Caffeine Half-Life: How Long It Lasts and When to Stop

Caffeine’s half-life, the time your body takes to clear half of a dose, is about 3 to 6 hours in most healthy adults. Individual values run from about 2 to nearly 10 hours.¹² Pregnancy, birth control pills, liver disease and some medications can push it past 10 hours. Smokers clear caffeine faster.³

Caffeine half-life in healthy adults

Caffeine is absorbed fast and almost completely. In 10 healthy men given caffeine in water, blood levels peaked about 30 minutes later and bioavailability was essentially 100%.² Half-lives in that small group ranged from 2.7 to 9.9 hours.

A saliva study of 25 men and non-pregnant women found a mean of 3.4 hours, with a range of 2 to 5.¹ Averages in other small groups of healthy adults sit between about 4 and 6 hours.³⁴⁵ The table pulls the numbers together.

Group	Average half-life	Study size
Healthy men and non-pregnant women	3.4 h (range 2 to 5 h)	251
Non-smokers vs smokers	6.0 h vs 3.5 h	13 vs 133
Women not on vs on the pill	6.2 h vs 10.7 h	9 vs 96
Pregnant women	8.3 h (range 3 to 16 h)	571
Last 4 weeks of pregnancy	10.5 h	Saliva study7
Healthy vs liver cirrhosis	3.8 h vs 13.7 h	10 vs 154
Before vs during fluvoxamine	5 h vs 31 h	88

Most of these studies are small and decades old. They agree on direction, and the spread between people is large enough that a single “caffeine lasts X hours” figure misleads.

How long does caffeine stay in your system?

Each half-life removes half of what is left. After one half-life, 50% of a dose remains. After two, 25%. After four, about 6%.

The table shows the share of any drink still in circulation, for three half-lives that match the groups above. Multiply by your drink’s milligrams: a 200 mg coffee at 33% leaves 66 mg.

Hours since drink	3 h half-life	5 h half-life	10.5 h half-life
4	40%	57%	77%
6	25%	44%	67%
8	16%	33%	59%
10	10%	25%	52%
12	6%	19%	45%

These figures are simple decay math. They ignore the time caffeine takes to peak, which ran from about 30 minutes to 1.5 hours in the studies above, so real levels late in the day run slightly higher.²⁵

A worked example: a 200 mg coffee-shop coffee at 2 pm, with a 5-hour half-life, leaves about 66 mg at 10 pm. That is about two cans of cola. The caffeine calculator runs this math for your own drinks, half-life and bedtime.

What makes caffeine’s half-life longer or shorter?

One liver enzyme, CYP1A2, metabolizes about 95% of the caffeine you take in.⁹ Anything that speeds it up or slows it down changes how long a coffee lasts.

Pregnancy

Half-life climbs during pregnancy, and the rise starts in the first part of it. In one study it went from an average of 3 hours in non-pregnant women to 10.5 hours in the last 4 weeks of pregnancy.⁷ Across 57 pregnant women the range was 3 to 16 hours, and in most cases values returned to normal within a month of delivery.¹

At a 10.5-hour half-life, about 40% of an 8 am coffee is still circulating at 10 pm.

Birth control pills

The pill slows caffeine clearance. One study measured average half-lives of 10.7 hours in 9 pill users vs 6.2 hours in 9 non-users.⁶ Another, in women on low-dose estrogen pills for over 3 months, found 7.9 vs 5.4 hours, and caffeine took longer to peak (about 1.5 vs 0.8 hours).⁵ A twin study of 378 Danes also found lower CYP1A2 activity in women on the pill.¹⁰

Smoking, and quitting

Tobacco smoke speeds up caffeine breakdown. In one study smokers had an average half-life of 3.5 hours vs 6.0 hours in non-smokers (13 per group).³

The flip side matters if you quit. When 7 healthy smokers stopped for 3 weeks, their caffeine clearance fell by about a third, from 118 to 77 ml per minute.¹¹ The same coffee habit then leaves more caffeine in your system at night.

Your genes (CYP1A2)

How fast your CYP1A2 enzyme runs is largely inherited. In the Danish twin study, genes accounted for an estimated 73% of the variation in CYP1A2 activity among non-smokers not on the pill.¹⁰ The enzyme was also more active in men than in women. Common variants in the CYP1A2 gene add to the spread between people.⁹

Liver disease

Because the liver does the work, liver disease slows clearance. In 15 people with cirrhosis, half-life averaged 13.7 hours vs 3.8 hours in 10 healthy volunteers, with very wide variation between patients.⁴

Medications that block caffeine breakdown

Fluvoxamine, an SSRI antidepressant, strongly blocks CYP1A2. In 8 healthy volunteers it raised caffeine’s median half-life from 5 to 31 hours, and the authors warned that caffeine during fluvoxamine treatment could lead to caffeine intoxication.⁸ In a second trial of 7 people, half-life went from 4.9 to 56 hours.¹² If you take a prescription drug, ask your pharmacist whether it interacts with caffeine.

When to stop drinking caffeine before bed

Three studies give the most useful cutoffs. They differ in dose, and dose turns out to matter as much as timing.

6 hours (Drake 2013). In a randomized, double-blind, placebo-controlled crossover study of 12 healthy sleepers, 400 mg of caffeine taken 0, 3 or 6 hours before bed disturbed sleep at every time point.¹³ Even at 6 hours, objectively measured sleep fell by more than an hour, while diary-reported sleep fell by 41 minutes.

The authors recommended no substantial caffeine for at least 6 hours before bed. For scale, 400 mg is about four 8 oz cups of brewed coffee.¹⁴

8.8 hours for a coffee, 13.2 for a pre-workout (Gardiner 2023). A meta-analysis of 24 studies found caffeine reduced total sleep by 45 minutes and sleep efficiency by 7%.¹⁵ It added 9 minutes to falling asleep and 12 minutes of waking during the night, and cut deep sleep by about 11 minutes. To avoid losing sleep time, the authors calculated that a 107 mg coffee should come at least 8.8 hours before bed and a 217.5 mg pre-workout serving at least 13.2 hours before.

100 mg vs 400 mg (Gardiner 2025). In a double-blind crossover trial, 23 young men who usually had under 300 mg a day took 100 or 400 mg at 12, 8 or 4 hours before bed.¹⁶ The 100 mg dose had no measurable effect, even at 4 hours. The 400 mg dose delayed sleep onset and cut deep sleep even when taken 12 hours before bed, and fragmented sleep when taken 8 hours or less before.

Putting it together:

• One regular coffee (about 100 mg): stop about 9 hours before bed. For an 11 pm bedtime, that is 2 pm.
• Large coffee-shop drinks, energy drinks, anything near 200 mg or more: 12 to 13 hours, which for most people means morning only.
• Pregnant, on the pill, on a CYP1A2-blocking drug, or recently quit smoking: add margin, because the same drink lasts longer.

Count back from the time you actually go to sleep. The sleep calculator helps pick that bedtime. On nights, use your planned sleep time instead of the clock; the shift work sleep guide covers the rest.

Does caffeine shift your body clock?

It did in a tightly controlled lab study. In a double-blind, placebo-controlled protocol lasting about 49 days per person, 5 adults took a dose equal to a double espresso (2.9 mg per kg, about 200 mg for a 69 kg person) 3 hours before bedtime.¹⁷ Their melatonin rhythm shifted about 40 minutes later. That is nearly half the delay caused by 3 hours of bright evening light at about 3,000 lux.

The same paper found caffeine lengthened the clock’s cycle in human cells by acting on adenosine receptors. Five people is a small sample, so treat 40 minutes as an estimate rather than a fixed number.

The authors suggested avoiding evening caffeine could help people whose sleep timing has drifted late. If that is you, see how to fix your sleep schedule and the morning sunlight guide for the light side. They also floated well-timed caffeine as a possible aid for the large delays westward jet lag requires, which still needs testing.

The same comparison shows that evening light delayed the clock about twice as much as caffeine did. If you are cutting evening caffeine to protect your clock, bright evening light is the bigger lever.¹⁷ Our circadian light bulbs page compares warm, dimmable options.

How much caffeine is in coffee, tea and soda?

Drink	Serving	Caffeine
Espresso	1 fl oz shot	about 63 mg14
Brewed coffee	8 fl oz	about 95 mg14
Brewed coffee	12 fl oz	113 to 247 mg18
Coffee-shop brewed coffee	16 fl oz	188 mg average19
Black tea	8 fl oz	about 47 mg14
Black tea	12 fl oz	71 mg18
Green tea	1 cup	about 29 mg14
Green tea	12 fl oz	37 mg18
Cola	12 fl oz can	about 33 mg14
Caffeinated soft drinks	12 fl oz	23 to 83 mg18
Energy drinks	16 fl oz	54 to 328 mg18
Decaf coffee	8 fl oz	2 to 15 mg18

Coffee is the hardest to pin down. In a lab analysis of coffee-shop drinks, caffeinated coffees ranged from 58 to 259 mg per serving, and the same 16 oz drink from the same shop ranged from 259 to 564 mg over six consecutive days.¹⁹ If your afternoon cup comes from a café, assume the high end.

Decaf is low but not zero: 2 to 15 mg per 8 oz cup.¹⁸

Why some people sleep fine after coffee

Part of it is clearance: a smoker with a 3-hour half-life and a pill user with a 10-hour half-life are living with very different evening caffeine levels.

Part of it is the brain. Caffeine works on adenosine receptors, and sensitivity there differs.¹⁷²⁰ People who call themselves caffeine-sensitive and those who don’t tend to carry different versions of the adenosine A2A receptor gene, ADORA2A.²⁰ In that study, the gene version predicted how closely caffeine’s effect on sleep brain waves resembled the pattern seen in insomnia.

And part of it is perception. In the 2025 trial, 400 mg changed objectively measured sleep up to 12 hours later, but people only rated their sleep quality worse when they took it 4 hours before bed. Their sleep diaries did not pick up the extra night waking compared with placebo.¹⁶ Feeling fine is not proof that caffeine left your sleep alone.

Who should be more careful with caffeine

• Pregnant or trying to conceive. Average half-life went from about 3 to 10.5 hours late in pregnancy in one study.⁷ The FDA advises talking with your health care provider about whether to limit caffeine.¹⁸
• On birth control pills. Expect a longer half-life and an earlier cutoff.⁶⁵
• Taking fluvoxamine or other drugs that affect CYP1A2. Ask your pharmacist.⁸
• Liver disease. Half-life averaged more than three times normal in cirrhosis.⁴
• Recently quit smoking. The same intake now lasts longer.¹¹
• Using caffeine pills or powder. The FDA notes toxic effects such as seizures can occur with rapid intake of about 1,200 mg, less than half a teaspoon of pure caffeine.¹⁸

For most healthy adults, the FDA cites 400 mg a day as an amount not generally associated with negative effects.¹⁸ This guide is general information, not medical advice.

Footnotes

1. Knutti R, Rothweiler H, Schlatter C. Effect of pregnancy on the pharmacokinetics of caffeine. European Journal of Clinical Pharmacology. 1981;21(2):121-126. doi:10.1007/BF00637512 ↩ ↩² ↩³ ↩⁴ ↩⁵

2. Blanchard J, Sawers SJ. The absolute bioavailability of caffeine in man. European Journal of Clinical Pharmacology. 1983;24(1):93-98. doi:10.1007/BF00613933 ↩ ↩² ↩³

3. Parsons WD, Neims AH. Effect of smoking on caffeine clearance. Clinical Pharmacology & Therapeutics. 1978;24(1):40-45. doi:10.1002/cpt197824140 ↩ ↩² ↩³ ↩⁴

4. Renner E, Wietholtz H, Huguenin P, Arnaud MJ, Preisig R. Caffeine: a model compound for measuring liver function. Hepatology. 1984;4(1):38-46. doi:10.1002/hep.1840040107 ↩ ↩² ↩³ ↩⁴

5. Abernethy DR, Todd EL. Impairment of caffeine clearance by chronic use of low-dose oestrogen-containing oral contraceptives. European Journal of Clinical Pharmacology. 1985;28(4):425-428. doi:10.1007/BF00544361 ↩ ↩² ↩³ ↩⁴

6. Patwardhan RV, Desmond PV, Johnson RF, Schenker S. Impaired elimination of caffeine by oral contraceptive steroids. Journal of Laboratory and Clinical Medicine. 1980;95(4):603-608. PubMed 7359014 ↩ ↩² ↩³

7. Knutti R, Rothweiler H, Schlatter C. The effect of pregnancy on the pharmacokinetics of caffeine. Archives of Toxicology Supplement. 1982;5:187-192. doi:10.1007/978-3-642-68511-8_33 ↩ ↩² ↩³

8. Jeppesen U, Loft S, Poulsen HE, Brøsen K. A fluvoxamine-caffeine interaction study. Pharmacogenetics. 1996;6(3):213-222. doi:10.1097/00008571-199606000-00003 ↩ ↩² ↩³

9. Nehlig A. Interindividual differences in caffeine metabolism and factors driving caffeine consumption. Pharmacological Reviews. 2018;70(2):384-411. doi:10.1124/pr.117.014407 ↩ ↩²

10. Rasmussen BB, Brix TH, Kyvik KO, Brøsen K. The interindividual differences in the 3-demethylation of caffeine alias CYP1A2 is determined by both genetic and environmental factors. Pharmacogenetics. 2002;12(6):473-478. doi:10.1097/00008571-200208000-00008 ↩ ↩²

11. Joeres R, Klinker H, Heusler H, Epping J, Zilly W, Richter E. Influence of smoking on caffeine elimination in healthy volunteers and in patients with alcoholic liver cirrhosis. Hepatology. 1988;8(3):575-579. doi:10.1002/hep.1840080323 ↩ ↩²

12. Culm-Merdek KE, von Moltke LL, Harmatz JS, Greenblatt DJ. Fluvoxamine impairs single-dose caffeine clearance without altering caffeine pharmacodynamics. British Journal of Clinical Pharmacology. 2005;60(5):486-493. doi:10.1111/j.1365-2125.2005.02467.x ↩

13. Drake C, Roehrs T, Shambroom J, Roth T. Caffeine effects on sleep taken 0, 3, or 6 hours before going to bed. Journal of Clinical Sleep Medicine. 2013;9(11):1195-1200. doi:10.5664/jcsm.3170 ↩

14. US Department of Agriculture, FoodData Central (SR Legacy), caffeine per 100 g multiplied by listed portion weight: espresso, brewed coffee, black tea, green tea, cola. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

15. Gardiner C, Weakley J, Burke LM, et al. The effect of caffeine on subsequent sleep: a systematic review and meta-analysis. Sleep Medicine Reviews. 2023;69:101764. doi:10.1016/j.smrv.2023.101764 ↩

16. Gardiner CL, Weakley J, Burke LM, et al. Dose and timing effects of caffeine on subsequent sleep: a randomized clinical crossover trial. Sleep. 2025;48(4):zsae230. doi:10.1093/sleep/zsae230 ↩ ↩²

17. Burke TM, Markwald RR, McHill AW, et al. Effects of caffeine on the human circadian clock in vivo and in vitro. Science Translational Medicine. 2015;7(305):305ra146. doi:10.1126/scitranslmed.aac5125 ↩ ↩² ↩³

18. US Food and Drug Administration. Spilling the beans: how much caffeine is too much? Updated August 28, 2024. fda.gov ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰

19. McCusker RR, Goldberger BA, Cone EJ. Caffeine content of specialty coffees. Journal of Analytical Toxicology. 2003;27(7):520-522. doi:10.1093/jat/27.7.520 ↩ ↩²

20. Rétey JV, Adam M, Khatami R, et al. A genetic variation in the adenosine A2A receptor gene (ADORA2A) contributes to individual sensitivity to caffeine effects on sleep. Clinical Pharmacology & Therapeutics. 2007;81(5):692-698. doi:10.1038/sj.clpt.6100102 ↩ ↩²