Caffeine withdrawal symptoms: the headache, how long it lasts, and how to cut back

Short answer

Caffeine withdrawal symptoms — most often a dull headache, tiredness, low mood, irritability and foggy concentration — usually start 12–24 hours after your last caffeine, peak between roughly 20 and 51 hours, and fade within 2–9 days. If a headache fits that window and eases soon after caffeine, withdrawal is the likely cause; if it is sudden, severe or comes with weakness, confusion or vision changes, get urgent care.

If day two without coffee leaves you slow, snappy and convinced you cannot function, that is not weakness and not a character flaw. It is a well-measured response of a brain that adapted to a daily drug and is now adjusting back. Before you blame yourself, check what your body already recorded: the timing of your last cup, and the short nights the coffee may have been covering.

Note: this article is educational and is not medical advice. A headache that is new, severe, sudden, or different from your usual pattern needs a clinician rather than a caffeine explanation. If you are pregnant, take regular medication, or have a heart or anxiety condition, talk to your doctor before changing your caffeine intake.

What are the symptoms of caffeine withdrawal?

The most useful source here is a systematic review by Laura Juliano and Roland Griffiths at Johns Hopkins, published in Psychopharmacology in 2004. They went through 57 experimental and 9 survey studies and asked a strict question: which reported symptoms hold up under controlled, preferably blinded, conditions? (doi.org)

Of 49 symptom categories reported somewhere, 10 met their validity criteria: headache, fatigue, decreased energy, decreased alertness, drowsiness, decreased contentedness, depressed mood, difficulty concentrating, irritability, and feeling foggy or "not clear-headed". They judged three more — flu-like feelings, nausea or vomiting, and muscle pain or stiffness — as likely to be real, with less evidence behind them. That group is why some people describe the second day off coffee as "like coming down with something".

The headline number is headache: it appeared in 50% of people across the experimental studies. A smaller group, 13%, had clinically significant distress or trouble functioning — the kind of withdrawal that makes you leave work early.

This is not a folk syndrome. Caffeine withdrawal is a recognised diagnosis in the DSM-5, the manual psychiatrists use, and also in the ICD-10 (doi.org). The DSM-5 version requires prolonged daily use, then an abrupt stop or reduction, followed within 24 hours by three or more of five symptoms: headache; marked fatigue or drowsiness; low, irritable or dysphoric mood; difficulty concentrating; and flu-like symptoms such as nausea, vomiting or muscle pain. The symptoms also have to cause real distress or impairment, and not be better explained by something else, such as a migraine or a viral illness.

The best-known single experiment is older. In 1992, Kenneth Silverman and colleagues published a double-blind study in the New England Journal of Medicine of 62 adults whose daily intake averaged the equivalent of about 2.5 cups of coffee — hardly heavy. In two separate two-day periods they ate a caffeine-free diet and took capsules that contained either their usual amount of caffeine or a placebo, without knowing which. (doi.org)

On the placebo days:

• 52% reported a moderate or severe headache, against 2% at baseline and 6% on caffeine;

• 11% had abnormally high depression scores and 11% abnormally low vigour;

• 8% had high anxiety scores and 8% high fatigue scores;

• 13% took medication they had not been cleared to use, against 2% on caffeine;

• finger-tapping speed, a simple test of motor performance, slowed.

Because nobody knew when they were off caffeine, expectation is unlikely to explain these effects. You are not imagining the fog because you read that you would feel foggy — and across the wider literature, Juliano and Griffiths also concluded that expectancy is not a prime driver of withdrawal.

Why does quitting caffeine cause a headache?

The short version: caffeine blocks a signal your brain produces all day, and when you use it daily, the brain turns that signal up to compensate. Take the caffeine away and the turned-up signal is suddenly unopposed.

The signal is adenosine. Caffeine sits on adenosine receptors and blocks them. With regular, heavy use, the brain can upregulate adenosine receptors, becoming more sensitive to normal adenosine. When caffeine intake drops suddenly, adenosine activity jumps, which leads to widening of blood vessels in the brain and less stimulation centrally (NCBI Bookshelf).

That blood-vessel part has been measured directly. In a double-blind study published in 2009, Stacey Sigmon and colleagues kept 16 regular caffeine users on daily caffeine for at least 14 days, and on placebo for at least 14 days, and used ultrasound to measure blood flow speed in major arteries of the brain. Twenty-four hours without caffeine increased blood flow velocity in the middle and anterior cerebral arteries. EEG showed more slow, drowsy theta activity, and people rated themselves more tired, sluggish and weary, and less energetic and lively. (doi.org)

Two more findings from that study are easy to miss:

Tolerance was only partial. People stopped feeling much from their daily dose, but the physiological effects — on blood flow and brain activity — kept showing up. Your body keeps responding to caffeine even when you no longer notice it.

Daily caffeine gave almost no net benefit on those measures, compared with being fully off it. That fits Juliano and Griffiths' conclusion that avoiding withdrawal is a central reason people keep drinking caffeine daily. For a habitual drinker, part of the morning lift is not a boost above normal. It is relief from a small overnight withdrawal.

What the headache usually feels like. Juliano and Griffiths describe it as gradual in onset, diffuse and throbbing, and made worse by physical exertion and straining. It can be mild enough to ignore or strong enough to send you to bed. Citing early clinical work, the same review describes it as distinct from a migraine attack; if you are not sure which you are dealing with, telling a headache from a migraine walks through the differences.

How headache specialists define it. The International Classification of Headache Disorders (ICHD-3) — the reference headache specialists use — has a specific entry, caffeine-withdrawal headache. It applies after regular daily caffeine above a minimum amount set in the criteria, for more than two weeks, which is then interrupted or delayed. The headache develops within 24 hours of the last caffeine, and either eases within an hour of taking a set amount of caffeine, or resolves within 7 days of stopping completely. (doi.org)

Note the word delayed. You do not have to quit to get this headache; a morning coffee much later than usual can be enough.

How long does caffeine withdrawal last?

For most people, from two days to just over a week. The review's typical pattern: onset 12–24 hours after your last caffeine, peak at 20–51 hours, and a total duration of 2–9 days. The ICHD-3 definition of the headache expects it to resolve within 7 days of stopping completely.

In practical terms, if your last coffee was on Monday morning:

Monday evening to Tuesday morning (12–24 hours). The first symptoms usually appear: a building headache, heavier eyelids, less patience.

Tuesday and Wednesday (roughly 20–51 hours). The peak window, and the hardest stretch for most people. Headache, low energy, patchy concentration and irritability tend to be at their worst here.

Thursday to the following week (up to about 9 days). Symptoms fade. For many people this happens within a few days; for some, tiredness lingers toward the end of the range.

What makes it harder. Juliano and Griffiths found that the incidence and severity of symptoms rise with daily dose. More caffeine, more withdrawal — which is also why clinical guidance prefers tapering to an abrupt stop. Anything else in the same days — short nights, a heavy work stretch, a cold — piles on and blurs the picture.

If it lasts longer than about two weeks, or the headache is getting worse rather than better, step back from the caffeine explanation. Withdrawal follows a curve; a headache that does not follow it needs a different explanation.

Can one cup a day cause withdrawal?

Yes. This is the part that surprises people most.

In the Juliano and Griffiths review, stopping from intakes as low as roughly one cup of coffee a day produced symptoms. (The review states the threshold in milligrams of caffeine; it is slightly below what the 2023 sleep meta-analysis cited further down counts for a 250 mL cup of coffee.) The Silverman study, as above, was in people drinking about two and a half cups daily — a low to moderate habit by American standards — and half of them got a real headache.

The review also found that physical dependence can develop quickly: in one controlled study, switching to placebo after just 3 days of a fairly high daily caffeine intake was enough to produce low mood, fatigue, tiredness and flu-like symptoms, and a clear headache appeared after 7 or more days of daily caffeine. So even a short run of heavy coffee days during a deadline can be enough to set up withdrawal symptoms when it stops.

Coffee is also not the only source. Tea, cola, energy drinks, pre-workout powders and some over-the-counter headache products contain caffeine — worth checking on the label, because reaching for one during withdrawal quietly puts back what you were removing.

The weekend headache, the post-surgery headache, and fasting

Two situations where caffeine withdrawal hides in plain sight, and one where it may be blamed unfairly.

The weekend headache. A headache that shows up on Saturday or Sunday, but rarely on workdays, is common enough to have a name. The main reasons are probably social or psychological, but caffeine timing is another piece. A 1992 study in Cephalalgia surveyed 151 patients at a migraine clinic; 21.9% reported attacks mainly at weekends. People with both above-average daily caffeine intake and much later waking at weekends had a 69% risk of weekend headache, compared with 4% in those with only one of the two, and none in those with moderate habits on both. Sleeping in on its own was not a significant cause. (doi.org)

The logic: a lie-in pushes your first coffee hours past its weekday time, and that delay alone may be enough. The fix is not to give up lie-ins — it is to keep your first coffee closer to your usual time, or to lower the weekday amount so the gap matters less. (If you tend to wake up with a headache on any day, not just weekends, caffeine timing is only one of several causes to rule out.)

The post-surgery headache. People fast before an operation, which usually also means no coffee. A study in Mayo Clinic Proceedings surveyed 233 surgical outpatients. 82% drank caffeinated drinks daily. Postoperative headaches occurred in 22% of daily caffeine drinkers but only 7% of those who did not drink caffeine. Among daily drinkers, those who had some caffeine on the day of surgery had fewer headaches than those who abstained — 17% versus 28%. (doi.org60691-0))

That was a survey, not a trial, and other things mattered too: a history of frequent headaches, being 50 or younger, and higher daily intake. The practical point is to tell your surgical or anaesthesia team how much caffeine you usually have. Do not decide on your own to drink coffee before a procedure; fasting rules exist for safety.

Fasting — where caffeine may be blamed unfairly. A headache during a religious fast, a fast before a blood test or an intermittent-fasting window is often put down to caffeine. Sometimes that is right. But the headache classification also has a separate entry for headache attributed to fasting: a diffuse, non-pulsating, usually mild to moderate headache that comes on after at least eight hours without food and is relieved by eating. The ICHD-3 notes that this fasting headache does not appear to be related to caffeine withdrawal, sleep duration or low blood sugar (doi.org). So if you fast and get a headache, not eating may be the cause in its own right — and keeping your coffee will not necessarily prevent it.

How to cut back without the crash

The goal is to take caffeine away more slowly than your brain can notice. Clinical guidance favours gradual tapering over several days to weeks rather than stopping abruptly (NCBI Bookshelf). For most people, one to two weeks is a workable window.

First, count what you actually have. For three or four days, change nothing and note every caffeinated drink: coffee, tea, cola, energy drinks, pre-workout, and any headache products with caffeine in them.

Cut the latest caffeine first. Keep your first cup and remove the afternoon ones before touching the morning. This does two jobs at once: it lowers your total intake, and it removes the caffeine most likely to be costing you sleep (see the next section).

Take away a small step every few days. Replace one cup with a smaller one, a half-caffeinated version or a decaf, then hold that level for a few days before the next step. If a step brings on a headache, stay at the current level for longer before going down again. The speed is yours to set — there is no prize for doing it in five days.

Keep the timing steady. Because a delayed dose can trigger the headache, drink what remains at the same times each day, including weekends.

Keep the ritual, change the contents. The warm cup, the break, the walk to the café can all stay; decaf or herbal tea keeps the habit without the drug.

Plan the hardest days. If you stop outright, avoid doing it before an exam, a long drive or a demanding shift, and let days two and three fall on a lighter stretch.

Protect sleep during the switch. Withdrawal makes you drowsy, and some of that drowsiness may be tiredness your coffee was hiding. Earlier nights in the first week help.

Be careful with pain relievers. In the Silverman study, 13% of people off caffeine reached for medication they had not been cleared to use. If you need something for the headache, ask a pharmacist or your doctor what suits you.

Or decide not to quit. Low to moderate caffeine is generally considered safe for most adults (doi.org), so a level and timing that leave your sleep alone may be a better goal than zero. If you want to understand what caffeine does to your sleep and heart rate before deciding, our piece on feeling wired and tired on caffeine covers the effects of the drug rather than of stopping it.

What happens to your sleep when you cut back?

Two things, in sequence, and the first can make you think the whole idea was a mistake.

First, you may feel sleepier. Drowsiness and fatigue are core withdrawal symptoms, and they peak in the first few days. Some of that is withdrawal itself. Some of it is likely sleep debt that caffeine was covering. A 2023 meta-analysis in Sleep Medicine Reviews notes the loop directly: people drink caffeine in response to insufficient sleep, and that caffeine may then impair the sleep that follows. When the caffeine goes, the underlying tiredness becomes visible. If that sounds like you, catching up on sleep debt explains what a run of short nights adds up to.

Then, sleep tends to improve — particularly if afternoon caffeine was part of your habit. That same meta-analysis pooled 24 studies. Caffeine reduced total sleep time by 45 minutes and sleep efficiency by 7%, lengthened the time to fall asleep by 9 minutes and added 12 minutes of wakefulness after first falling asleep. It also shifted sleep toward lighter stages: deep sleep fell by 11.4 minutes. To avoid losing total sleep time, the authors estimated that a 250 mL (roughly 8 oz) cup of coffee should be finished at least 8.8 hours before bedtime. (doi.org)

The earlier and often-quoted study by Christopher Drake and colleagues, published in the Journal of Clinical Sleep Medicine in 2013, tested a large dose — the authors note that a single 16-oz serving of brewed coffee can contain as much or more — at bedtime, 3 hours and 6 hours before bed. All three disrupted sleep compared with placebo (doi.org). Even at 6 hours, the 12 healthy sleepers lost more than an hour of objectively measured sleep — yet their sleep diaries did not pick up a significant change, suggesting they were largely unaware of the disruption.

You may not feel the sleep caffeine costs you, which is why the improvement after cutting back can also be subtle — a little less time lying awake, a little less waking in the night. That is where a record helps more than memory.

How to bring this up with your doctor

Most caffeine withdrawal never needs a clinician. A few situations do, and a few headaches that look like withdrawal are not.

Seek emergency care immediately if you have:

• a sudden, severe headache that peaks within about a minute — "the worst headache of my life";

• headache with weakness, numbness, drooping of the face, trouble speaking, confusion, loss of vision or double vision;

• headache with fever and a stiff neck, or a rash;

• headache after a head injury;

• headache with fainting or a seizure.

Book an appointment if:

• the headache has lasted beyond about two weeks, or is getting worse rather than better;

• it is a new kind of headache for you, or your usual pattern has changed;

• you are pregnant, or have a heart condition, high blood pressure, an anxiety disorder or a sleep disorder, and want to change your caffeine intake;

• low mood, anxiety or fatigue carry on well after the withdrawal window;

• you have tried to cut back several times and cannot, despite problems it is causing — the DSM-5 lists caffeine use disorder as a condition for further study, and it is a reasonable thing to raise.

What to track for a week or two before the visit:

• every caffeinated drink, with the time — including tea, soft drinks, energy drinks, pre-workout and headache products;

• when each headache started, how long it lasted and how bad it was on a 0–10 scale;

• whether caffeine or food relieved it, and how quickly;

• sleep: bedtime, wake time, and how rested you felt;

• anything else that changed in those days — illness, travel, alcohol, a stressful stretch, your cycle.

What to say: "I usually have about [number] cups of coffee and [other drinks] a day. I cut back on [date]. The headaches started [hours] later and have lasted [days]." That timeline lets a clinician check it against the withdrawal pattern in minutes.

What to ask:

• "Does this fit caffeine-withdrawal headache, or should we look for something else?"

• "Is there a pain reliever that is safe for me, and does it contain caffeine?"

• "Given my other medicines and conditions, how quickly should I taper?"

• "If I get migraines, could caffeine — or stopping it — be part of the pattern?" Our overview of common migraine triggers is a useful companion here.

Before you blame yourself, check what your body already recorded

The story people tell themselves on day two is familiar. I cannot even get through a morning without coffee. I have no willpower. Everyone else manages. Some go further: maybe I am just a tired, irritable person, and the coffee was the only thing holding it together.

The evidence says something much less personal. The feeling has a visible cause, and it is written in the timing. If your last cup was a day or two ago, you are sitting exactly where the peak of withdrawal sits. The research found the same pattern in people who did not even know they were off caffeine. Your brain spent weeks or months adapting to a daily drug; it takes days to adapt back. That doesn't make you weak. It makes you a person with a normal adenosine system.

There is often a second cause underneath the first. If coffee was carrying you through short nights, the tiredness you feel now is not new. It was there all along, and caffeine was covering it. Look back at the last few weeks of sleep — how many nights were short, how many were broken — and the fatigue of withdrawal starts to look less like a verdict on you and more like a bill arriving.

So give yourself permission to feel rough for a few days, and pick one lever — not five. For most people the best single lever is to move the last caffeinated drink of the day earlier, then hold that for a week before reducing anything else. It is easy to keep, it lowers your total, and it goes straight at sleep, which is where much of the benefit sits.

Then watch the trend for one to two weeks, not the morning. How long it takes you to fall asleep, how often you wake, how you feel at eleven in the morning. Day two is the worst possible day to judge the decision. Day ten is a fair one.

How Welltory helps — and what it cannot do

The limits first. Welltory is a general wellness product with no regulatory clearance. It does not diagnose caffeine withdrawal, headache disorders or anything else, and it does not know how much caffeine you drink unless you tell it. A headache that fits the red flags above needs a clinician, not an app.

What it can do is show the part memory handles badly: whether anything is changing. Cutting back is a slow experiment with small effects spread over a fortnight. Welltory records the signals likely to move:

• sleep — how long you slept and how regular your nights were;

• resting heart rate and heart rate variability compared with your own baseline, rather than someone else's;

• Battery, as a read on your energy through the day;

• stress minutes, the time your body showed signs of a stress response.

Tags and notes let you mark the experiment itself — "last coffee 2 pm", "switched to half-caf", "headache", "first day off caffeine". Over time, My Patterns can show which tagged situations tend to come before stressful stretches, which makes it easier to see whether, for example, late coffee days tend to come before more stress the next day; setting your tagged days against your sleep record shows whether your nights change too.

Two caveats. These signals are non-specific: alcohol, a late meal, a warm bedroom, illness and a stressful week can move sleep and heart rate as much as caffeine does, so read the pattern over one to two weeks rather than any single morning. And we do not have caffeine data across our users, so we cannot tell you what a "typical" change looks like — only what yours does.

How we made it

Made with AI tools, then edited and fact-checked by the Welltory team. See our Editorial & AI policy.

Data analysis by Jane Smorodnikova, co-founder of Welltory and the person who built the methodology behind how we read physiological data.

Written by Tatsiana Yashyna.

Discounts for blog readers: up to 36% off

See what affects your energy, stress, sleep, and daily state with Welltory

Explore your data