How long does caffeine last? Half-life, timing and sleep

The question has two answers: how long caffeine stays in the body, and how long it keeps affecting sleep. This guide covers both, using pharmacokinetic studies (which track how the body absorbs and clears a substance) and placebo-controlled sleep trials.

The European Food Safety Authority (EFSA) reports that caffeine taken by mouth is absorbed rapidly and completely, and that its stimulating effects may begin 15 to 30 minutes after ingestion and last a number of hours [30]. In a randomized crossover study of 10 healthy men given 5 mg/kg as an oral solution, plasma caffeine peaked after about 30 minutes (29.8 minutes on average) and oral bioavailability was essentially complete [1].

The drink made little difference in one small trial [2]. In a double-blind, placebo-controlled study, 13 regular users of coffee and cola took 400 mg of caffeine as coffee, cola or capsules; saliva caffeine peaked at 42 minutes after coffee and 39 minutes after cola, but later, at 67 minutes, after capsules [2].

A half-life is the time it takes for the amount of a substance in the body to fall by half. EFSA reports that in healthy adults caffeine's average half-life is approximately 4 hours, with a range of 2 to 8 hours, and that it varies with age, body weight, pregnancy, medicines and liver health [30].

Other sources reported a wider spread. In the 10-man bioavailability study, half-lives ranged from 2.7 to 9.9 hours [1], and a review of caffeine pharmacokinetics gave a range of 2.5 to 10 hours [3].

What a half-life means in practice shows up in a small sleep study. Nine healthy men took 200 mg of caffeine at 7:10 in the morning; their saliva caffeine fell from a peak of 17 µmol/l one hour after intake to 3 µmol/l, about a sixth of the peak, just before sleep at 11 p.m. [23].

The U.S. Food and Drug Administration (FDA) notes that there is wide variation both in how sensitive people are to caffeine and in how fast they eliminate it from the body [31].

Much of the variation traces to one liver enzyme [5]. A 2018 review reported that the CYP1A2 form of cytochrome P450 metabolizes 95% of the caffeine ingested, and that caffeine's metabolism and clearance are affected by age, sex and hormones, liver disease, obesity, smoking and diet [5]. A systematic analysis of caffeine pharmacokinetic data from 141 publications described large differences between individuals and analyzed smoking and oral contraceptives as two of the factors [6].

Genes and CYP1A2

A 2018 review attributes much of the variation in caffeine pharmacokinetics to a CYP1A2 polymorphism [5], but the two genotype studies below found narrower links than the "fast or slow metabolizer" label suggests. In a study of 185 healthy Caucasian non-smokers and 51 smokers given 100 mg of caffeine, a common CYP1A2 variant made no significant difference to a caffeine metabolism ratio in non-smokers; in smokers, people with the A/A genotype had a higher mean ratio (1.37) than heterozygotes (0.88) or C/C carriers (0.82) [14].

A second study, in Serbian and Swedish non-smokers, found that heavy coffee drinkers with the -163 A/A genotype had higher CYP1A2 activity, while enzyme activity did not differ between genotypes among people who were not heavy coffee drinkers [15]. Its authors estimated that the variant might explain 22% and 14% of the variation in enzyme activity among heavy coffee drinkers in the two countries [15].

A 2024 systematic review of 26 case-control studies covering 1,851,428 people found that variants in metabolism genes, including CYP1A2, were associated with differences of 3% to 32% in cups of caffeinated drink per day per effect allele [18].

Sleep effects have been linked more to a different gene [16][17]. A systematic review of 22 records in healthy adults, 15 of them randomized controlled trials, found that CYP1A2 variants were associated with cognitive function, while variants in ADORA2A, the gene for an adenosine receptor, were associated with anxiety and sleep disturbance [16]. An earlier study found that ADORA2A genotypes were distributed differently in adults who rated themselves sensitive or insensitive to caffeine's effects on sleep [17].

As of November 11, 2026 we found no randomized trials testing whether CYP1A2 genotype changes how caffeine affects sleep (PubMed and Europe PMC search, November 11, 2026). The CYP1A2 studies above measured enzyme activity or reported intake, not sleep [14][15][18].

Smoking

Smoking speeds caffeine clearance [7][8]. In one study, caffeine clearance was 114 ml per minute in 8 smokers and 64 ml per minute in 15 non-smokers without liver disease, and in healthy smokers it fell from 118 to 77 ml per minute after 3 weeks without cigarettes [7]. In a controlled study of 12 smokers, a week of smoking increased caffeine clearance compared with a week of breathing air (2.0 vs 1.5 ml per minute per kg of body weight), with no effect from carbon monoxide alone [8].

Hormonal contraceptives and the menstrual cycle

Estrogen-containing contraceptives slowed clearance in one study [9]. Women who had used low-dose estrogen oral contraceptives for more than 3 months had a mean caffeine half-life of 7.88 hours after a 162 mg dose, compared with 5.37 hours in matched controls, because plasma clearance was lower [9].

The menstrual cycle had a smaller effect [10]. In repeated 24-hour studies in 10 healthy women, caffeine clearance was slower in the luteal phase, although the half-life did not differ, and the authors wrote that the effect may be too small to matter clinically for most women [10].

Pregnancy

Caffeine lasts longer during pregnancy [11][12]. In a saliva study, the mean half-life was 3.4 hours (range 2 to 5 hours) in 25 men and non-pregnant women, compared with 8.3 hours (range 3 to 16 hours) in 57 pregnant women; after delivery, values returned to normal within one month in most cases [11]. A multicenter study of 59 pregnant women (93.2% white) found decreased caffeine metabolism during pregnancy, especially in the third trimester [12].

EFSA found that caffeine intakes from all sources up to 200 mg per day, consumed throughout the day, do not raise safety concerns for the fetus [30], and the NHS advises no more than 200 mg a day in pregnancy [32]. A dose-response meta-analysis of 53 observational studies linked higher intake in pregnancy with spontaneous abortion, stillbirth, low birth weight and small size for gestational age, found no threshold below which the associations disappeared, and noted they were modest within usual intakes and may reflect study bias [29]. Limits differ by country; the caffeine monograph lists them. Anyone who is pregnant can discuss caffeine with their midwife or doctor.

Babies

Newborns clear caffeine slowly [13]. In a study of 7 infants aged 2 1/2 weeks to 6 months, plasma clearance and half-life reached adult levels at 3 to 4 1/2 months of age, after the low elimination rates seen in newborns [13].

Medicines that slow caffeine clearance

Some medicines block CYP1A2 and can make caffeine last much longer [26][27]. A review of pharmacokinetic interactions named potent inhibitors of the enzyme, including certain selective serotonin reuptake inhibitor antidepressants (particularly fluvoxamine), the antiarrhythmic mexiletine, the antipsychotic clozapine, and quinolone antibiotics such as enoxacin [26].

In a randomized crossover study of 8 healthy volunteers, caffeine's median half-life rose from 5 to 31 hours while they took fluvoxamine [27]. In 12 healthy men, enoxacin slowed caffeine elimination about sixfold and high-range doses of ciprofloxacin about 1.5-fold [28].

If you take any of these medicines, talk to your doctor or pharmacist before combining them with caffeine. Full safety and interaction details are on the caffeine monograph.

As a mechanism, caffeine works by blocking adenosine receptors [4]. A review of cognitive and physical studies reported that low to moderate doses, from about 40 mg to about 300 mg, improved alertness, vigilance, attention and reaction time, with less consistent effects on memory and higher-order executive function [4]. Our focus hub compares caffeine with other options for attention.

For caffeine combined with L-theanine, see the caffeine and L-theanine guide and the L-theanine monograph.

Yes, in pooled trials. A 2023 meta-analysis of 24 studies found that caffeine reduced total sleep time by 45 minutes and sleep efficiency by 7%, increased the time to fall asleep by 9 minutes and time awake after falling asleep by 12 minutes, and reduced deep sleep by 11.4 minutes [19].

A 2025 meta-analysis of 22 controlled crossover trials in 956 healthy adults found effects in the same direction: total sleep time fell by 34.67 minutes and the time to fall asleep rose by 8.35 minutes compared with placebo [20]. Total sleep time fell in both young and middle-aged groups and at both high and low doses; other sleep measures changed significantly only in the young group. Differences between subgroups were not statistically significant [20].

In timing trials, a dose taken hours before bed still affected sleep that night [21][22]. In a placebo-controlled home study, 400 mg taken at bedtime, 3 hours before or 6 hours before each caused significant sleep disturbance [21]. In a 2025 double-blind randomized crossover trial, 23 young men with moderate habitual intake took placebo, 100 mg or 400 mg at 12, 8 or 4 hours before bed [22]. The 100 mg dose had no significant effect on objective or self-rated sleep compared with placebo, although EFSA notes that single doses of 100 mg may affect sleep duration and patterns in some adults, particularly close to bedtime [30]. The 400 mg dose delayed sleep onset and changed sleep structure when taken within 12 hours of bedtime, and fragmented sleep more when taken within 8 hours [22].

One small trial tested a single morning dose. In 9 healthy men, 200 mg at 7:10 a.m. reduced sleep efficiency and total sleep time that night compared with placebo [23].

These results come from healthy adults. See our sleep hub for other sleep evidence.

Studies report how sleep changed at a given dose and time; they do not set a cutoff for any one person. From pooled data, the 2023 meta-analysis's model estimated that total sleep time would not be reduced if a coffee containing 107 mg were consumed at least 8.8 hours before bedtime, or a 217.5 mg pre-workout serving at least 13.2 hours before [19].

In the 2025 trial, 400 mg affected sleep at every timing tested, up to 12 hours before bed, while 100 mg did not significantly change sleep even at 4 hours [22]. That trial was small and included only young men [22]. Because clearance varies with CYP1A2 activity, smoking, hormones, pregnancy and medicines [5][9][11], group averages may not predict how any one person responds.

In a review of 57 experimental and 9 survey studies in habitual caffeine users, withdrawal symptoms typically began 12 to 24 hours after stopping caffeine, peaked at 20 to 51 hours and lasted 2 to 9 days [25]. Headache occurred in 50% of people in the experimental studies, and stopping doses as low as 100 mg/day produced symptoms [25].

• "Coffee hits faster than cola." In a placebo-controlled study of 13 users, peak saliva caffeine and time to peak did not appear to differ between coffee and cola [2].
• "A genetic test can set your caffeine cutoff." As of November 11, 2026 we found no such trial (see Genes and CYP1A2), and the variant studied was linked to differences in measured enzyme activity mainly in smokers and heavy coffee drinkers [14][15].
• "You can tell whether caffeine hurt your sleep." Results depend on the measure used. In the 2025 timing trial, 400 mg changed objective sleep 12 hours before bed, but self-rated sleep quality fell significantly only at 4 hours (by 34.02%) [22]. A meta-analysis of 10 studies in 128 mostly male athletes found the opposite gap: athletes consistently reported poorer sleep after evening caffeine, while measured changes were small and not robust; the authors urged caution because the studies were few [24].

Grades on this page come from our evidence matrix and are pulled live, not retyped. Our page on how we grade evidence explains the rubric and what each grade means.

How long does caffeine stay in your system?

EFSA puts the average half-life in healthy adults at about 4 hours (range 2 to 8 hours) [30]. In 9 healthy men, saliva caffeine fell to about a sixth of its peak between 1 hour after a 200 mg dose at 7:10 a.m. and 11 p.m. [23].

How long after coffee does caffeine peak?

In 13 regular users, saliva caffeine peaked 39 to 42 minutes after cola or coffee [2]; in 10 healthy men given an oral solution, plasma levels peaked at about 30 minutes [1].

How many hours before bed does caffeine affect sleep?

A meta-analysis's model estimated that total sleep time would not be reduced if a 107 mg coffee were consumed at least 8.8 hours before bedtime [19]. In a trial of 23 young men, 100 mg did not significantly change sleep even 4 hours before bed, while 400 mg affected sleep when taken up to 12 hours before [22]. That trial was small, and EFSA notes that single 100 mg doses may affect sleep in some adults, particularly close to bedtime [30].

Does a CYP1A2 gene test tell me how long caffeine lasts for me?

CYP1A2 gene variants were linked to differences in measured enzyme activity mainly in smokers and heavy coffee drinkers, not in non-smokers overall [14][15]. As of November 11, 2026 we found no randomized trials testing whether CYP1A2 genotype changes how caffeine affects sleep (PubMed and Europe PMC search, November 11, 2026).