approach.
11. Resuming intense training immediately after several days without intake, before refeeding has stabilised.
12. Stacking several prolonged fasts close together, banking on experience from a previous attempt, when each practice deserves its own reassessment.
13. Underestimating the psychological weight of a fast whose end is not fixed in advance, a risk factor in its own right for people already predisposed to it.
14. Searching online for “how many days can you fast” instead of asking a professional, when no generic answer can fit an individual situation.
15. Mistaking the support of an online community of people who fast for genuine medical supervision, two things that serve a different purpose and a different level of clinical rigour.

Comparing prolonged fasting with other fasting practices

The table below places prolonged fasting among some of the forms it can take or sit alongside, and adds a daily practice to give a point of reference at the other end of the scale. The safety gap between a daily 16:8 and prolonged fasting is not a nuance: it is a change of category altogether.

This ranking confirms what this article has argued since its introduction: difficulty and risk rise sharply with the absence of a fixed duration, without the demonstrated benefits following the same curve. Our page on BMI limitations makes a point that applies directly here: pushing a limit further is not, on its own, a sign of relevance for an individual situation, and sometimes the opposite is true.

Conclusion

Prolonged fasting is not a method you would choose among others: it is the collective name given to a set of heterogeneous practices that share one level of risk and one requirement — never deciding alone. Whether it takes the form of a water fast, a dry fast, a therapeutic fast, or simply an extension of a 72-hour fast, that conclusion does not change.

If this practice interests you, in any of its forms, the first useful step is not fixing a start date or a duration: it is talking to a health professional, who can assess your situation and determine with you whether, and how, appropriate supervision can be put in place.