Caffeine
Caffeine is the stimulant in coffee, tea, cola and energy drinks. At up to 400 mg a day it is safe for most healthy adults; its real costs are lost sleep, a blood-pressure rise that lasts hours, interactions with a handful of medicines, and pregnancy risks that, in observational studies, rise with the dose.
Keep to 400 mg a day and no more than 200 mg at once, or 200 mg a day in pregnancy. Size your bedtime cut-off to the dose: one afternoon cup is fine for most people, but a heavy day's total wants 10–12 hours. If you want to cut back, taper rather than stop dead.
What it is and how fast you clear it
A brewed cup of coffee (240 mL) carries 75–100 mg of caffeine, a single espresso 60–80 mg and a cup of decaf 2–15 mg. Tea usually carries less, roughly 15–70 mg a cup depending on type and brewing, and a teaspoon of matcha 60–80 mg; an analysis of 20 commercial teas found 14 to 61 mg per serving with no consistent trend between white, green and black, because steeping time matters more than the variety.[1] Caffeine is cleared almost entirely by one liver enzyme, called CYP1A2, whose activity varies several-fold between people. Its half-life — the time your body takes to clear half a dose — is usually 3–6 hours in healthy adults but ranges from about 2 to about 10, depending on your genes, whether you smoke, and pregnancy. That spread is why no single cut-off time suits everyone.
How much is safe
Europe's food-safety authority (EFSA), after a full review of caffeine's toxicology, and the US Food and Drug Administration both put the safe daily amount for healthy adults at 400 mg — about four brewed cups. EFSA adds no more than 200 mg in a single dose.[2] Above 400 mg a day, anxiety, insomnia, palpitations and stomach upset become more likely.
Some people react badly well below that. Caffeine often worsens anxiety, and 400 mg or more can trigger panic attacks in susceptible people. A common variant in the gene for caffeine's target in the brain, the adenosine A2A receptor, governs sensitivity rather than clearance: carriers of one genotype report markedly more caffeine-induced anxiety.[3]
For children, the American Academy of Pediatrics sets no milligram figure: it says caffeine and other stimulants contained in energy drinks "have no place in the diet of children and adolescents."[4] The often-quoted limits of 100 mg a day for teenagers and 2.5 mg per kilogram for younger children come from a separate review in the same journal, drawing on Health Canada's numbers.[5]
Sleep: the dose decides the cut-off
The popular "six hours before bed is fine" doesn't hold for big doses. In a controlled trial, 400 mg taken six hours before bed still cut total sleep by more than an hour, and the sleepers didn't notice it in their own ratings.[6] A later crossover trial tested dose and timing together: 100 mg, about one cup, had no measurable effect even four hours before bed, while 400 mg altered the structure of sleep when taken within 12 hours of bedtime and fragmented it within 8.[7]
So a single afternoon cup is defensible for most people, but a heavy total wants a 10–12-hour buffer — stricter still if caffeine lingers in you or you sleep badly. In a trial in everyday life, people slept about 36 minutes less on days they drank coffee.[8] See circadian rhythms.
Blood pressure, the heart and your genes
Caffeine raises blood pressure for hours. In people who already have high blood pressure, 200–300 mg raised it by roughly 8 mmHg systolic and 6 diastolic for at least three hours.[9] Daily use blunts this only partly: in a four-week crossover trial in healthy adults, a 250 mg dose still raised blood pressure at every habitual intake tested, up to 600 mg a day, and about half the participants kept the full response.[10] A companion trial with all-day monitoring was titled "Caffeine tolerance is incomplete."[11] If your high blood pressure is severe or poorly controlled, keep caffeine moderate, and don't have it shortly before a blood-pressure reading: the rise distorts the result.
Eye pressure, the main changeable risk for glaucoma, is a different story: in about 121,000 UK Biobank participants, the heaviest caffeine users had eye pressure just 0.1 mmHg lower than the lightest, a meaningless difference, and a genetic analysis found no causal effect of coffee on it. The exception was the quarter of people with the highest genetic risk of raised eye pressure: in them, more than about 480 mg a day went with slightly higher eye pressure, and 321 mg or more with about four times the glaucoma prevalence of low-risk non-drinkers, a cross-sectional comparison that mixes genes with coffee.[12]
On heart rhythm, the trials on coffee are reassuring for atrial fibrillation but found more extra beats from the heart's lower chambers on coffee days; the DECAF trial page covers both.
On current evidence, genetic "slow metaboliser" tests don't change the advice. Roughly half the population carries a common variant that slows caffeine clearance. A 2006 Costa Rican heart-attack study found slow metabolisers drinking four or more cups a day at about 64% higher risk of a non-fatal heart attack, while fast metabolisers showed neutral-to-protective results.[13] That has not held up at scale. In 347,077 people in UK Biobank, the large British health study, neither the gene nor a broader caffeine-metabolism score was linked to cardiovascular disease, and neither changed coffee's association with it;[14] in 498,134 people the mortality benefit held regardless of genetic score;[15] and a 34,992-person heart-imaging study concluded that customary caffeine limits aimed at reducing arrhythmia risk "might be unnecessary."[16]
Pregnancy and breastfeeding
Pregnancy is where the limit matters most. EFSA concluded that up to 200 mg a day, counting all sources — tea, cola, chocolate and energy drinks — raises no safety concern for the fetus. The American College of Obstetricians and Gynecologists says intake below 200 mg "does not appear to be a major contributing factor in miscarriage or preterm birth," and leaves growth restriction undetermined.[17] The FDA has issued no pregnancy figure.
The observational evidence behind the limit is graded, with no clean cut-off. Pooling 53 studies, each extra 100 mg a day went with about 14% higher risk of miscarriage, 19% of stillbirth, 10% of a small-for-dates baby and 7% of low birth weight, with no threshold below which the links disappeared and no link with preterm birth. The authors note these modest associations could be produced by study bias, and recommend keeping the current limit rather than tightening it.[18] The only trial evidence is close to empty: a Cochrane review found a single study with usable data, too little to confirm or refute an effect on birthweight or preterm birth.[19] So 200 mg is a precaution, not a measured safe line. Clearance roughly halves late in pregnancy, and the fetus has essentially no capacity to clear caffeine itself.
Breastfeeding is more forgiving. Up to 200–300 mg a day is generally considered compatible with nursing: EFSA sets 200 mg as its level of no concern, while the US National Institutes of Health lactation database puts a safe range at 300–500 mg for most mothers. Infant fussiness, jitteriness and poor sleep have been reported at very high intakes, equivalent to about ten or more cups a day; the one cohort that tested more than 300 mg a day found slightly more night waking, short of statistical significance, and a double-blind trial adding 500 mg a day for five days found no effect on infant sleep or heart rate.[20] Newborns clear caffeine extremely slowly, with a half-life of about 100 hours,[21] and reach adult clearance only by 3–5 months of age.[22]
Medicines and medical tests
Because caffeine and several drugs share the CYP1A2 enzyme, they can change each other's levels.
- Clozapine. In the largest clinical dataset, 16,558 blood samples from 8,833 patients, coffee had no discernible effect on clozapine dose or blood levels once smoking was accounted for.[23] What matters is the enzyme's activity: a measure of it explains up to 15% of the variation in clozapine levels between patients, and each one-unit rise carried about 26% higher odds of hospital admission.[24] Smoking speeds up that enzyme, and coffee drinking tracks with smoking, so the practical rule is about keeping your smoking steady: tell your prescriber before you start, stop or cut down smoking.
- Fluvoxamine, ciprofloxacin, oral contraceptives, verapamil and cimetidine slow caffeine clearance, so the same cup hits harder. Theophylline shares the pathway, so the stimulant effects add up.
- Levothyroxine and alendronate are absorbed less well when swallowed with coffee; separate them by 30–60 minutes.
- Oral iron is blocked by coffee's polyphenols rather than its caffeine, and only in one direction: have the coffee first and the iron an hour later. See Coffee.
- Adenosine heart stress tests. Caffeine blocks the drug used in the test, and the result is a scan that can miss a problem. European cardiac-imaging guidance is to stop all caffeine-containing drinks, foods and medicines for at least 12 hours beforehand, and to be wary even of decaf.[25] Follow your centre's instructions.
Exercise, and what happens when you stop
Exercise is one of the few benefits tested in randomised trials. Pooling 44 studies, 3–6 mg of caffeine per kilogram of body weight — about 210–420 mg for a 70-kg adult — taken about an hour before exercise improved endurance performance by roughly 2–3%.[26] Doses as low as 3 mg per kilogram are said to give most of the benefit with fewer side effects. That low end is about EFSA's 200 mg single-dose level for a 70-kg adult; the top of the range is about twice it.
Withdrawal is real but mild and short. It is a recognised diagnosis in the psychiatric manual DSM-5. In experimental studies, about half of people got a headache when they stopped, and roughly one in eight had distress or trouble functioning; fatigue, low alertness and poor concentration were also common. Symptoms start 12–24 hours after the last dose, peak at one to two days and last two to nine days, and a dose of caffeine relieves them within 30–60 minutes.[27]
What to do
- Count your total. Under 400 mg a day from all sources and no more than 200 mg at once; 200 mg a day if you are pregnant.
- Match the cut-off to the dose. One cup in the afternoon is fine for most people; a heavy day's total needs 10–12 hours before bed.
- Check your medicines against the list above, and ask before an adenosine stress test. Keep caffeine moderate if your blood pressure is poorly controlled.
- Skip the gene test. On current evidence it wouldn't change what you should do.
- Taper, don't quit. Cut about 25% every three to four days to avoid withdrawal headaches.
For caffeine's main source, and what else is in the cup, see Coffee.
Further reading
- EFSA Scientific Opinion on the safety of caffeine.[28]
- Gardiner CL et al. Dose and timing effects of caffeine on subsequent sleep: a randomized clinical crossover trial. Sleep 2025.[29]
- Greenwood DC et al. Caffeine intake during pregnancy and adverse birth outcomes: a systematic review and dose-response meta-analysis. Eur J Epidemiol 2014.[30]
- Jahanfar S, Jaafar SH. Effects of restricted caffeine intake by mother on fetal, neonatal and pregnancy outcomes. Cochrane Database Syst Rev 2015.[31]
- Juliano LM, Griffiths RR. A critical review of caffeine withdrawal. Psychopharmacology 2004.[32]