The figure of 400 mg a day is repeated so often that it has become detached from its source. It gets quoted as a hard limit, as a target, and occasionally as a recommendation, and it is none of those things. This article covers where the number came from, exactly what the two bodies that produced it were claiming, and the several important ways it does not apply to individuals.
Two regulators, arriving separately at the same number
The US Food and Drug Administration[1] states that for healthy adults, 400 mg a day, roughly four or five cups of coffee, is an amount not generally associated with dangerous negative effects. Working independently and reviewing the evidence on its own terms, the European Food Safety Authority reached the same figure in its 2015 Scientific Opinion on the safety of caffeine[2], concluding that intakes up to 400 mg per day consumed throughout the day do not raise safety concerns for healthy adults in the general population, except pregnant women.
Two bodies with different mandates, different data-weighting conventions, and no obligation to agree, landing on the same number is meaningful. It is the closest thing caffeine has to a consensus ceiling.
EFSA also produced a figure the FDA did not: single doses of caffeine up to 200 mg, about 3 mg per kilogram of body weight for a 70 kg adult, do not raise safety concerns for healthy adults. That distinction between a daily total and a single dose is genuinely useful, because 400 mg spread across a morning behaves quite differently from 400 mg taken at once.
What the number actually means
This is where most usage goes wrong. The 400 mg figure is a statement that no consistent evidence of harm has been demonstrated at or below that level in healthy adult populations. It is not a threshold above which harm begins, and it is not a level at which everyone is comfortable.
What the evidence behind the number looks like
It is worth knowing what these bodies were actually reading, because 400 mg is not an arbitrary round figure. EFSA's opinion assessed a wide literature covering caffeine's effects on the cardiovascular system, on the central nervous system, on bone, on pregnancy outcomes and on children and adolescents, and set its figures where the evidence for adverse effects became consistent enough to matter.
The two figures answer different questions. The 400 mg daily total addresses chronic intake, principally cardiovascular and general safety concerns. The 200 mg single-dose figure addresses acute effects, principally on sleep, anxiety and blood pressure. That distinction is more useful in daily life than the headline number, because it tells you that spreading intake across the day is materially different from concentrating it.
Neither figure is a no-effect threshold. Caffeine has measurable effects at doses far below both, on alertness, on sleep architecture and on blood pressure. The guidance describes where the evidence of harm becomes consistent, not where the evidence of any effect begins.
Plenty of people experience jitteriness, anxiety, palpitations, reflux or disrupted sleep at 200 mg, well under the guidance. That is not a contradiction of the guidance; it is what a population-level safety figure looks like when applied to individuals who vary. The relevant limit for you is whichever is lower: the regulatory ceiling, or the amount above which you feel unwell.
There is also a wide gap between the daily guidance figure and the level associated with acute toxicity, which is far higher and in practice essentially unreachable through brewed coffee. Documented caffeine toxicity cases overwhelmingly involve tablets, powders and concentrated energy products rather than coffee, because the volume of liquid required makes coffee self-limiting. This is worth knowing mainly to keep the risk in proportion.
Translating milligrams into actual drinks
The number is only useful if you can map it onto what you drink, and the mapping is looser than most charts suggest. As a working guide, an 8-ounce cup of brewed drip coffee contains roughly 95 mg of caffeine, a single espresso shot roughly 63 mg, instant coffee somewhat less than drip at around 60 to 80 mg, and brewed tea in the range of 30 to 50 mg. Cola sits around 30 to 40 mg per can, and energy drinks vary enormously.
Every one of those figures is an average concealing wide variation. Actual caffeine content shifts with the species of bean, since Robusta contains roughly double the caffeine of Arabica, with roast level, with grind size, with brew time, and above all with serving size. The 8-ounce cup that these figures assume is small by modern cafe standards; a large takeaway drink is frequently 16 ounces or more and may contain two or three espresso shots. Our guide to Arabica and Robusta covers the species difference, which is the largest single variable.
For most people the practical implication is that three or four ordinary coffees a day sits inside the guidance, while a habit built on large multi-shot cafe drinks can exceed it without feeling excessive.
Who should be below 400 mg
The guidance explicitly excludes several groups, and this is the part most often dropped when the number gets quoted.
The gap between guidance and real intake
There is a practical problem with a milligram-based limit, which is that almost nobody consumes caffeine in milligrams. They consume drinks of unknown strength, and the labelling that would let them convert generally does not exist.
Brewed coffee is the worst offender, because its caffeine content depends on species, roast, grind, dose, brew method, brew time and serving size, and none of those are printed anywhere. Two cafes pouring what they both call a large coffee can differ by a factor of two or more. Energy drinks and soft drinks are the exception in that they usually do declare caffeine content, which is why intake research often finds them easier to quantify than coffee.
The practical consequence is that anyone genuinely trying to stay under a limit has to work from conservative estimates. Assuming an ordinary cafe brewed coffee is around 150 mg rather than 95 mg, and that a large multi-shot drink may carry 200 mg or more, will keep you closer to the truth than the standard tables do.
Pregnant women are the clearest case. EFSA concluded that caffeine intakes from all sources up to 200 mg per day do not raise safety concerns for the foetus, which is half the general adult figure. We cover the evidence behind this in our article on caffeine and pregnancy, where the picture is more contested than a single number suggests.
Children and adolescents are a second group, for whom EFSA works from body weight rather than a flat figure, at around 3 mg per kilogram per day. Adolescents in particular consume a great deal of caffeine through energy drinks rather than coffee, and their intakes are frequently poorly tracked.
Beyond those, several groups have condition-specific reasons to sit lower: people with cardiac arrhythmias, with anxiety disorders, with gastro-oesophageal reflux, with insomnia, and anyone on medication that interacts with caffeine metabolism. van Dam, Hu and Willett's review[3] is explicit that general population figures do not translate into clinical guidance for people with existing conditions.
Why your personal number differs so much
The single largest reason people misapply this figure is that caffeine clearance varies severalfold between individuals. Nehlig's review in Pharmacological Reviews[4] documents this in detail: variation in the CYP1A2 enzyme is the main genetic driver, but smoking accelerates clearance substantially, oral contraceptives and pregnancy slow it considerably, and liver function matters. Reported half-lives range from around two hours to eight or more.
Children, teenagers and energy drinks
The group most likely to exceed sensible caffeine intake is not coffee drinkers at all. EFSA works from body weight for children and adolescents, at around 3 mg per kilogram per day, which for a 40 kg child is about 120 mg, and for a 55 kg teenager about 165 mg.
A single large energy drink can approach or exceed that on its own, and they are marketed to exactly this age group, sold without age restriction in most places, and frequently consumed alongside cola and iced tea. Caffeine intake in adolescents is also much harder to track than in adults, because it arrives through products that are not thought of as caffeinated in the way coffee is.
The reason this matters more in teenagers than in adults is sleep. Adolescents need more sleep than adults and their circadian rhythms already run late, so caffeine late in the day compounds a problem that biology has created. If you are worried about caffeine in a household, the energy drinks in the fridge are a more productive place to look than the coffee in the kitchen.
That means two people drinking identical amounts can have very different caffeine exposure over a day. Our article on caffeine genetics covers the evidence, including a study finding that the cardiovascular risk associated with coffee differed by genotype.
There is a second, subtler reason a daily total is not the whole story: timing. Drake and colleagues found that 400 mg taken 6 hours before bed reduced measured sleep by over an hour[5], and participants did not notice. A person drinking 300 mg spread across morning and early afternoon and a person drinking 300 mg with the last cup at 6pm are within the same daily guidance and having very different experiences. Our sleep article covers this.
Recognising when you are over your own limit
The symptoms are specific rather than vague, and worth knowing: jitteriness, tremor, a noticeably faster or irregular heartbeat, headache, restlessness, anxiety, difficulty sleeping, and in more pronounced cases nausea or stomach upset. Reflux and heartburn are common at lower doses in susceptible people.
If these appear at your usual intake, that is your body providing better information than any guideline. The response supported by the research is to reduce gradually rather than abruptly, because Juliano and Griffiths documented a validated withdrawal syndrome[6] that follows habitual intakes as low as 100 mg a day. Our article on caffeine withdrawal covers what to expect.
The summary
Two independent regulators concluded that up to 400 mg of caffeine a day is not associated with negative effects in healthy adults, with EFSA adding that single doses up to 200 mg are also unproblematic and that pregnant women should sit at 200 mg a day from all sources. These are population statements about the absence of demonstrated harm, not personal thresholds, and individual tolerance varies severalfold for reasons that are partly genetic and partly circumstantial.
It is a ceiling with room underneath it, not a target to aim at.
A note on caffeine in medicines and other sources
One category that routinely escapes people's mental accounting is medication. Caffeine is a common ingredient in over-the-counter combination painkillers, particularly those marketed for headache and migraine, where it is included because it constricts blood vessels and enhances the absorption of the analgesic alongside it.
That matters for two reasons. It contributes to a daily total that someone may believe they are tracking accurately from drinks alone. And regular use of caffeine-containing analgesics is associated with medication overuse headache, a recognised clinical problem in which the treatment sustains the condition. Someone taking such a painkiller several times a week for headaches they attribute to stress may be maintaining a cycle rather than breaking it, a mechanism covered in our article on caffeine withdrawal.
Chocolate, cocoa, some teas marketed as herbal, pre-workout supplements and weight-loss products all contribute as well. The phrase from all sources in the official guidance is doing real work.
The useful version is short. Know roughly what you are drinking, keep well clear of the ceiling if you are pregnant or have a relevant condition, and treat your own symptoms as more authoritative than the number. This is a summary of published guidance, not medical advice.
Both figures come from bodies with no commercial interest in the answer, which is worth noting in a field where a great deal of published research is industry-funded.