What Actually Breaks a Fast?

Ask ten sources whether a splash of milk breaks your fast and you will get ten confident answers, several of them contradictory. That is not because the topic is mysterious. It is because the question is missing a word.

Breaks a fast for what? A fast is not one thing. It is a period of not eating that produces several different effects at once, on different timescales, and the things that interrupt one effect leave another completely untouched. Once you name which effect you are after, most of the arguments dissolve.

Three reasons people fast, and three different answers

Metabolic rest — the reason most people are actually here

You are fasting to keep insulin low for a stretch of the day, so that stored fat is accessible rather than locked away. This is the mechanism behind almost every practical benefit people report from 16:8: steadier energy, quieter appetite, easier fat loss without counting.

For this purpose, the threshold is simple: does it raise insulin meaningfully? Protein and carbohydrate do. Fat barely does. Water, salt, black coffee and plain tea do not.

Digestive rest

You are fasting to give your gut a genuinely empty stretch. Here the threshold is different and blunter: anything with calories counts, including the fat in a bulletproof coffee, because digestion is work regardless of the insulin response.

Autophagy

You are fasting for cellular housekeeping — the process by which cells break down and recycle damaged components. This is the reason people get most dogmatic, and it is also the one where the honest answer is the least satisfying. Most of what is known about autophagy thresholds in humans is extrapolated from animal work and short-term studies; nobody can currently tell you that seventeen grams of cream at hour fourteen cost you a specific percentage of anything. Anyone who quotes you a precise figure is guessing with confidence.

The defensible summary: protein is the most likely of the three macronutrients to blunt autophagy, so if this is your goal, keep the fasting window genuinely clean. But do not let an unmeasurable benefit dictate a schedule you cannot keep.

The list, from safest to least

Does not break a fast, on any reasonable definition

• Water, still or sparkling.
• Salt and electrolytes, provided they contain no sugar. Salt in water is not food; it is the thing that makes fasting comfortable. This is the single most useful item on the list for anyone doing 16:8 on a low carbohydrate intake.
• Black coffee. A couple of calories, no meaningful insulin response. It may even nudge fat oxidation upward slightly.
• Plain tea, black, green or herbal, unsweetened.

Fine for metabolic purposes, ends the fast on a strict reading

• A splash of milk in coffee. Fifteen millilitres carries well under a gram of carbohydrate and a trivial amount of protein. For insulin and fat access, it is noise. For a purist's autophagy window, it is a technical break. If a dash of milk is what makes a sixteen-hour fast something you will still be doing in a year, take the milk.
• Cream, butter or MCT in coffee. Little insulin response, but real calories and real digestion. Fine for metabolic rest and genuinely useful for appetite in the first weeks; it ends the fasted state for autophagy and digestive-rest purposes. Worth knowing that a large fatty coffee also removes some of the training effect of fasting — you are teaching your body to expect fuel at hour twelve.
• Apple cider vinegar, a squeeze of lemon. Negligible.

Breaks a fast for most purposes

• Anything with protein. Bone broth, collagen, BCAAs, a protein shake, a handful of nuts. Protein raises insulin — more than most people expect — and is the macronutrient most likely to interrupt autophagy. Bone broth is a common casualty here: it is nourishing, but it is food.
• Anything with sugar, including the honey in a herbal tea and the fruit juice in a "healthy" drink.
• Most flavoured or "zero" pre-workouts. Read the label rather than the marketing.

Genuinely uncertain

• Artificial and non-nutritive sweeteners. The evidence is mixed and individual. Some people show no insulin response at all; some respond to sweetness itself; some find that anything sweet in the fasting window makes them hungrier two hours later, which matters more than the biochemistry. The practical test is your own appetite: if diet drinks make the back half of your fast harder, that is your answer regardless of what a study says.
• Chewing gum. A gram or two of sweetener and some digestive anticipation. Metabolically trivial; behaviourally, it depends whether it settles you or starts you thinking about food.

Not a question worth asking

Medication. Do not delay or skip prescribed medicine to protect a fasting window. If something must be taken with food, take it with food and shift the window instead. This is not a close call, and no fasting benefit is worth the trade. If your medication schedule and your eating window genuinely conflict, that is a conversation with your prescriber, not a decision to make alone — and if you take insulin or any glucose-lowering medication, fasting needs their supervision from the start.

Why this matters much less on 16:8 than the internet suggests

Here is the part that rarely gets said. Most of the fast-purity debate is imported from people doing twenty-four to seventy-two hour fasts, where autophagy is the point and the margins are fine. On a sixteen-hour daily rhythm, the mechanism doing nearly all the work is simply the length and consistency of the low-insulin stretch — and that survives a splash of milk without complaint.

The things that actually determine whether 16:8 works for you, in order:

1. Whether you keep the window at roughly the same times, most days, for months. Rhythm beats duration.
2. What you eat inside it. Two structured meals with adequate protein and fat will do more than an extra hour of fasting ever will.
3. Whether you are salted well enough to make the fasting hours comfortable. Most "I can't get past hour thirteen" problems are sodium, not willpower.
4. Whether the schedule fits your life — because a window you abandon in December is worth less than a slightly imperfect one you keep all year.

Somewhere below all of that sits the question of whether your coffee had milk in it.

The rule worth keeping

If you want one sentence to carry around:

If it contains protein or carbohydrate, it ends the fast. If it is water, salt, black coffee or plain tea, it does not. Everything else is a judgement call about what you are fasting for.

And a second, less tidy one: the person who drinks a milky coffee at hour twelve and eats two good meals in an eight-hour window, every day for a year, will be in better shape than the person who kept a perfectly clean fast for three weeks and then stopped.

Where this fits

In Keto Dive, the eating window is not a rule to be defended but the frame that holds two meals in place. The book's version of 16:8 pairs the window with a two-meal rhythm, an electrolyte baseline that makes the fasting hours uneventful, and an evening protocol that closes the window early enough to protect sleep.

For the full picture of the window itself, that is the 16:8 guide. To see your own times laid out — window, both meals, the electrolyte moment — the rhythm planner takes about a minute and nothing you type leaves the page.

And if hour thirteen is where your day falls apart, read the electrolytes guide before you shorten the window. It is usually salt.

This guide is general information, written to accompany a book, and is not medical advice. Everyone's health context is different — medication, pregnancy, diabetes, kidney or heart conditions, and a history of disordered eating all change what is sensible. Before making significant changes to how you eat, particularly if any of those apply, talk to your GP or a registered dietitian.