Skip to content
Run

Caffeine for runners: what's being used and what the research says

We report what runners are using and what the research says. We don't recommend taking any supplement. Talk to your GP or a registered sports dietitian before trying one.

Quick answer

Research suggests yes, modestly. A meta-analysis of 21 trials in runners found caffeine extended time to exhaustion by about 17% and made time trials about 0.7% faster on average[1]. Studies consistently found benefits at 3 to 6 mg per kg of body mass, usually taken an hour before[2], and responses vary between people[3].

Research
Strong (proposed)Longer time to exhaustion, and small gains when time trials are pooled. Responses vary between people.
Proposed by our editor; not yet confirmed by an expert reviewer.
WADA status
Permitted
Distances in the research
800 m–5K: studied10K–half: limited researchMarathon: not studiedUltra: not studied

What is caffeine?

Caffeine (1,3,7-trimethylxanthine) is a stimulant found naturally in coffee, tea and cocoa. It is also added to soft drinks, energy drinks and a growing range of sports products, including gels, chews, gum, bars, mouth rinses and pre-workout powders[2].

Most research has used anhydrous (dry) caffeine in capsules, because the dose is easy to standardise and to match with a placebo[2]. Coffee is harder to pin down: the ISSN notes 8- to 9-fold differences in caffeine between coffees bought from similar shops, and similar spread between pre-workout supplements[2].

The NHS gives typical amounts: 100 mg in a cup of instant coffee (filter and coffee-shop coffee can contain more), 75 mg in a cup of tea, 40 mg in a can of cola, 80 mg in a 250 ml can of energy drink and less than 25 mg in a 50 g bar of plain dark chocolate[4].

The Australian Institute of Sport puts caffeine in Group A of its supplement framework, meaning strong scientific evidence for use in specific situations in sport[5].

Also called: coffee, caffeine tablets, caffeinated gels.

Why is caffeine getting attention?

Caffeine is one of the most widely used performance-enhancing substances in sport[1], and one of the few supplements the IOC consensus statement says has good evidence behind it[6]. The international athletics federation's 2019 consensus lists it among the evidence-based supplements for athletics[7]. Outside sport, about 90% of adults in Western countries consume caffeine regularly[2].

Doping-control samples give a rough measure of use in competition. Of 20,686 urine samples collected after competitions in 2004 to 2008, 73.8% contained caffeine, with the highest levels in triathlon, cycling and rowing, and only 0.6% above the old doping threshold of 12 µg per ml[8]. In 7,488 Spanish samples, detectable caffeine rose from about 70% to 75.7% by 2015, with the highest urine levels in cycling, athletics and rowing[9].

The claims, checked against the research

  • Faster race timesSupported

    Across 14 time-trial studies in runners, caffeine made runs about 0.7% faster, a small but significant effect; neither the 800 m to 5 km nor the longer trials were significant on their own[1].

  • Running longer before tiringSupported

    In 7 trials of running at a set pace to exhaustion, caffeine extended run time by about 17% overall, helping both recreational and trained runners[1]. Races, though, are run over a set distance[1].

  • Hard running feels easierSupported

    Across 21 studies, caffeine lowered perceived exertion during exercise by 5.6%, which explained about 29% of the variation in performance gains[10].

  • A lift late in a long raceMixed

    The IOC says 100 to 300 mg taken during endurance exercise may improve cycling time trials by 3 to 7%[6]. The running trials all gave caffeine before the run[1].

  • Sharper focus and alertnessMixed

    The ISSN says caffeine improves attention and vigilance in most people[2], and low doses of about 200 mg improved alertness and mood with few side effects[11]. But in a trial of 379 people, caffeine didn't raise alertness in those who rarely had it, and in regular users it mainly reversed the drop in alertness caused by going without[12].

  • Mouth rinse without swallowingMixed

    Rinsing with caffeine improved performance in 3 of 11 studies[13], and rinses during a 1-hour cycling time trial didn't help[14].

  • Burns more fat, saving glycogenNot supported

    The ISSN says the glycogen-sparing theory was challenged and eventually dismissed; the brain and nervous system are now accepted as the main target[2].

How does caffeine work?

The main effect is thought to be on the brain. Caffeine is similar in shape to adenosine, a chemical that dampens the release of neurotransmitters such as dopamine. By blocking adenosine's A1 and A2A receptors, caffeine raises alertness and may reduce the sense of effort and pain during exercise[2].

The older theory, that caffeine spares glycogen by increasing fat burning, did not hold up[2]. Caffeine may also act on muscle directly, by helping calcium release, but the strongest evidence points to the central nervous system[2].

It is absorbed quickly from the gut and peaks in the blood 30 to 120 minutes after it is taken[2]. Its half-life is usually 4 to 6 hours but ranges from about 1.5 to 10 hours in adults, and over 95% is broken down in the liver by the CYP1A2 enzyme[2].

Gum is partly absorbed through the lining of the mouth[15]. In 84 healthy men, caffeine from gum peaked in the blood sooner than from capsules, with comparable amounts absorbed at 100 and 200 mg[16].

Does caffeine improve running performance?

Yes, modestly: in a meta-analysis of 21 trials in runners, time to exhaustion improved by a medium amount and time trials got about 0.7% faster when all the trials were pooled[1]. Most endurance evidence comes from cycling in the laboratory[1]; across 46 endurance time-trial studies, completion times were about 2.2% faster[17], and an umbrella review rated the evidence mostly moderate in quality[18].

800 m to 5K (Studied)

In the running meta-analysis, 14 comparisons over 800 m to 5 km were not significantly faster with caffeine when analysed on their own[1]. Individual trials varied. Well-trained and recreational runners each ran a 5 km time trial about 1% faster after 5 mg per kg[19], and 18 recreational runners ran a treadmill 5 km in 1,413 seconds after an energy drink containing caffeine, sugar and taurine, against 1,444 seconds after a placebo[20].

Two 800 m trials, one using coffee and one using capsules, found no significant benefit. Of the other trials up to 5 km, those over a mile, 3 km and two over 5 km found benefits, while trials over 1,600 m and 4 km did not[1].

10K to half marathon (Limited research)

Eight comparisons over more than 5 km, from six studies (three at 10 km, two at 8 km and one at 21 km), were not significantly faster with caffeine when analysed on their own; the small overall benefit came from pooling all distances[1].

Both 8 km track trials found benefits, one of them in 8 trained men after 3 mg per kg, while none of the three 10 km trials did[1]. The 21 km trial was a road race in hot, humid conditions, and 7 trained runners' times after 0, 5 and 9 mg per kg were within 1.1% of each other[21].

Marathon (Not studied)

We found no controlled trials of caffeine in marathons[1]. The nearest evidence comes from other long events. In a field trial, 26 triathletes finished an Olympic-distance triathlon 1.3% faster after 6 mg per kg[22], and in a meta-analysis of endurance time trials not limited to running, the benefit grew with the length of the event[23][2].

Ultra (Not studied)

We found no controlled trials in ultra running. The authors of the running meta-analysis suggest the longer time to exhaustion seen with caffeine might apply to 24-hour, 100-mile and mountain races, but that has not been tested[1].

The ISSN notes that caffeine may help both thinking and physical performance in some people who are short of sleep, as in multiday events, while adding that caffeine can't make up for poor sleep[2].

Study by study

The main studies on this page. What they took is what the study gave, not advice for any individual.
StudyWho took partWhat they tookWhat was measuredWhat they found
Wang 2022 [1]Meta-analysis21 randomised crossover trials; 254 runners (220 men, 19 women, 15 not stated)3 to 9 mg/kg, or 200 to 300 mg in gum, strips or powder, mostly 60 minutes before, v placeboTime to exhaustion (7 trials) and time trials of 800 m to 21 km (14 trials)Time to exhaustion up about 17% (medium effect); time trials about 0.7% faster (small effect). Neither distance subgroup was significant on its own.
Southward 2018 [17]Meta-analysis46 randomised placebo-controlled studiesAcute caffeine v placebo; effect evident at 3 to 6 mg/kgEndurance time-trial time and mean powerTime about 2.2% better and power about 3.0% higher, small effects; a few studies found slower times.
Shen 2019 [23]Meta-analysis56 placebo-controlled comparisons from 40 articlesCaffeine v placeboEndurance time trials, by length of eventEffect size 0.33, growing by 0.15 for every extra 30 minutes of event.
O'Rourke 2008 [19]Randomised controlled trial30 runners: 15 well-trained and 15 recreational5 mg/kg of caffeine or a placebo, double-blind5 km time trial1.1% faster (well-trained) and 1.0% faster (recreational) than placebo.
Prins 2016 [20]Randomised controlled trial18 recreational endurance runners (13 men, 5 women)500 ml of a caffeinated energy drink with glucose and taurine, or a placebo, 60 minutes before5 km treadmill time trial1,413 v 1,444 seconds, faster with the energy drink; effort ratings did not differ.
Cohen 1996 [21]Randomised controlled trial7 trained road runners (5 men, 2 women)0, 5 or 9 mg/kg of caffeine, double-blind, in random orderThree 21 km outdoor road races in hot, humid conditionsNo significant difference: average race times were within 1.1% of each other.
Potgieter 2018 [22]Randomised controlled trial26 triathletes (14 men, 12 women); average habitual intake 413 mg a day6 mg/kg of microencapsulated caffeine or a placebo, 60 minutes beforeOlympic-distance triathlon, in the fieldTotal time 1.3% shorter and swim 3.7% shorter; effort ratings unchanged.
Guest 2018 [3]Randomised controlled trial101 competitive male athletesPlacebo, 2 mg/kg or 4 mg/kg of caffeine10 km cycling time trial, analysed by CYP1A2 genotype3% faster overall at 4 mg/kg. AA genotype: 4.8% faster at 2 mg/kg and 6.8% at 4 mg/kg; AC: no effect; CC: 13.7% slower at 4 mg/kg.
Carvalho 2022 [24]Meta-analysis60 studies reporting participants' habitual caffeine intakeA single dose of caffeine v placeboEndurance, power and strengthSmall benefit (SMD 0.25), unaffected by habitual intake; significant below 3 and at 3 to 6 mg/kg, not above 6 mg/kg.
Lara 2020 [25]Randomised controlled trial13 well-trained female triathletes with regular menstrual cycles3 mg/kg of caffeine or a placebo, 60 minutes before, in three phases of the cyclePeak power in an incremental cycling testPeak power up 2.7 to 3.6%, with no difference between phases.
Irwin 2011 [26]Randomised controlled trial12 well-trained male cyclists who habitually consumed caffeine4 days with or without caffeine, then 3 mg/kg or a placeboTime trial equal to 1 hour of cycling3.0% faster after the withdrawal period and 3.6% faster without it; the two caffeine trials did not differ.
Lara 2019 [27]Randomised controlled trial11 healthy active people3 mg/kg of caffeine or a placebo every day for 20 daysPeak cycling power, three times a weekPeak power about 4% higher over the first 15 days, then less; the effect was largest on day 1 and shrank progressively, suggesting tolerance.

The key studies, in order

  1. 1996Trial7 trained runners, three 21 km road races in heat and humidity: 5 or 9 mg/kg of caffeine did not change race times significantly. [21]
  2. 2002Trial84 healthy men: caffeine from gum peaked in the blood sooner than from capsules, with comparable amounts absorbed at 100 and 200 mg. [16]
  3. 2004Expert reviewWithdrawal symptoms such as headache and fatigue typically start 12-24 hours after stopping caffeine and last 2-9 days. [36]
  4. 2005Meta-analysis21 studies: caffeine cut perceived exertion during exercise by 5.6%, explaining about 29% of the variation in performance gains. [10]
  5. 2008Trial30 runners: 5 km time trials were about 1% faster after 5 mg/kg of caffeine than after a placebo, in both well-trained and recreational runners. [19]
  6. 2010Trial379 people after 16 hours without caffeine: the ADORA2A TT genotype was linked to more caffeine-induced anxiety. Caffeine didn't raise alertness in low consumers; in regular ones it returned alertness to baseline. [12]
  7. 2011Meta-analysis21 studies: caffeine added to carbohydrate gave a small extra endurance benefit, about half that of caffeine v placebo. [35]
  8. 2011Observational study20,686 post-competition urine samples, 2004-2008: 73.8% contained caffeine, with the highest levels in endurance sports. [8]
  9. 2011Trial12 trained male cyclists: 3 mg/kg improved a 1-hour time trial by about 3% with or without 4 days' prior caffeine withdrawal. [26]
  10. 2013Trial400 mg of caffeine at bedtime, 3 hours before or 6 hours before each disrupted sleep compared with placebo. [40]
  11. 2013Trial13 resistance-trained men: 3 mg/kg improved fast, light-load movements; 9 mg/kg helped heavy loads but raised side effects to 15-62%. [38]
  12. 2014Trial10 trained male cyclists: caffeine mouth rinses during a 1-hour time trial didn't improve performance or raise blood caffeine. [14]
  13. 2014Expert reviewLow doses (under 3 mg/kg, about 200 mg) can be ergogenic and improve alertness and mood, with few if any side effects. [11]
  14. 2016Expert review9 studies: coffee providing 3-8.1 mg/kg of caffeine improved endurance in 5, with moderate evidence for cycling and running. [33]
  15. 2016Trial38 recreational cyclists: 6 mg/kg improved 3 km performance only in CYP1A2 AC carriers; adding a mouth rinse helped both genotypes. [31]
  16. 2016Trial18 recreational runners: a caffeinated energy drink 60 minutes before made a treadmill 5 km faster than a placebo (1,413 v 1,444 seconds). [20]
  17. 2017Meta-analysis34 studies: caffeinated energy drinks improved endurance, strength and jumping; gains tracked taurine dose, not caffeine dose. [34]
  18. 2018Observational study20 elite rugby players: after an evening game with their usual caffeine, they went to bed about 3 hours later and slept 1.5 hours less. [41]
  19. 2018Trial101 athletes, 10 km cycling: caffeine helped those with the AA version of the CYP1A2 gene, did nothing for AC and slowed CC at the higher amount. [3]
  20. 2018ConsensusCaffeine is among the few performance supplements with good evidence. Contamination is a known anti-doping risk of supplement products. [6]
  21. 2018Trial26 triathletes (14 men, 12 women): 6 mg/kg before an Olympic-distance triathlon cut total time by 1.3% compared with placebo. [22]
  22. 2018Meta-analysis46 studies of endurance time trials: overall, caffeine improved completion time by about 2.2% and mean power by about 3.0%; a small effect at 3-6 mg/kg. [17]
  23. 2018Expert reviewGum is absorbed faster than capsules; mouth rinses mostly haven't helped aerobic exercise; energy-drink studies rarely isolate caffeine. [15]
  24. 2019Observational study7,488 samples: detectable caffeine rose from about 70% (2004-2008) to 75.7% (2015); highest urine levels in cycling, athletics and rowing. [9]
  25. 2019ConsensusEvidence-based supplements for athletics include caffeine, bicarbonate, beta-alanine, nitrate and creatine, with value specific to each event. [7]
  26. 2019Trial11 active people taking 3 mg/kg daily for 20 days: the gain in peak cycling power shrank progressively, suggesting tolerance. [27]
  27. 2019Systematic review10 studies comparing the sexes: similar aerobic benefit in men and women; 4 of 7 found larger anaerobic gains in men. [30]
  28. 2019Meta-analysis56 comparisons in endurance time trials: caffeine's effect grew with event duration, by 0.15 (Cohen's d) per extra 30 minutes. [23]
  29. 2020Systematic review11 crossover studies: caffeine mouth rinsing improved performance in 3; 8 found no or only suggestive benefits. [13]
  30. 2020Meta-analysisAcross 21 meta-analyses, caffeine improved aerobic endurance, strength and power, with moderate-quality evidence; most studies were in young men. [18]
  31. 2020Trial13 female triathletes: 3 mg/kg raised peak cycling power by 2.7-3.6%, similarly in three phases of the menstrual cycle. [25]
  32. 2021Systematic review17 studies of CYP1A2 genotype: differences in response were small, inconsistent or limited to specific tests. [32]
  33. 2021ConsensusMost consistent benefits for aerobic endurance; 3-6 mg/kg improved performance; very high amounts brought side effects; responses vary with genetics and habitual intake. [2]
  34. 2022Meta-analysis60 studies: habitual intake didn't change caffeine's acute benefit. Benefits were significant below 3 and at 3-6 mg/kg, not above 6 mg/kg. [24]
  35. 2022Systematic review25 studies, 421 athletes: individual side effects affected 6-34% at low doses and 8-83% at high doses; palpitations and poor sleep onset were most common. [37]
  36. 2022Meta-analysis21 randomised trials in runners (254 people, 220 men): caffeine improved time to exhaustion (medium effect) and time-trial times (small effect). [1]
  37. 2023Meta-analysis24 studies: caffeine cut total sleep by 45 minutes and sleep efficiency by 7%; coffee would need drinking 8.8 hours before bed to avoid lost sleep. [39]

Who has the research been done in?

The running meta-analysis pooled 254 runners: 220 men, 19 women and 15 whose sex wasn't reported. 167 were recreational and 87 trained, and average ages in the trials ranged from about 21 to 40[1]. Across caffeine research more broadly, most studies were in young men, and reviewers have called for more work in women and in middle-aged and older adults[18].

Training status doesn't seem to change the effect much. Time to exhaustion improved in both recreational and trained runners[1], and the ISSN concludes that trained and untrained people seem to gain similarly[2]. Regular coffee drinkers also seem to respond: in a meta-analysis of 60 studies, habitual intake did not change the acute benefit[24].

Adolescents and runners over 50 have hardly been studied, so the findings may not apply to them[1][18]. For children and adolescents, EFSA proposed a safe level of 3 mg per kg of body weight a day[28]. In England, the government plans to ban sales to under-16s of drinks, other than tea or coffee, with more than 150 mg of caffeine per litre, expected from April 2027[29].

What we know about women

Women made up 7.5% of the pooled sample in the running meta-analysis, so its authors say the benefit is supported in men and still needs study in women runners[1]. A systematic review of 10 studies comparing the sexes found a similar benefit for aerobic performance, though 4 of 7 found larger gains in men for anaerobic performance[30]. In 13 well-trained female triathletes with regular cycles, 3 mg per kg raised peak cycling power by a similar 2.7 to 3.6% in the early follicular, pre-ovulation and mid-luteal phases[25].

Hormonal contraceptives and pregnancy slow the enzyme that clears caffeine, so it stays in the body for longer[2]. In pregnancy, the NHS says no more than 200 mg of caffeine a day, because too much has been linked to a higher chance of miscarriage and stillbirth[4]. EFSA reached the same 200 mg a day figure for pregnancy[28].

What the research can't tell us yet

Most participants were men: 220 of 254 in the running meta-analysis[1]. The running trials were small, none was registered in advance, and the reviewers rated their overall risk of bias as unclear to low[1]. Across published meta-analyses, the quality of evidence was mostly rated moderate[18].

Placebos and mixtures varied. Some trials gave caffeine with sugar or sweeteners, or in coffee, which could mask or exaggerate its effect, and the meta-analysis couldn't test whether sex, form or dose changed the result[1]. Time-to-exhaustion tests, where the effect was largest, are an unusual format for runners, since races are run over a set distance[1].

Responses vary. In a trial of 101 athletes, caffeine helped those with one version of the CYP1A2 gene, did nothing for another and slowed a third[3]. Other studies disagree: in 38 recreational cyclists, swallowed caffeine improved a 3 km time trial only in AC carriers, the group that didn't benefit in the larger trial[31]. A systematic review of 17 studies found genotype differences small, inconsistent or limited to particular tests, and called for larger studies[32].

Belief matters too: the ISSN describes studies in which athletes' performance tracked what they thought they had taken[2].

Train smarter

Most gains come from consistent training. See how RunV builds your plan.

Start your free 7-day trial

Research alerts

Only when the rating, the anti-doping status or the key findings change. One click to stop.

How and when is caffeine used in studies?

Caffeine has consistently improved performance in studies at 3 to 6 mg per kg of body mass. The lowest effective amount may be around 2 mg per kg, and very high amounts, such as 9 mg per kg, brought more side effects without being needed[2]. The IOC describes two protocols: 3 to 6 mg per kg of anhydrous caffeine about 60 minutes before exercise, or lower doses (under 3 mg per kg, about 200 mg) before and during exercise, taken with carbohydrate[6].

The running trials used 3 to 9 mg per kg, or 200 to 300 mg in gum, mouth strips or powder[1]. A meta-analysis of 60 studies found significant benefits below 3 mg per kg and at 3 to 6 mg per kg, but not above 6 mg per kg[24].

Coffee providing 3 to 8.1 mg per kg improved endurance in 5 of 9 studies in one review[33], and the ISSN notes the gains at higher amounts were similar to those at 3 mg per kg[2]. Caffeinated energy drinks improved endurance tests in a meta-analysis of 34 studies, although the gains tracked the drinks' taurine content rather than their caffeine[34]. Adding caffeine to carbohydrate gave a small extra benefit, about half the size of caffeine compared with a placebo[35].

Gels have been studied far less. With 100 mg of caffeine, a gel taken 10 minutes before a 2,000 m rowing test improved it, while one taken 60 minutes before intermittent sprints did not[2]. Energy-drink studies rarely separate the effect of caffeine from other ingredients[15].

When studies gave it

Studies most often gave caffeine 60 minutes before exercise, when blood levels were thought to peak[2]. Gum may need less time: in 8 male cyclists, 300 mg in gum improved a time trial when chewed just before, but not 1 or 2 hours before[2].

Caffeine taken later in a long event also worked in cycling studies. The ISSN describes an early study in which a small amount of caffeine from flat cola towards the end of exercise matched the effect of about 5 to 6 mg per kg taken an hour before[2]. The IOC says 100 to 300 mg taken during endurance exercise, after 15 to 80 minutes, may improve cycling time trials by 3 to 7%[6].

Daily use may blunt the effect over weeks: when 11 active people took 3 mg per kg every day, the gain in peak cycling power was largest on day 1, then shrank progressively, lasting about 15 to 18 days[27]. Cutting caffeine out beforehand isn't clearly needed, as 3 mg per kg improved a cycling time trial by about 3% whether or not habitual users had gone 4 days without it[26]. Withdrawal symptoms such as headache and fatigue typically start 12 to 24 hours after stopping and last 2 to 9 days[36].

What studies used. These are research protocols, not advice for any individual.
What studies usedWhenRef.
3 to 6 mg per kg of body massMost often 60 minutes before exercise[2]
Under 3 mg per kg (about 200 mg), or 100 to 300 mg during exerciseBefore and during exercise, with carbohydrate; during, after 15 to 80 minutes[6]
5 mg per kg of body massBefore a 5 km time trial[19]
3 to 9 mg per kg, or 200 to 300 mg in gum, strips or powderMostly 60 minutes before running trials[1]
6 mg per kg of body mass60 minutes before an Olympic-distance triathlon[22]
3 mg per kg a day for 20 days45 minutes before each test; the benefit shrank over time and lasted 15 to 18 days[27]

What are the side effects, and who should be careful?

Side effects rise with the dose. A systematic review of 25 studies in 421 athletes found that individual side effects affected 6 to 34% of people at low doses (3 mg per kg or less) and 8 to 83% at high doses (above 6 mg per kg). A fast heartbeat or palpitations and trouble getting to sleep were the most common and the most marked[37]. In 13 resistance-trained men, 9 mg per kg sharply increased the frequency of side effects, to between 15 and 62%[38].

The IOC lists nausea, anxiety, a faster heart rate, insomnia and restlessness, which become more likely at 9 mg per kg or more[6]. The ISSN adds headaches[2]. People with the TT version of the ADORA2A gene were more prone to anxiety after caffeine, and regular consumers had a smaller anxiety response[12].

Across 24 studies, caffeine cut total sleep by 45 minutes and sleep efficiency by 7%, and reduced deep sleep[39]. 400 mg taken 6 hours before bed still disrupted sleep in a home study[40]. After an evening game, elite rugby players using caffeine as usual went to bed about 3 hours later and slept 1.5 hours less, and bigger rises in saliva caffeine were linked to slower sleep onset[41].

For healthy adults who aren't pregnant, EFSA concluded that single doses up to 200 mg (about 3 mg per kg) raise no safety concerns, even within 2 hours before intense exercise in normal conditions, and that up to 400 mg a day raises none. It noted that 100 mg close to bedtime may affect sleep in some adults[28]. Caffeine is a mild diuretic, but the IOC says the effect is small at the doses that improve performance[6], and the ISSN says it doesn't significantly impair hydration or temperature control when exercising in the heat[2].

Medicines and health conditions

The NHS lists caffeine among possible triggers of heart palpitations, alongside strenuous exercise, stress, anxiety and some medicines, and says to call 999 or go to A&E if palpitations don't go away or come with chest pain, shortness of breath or fainting[42].

The ISSN notes that pregnancy and hormonal contraceptives slow the breakdown of caffeine, while smoking and heavy caffeine use speed it up[2]. Caffeine is an ingredient in some over-the-counter and prescription medicines, often combined with painkillers, so it can add to the caffeine from drinks and supplements[2].

The ISSN suggests caution with very high caffeine intakes (above 9 mg per kg) during creatine loading, as the two may have opposing effects on muscle relaxation[2].

Be careful if

  • You're pregnant: the NHS says no more than 200 mg of caffeine a day[4].
  • You train or race late in the day: across 24 studies, caffeine cut total sleep by 45 minutes on average[39].
  • Caffeine makes you anxious or jittery: sensitivity varies with genes and with how much you usually have[12].
  • You're thinking of taking very large amounts: 9 mg per kg raised side effects to as many as 62% of people in one trial[38].
  • You take hormonal contraceptives: they slow the breakdown of caffeine, so it lasts longer[2].
  • You get heart palpitations: the NHS lists caffeine among possible triggers[42].

Is caffeine allowed in competition?

Caffeine is not on the World Anti-Doping Agency's Prohibited List[43]. It was removed from WADA's controlled substances in 2004[2] and is in WADA's Monitoring Program, which tracks use without prohibiting it[43]. The Monitoring Program follows urine caffeine levels after competition[9].

The ISSN notes that athletes are encouraged to keep urine caffeine below 12 µg per ml, the old doping limit. That corresponds to about 10 mg per kg taken over several hours, more than three times the amounts shown to improve performance[2].

The anti-doping risk with any supplement product is contamination: products can contain prohibited substances that aren't on the label, and athletes are responsible for anything found in their sample[6].

Independent quality-assurance schemes, often called batch testing, check supplement products for prohibited substances[6]. WADA says using a product from one of these schemes can reduce, but not eliminate, the risk of an anti-doping violation[43], and the IOC says no product can be guaranteed entirely safe[6].

Myth or evidence?

5 things runners hear about caffeine. Pick an answer to see what the research says.

  1. 1.Caffeine improves endurance running performance.

    Show the answer

    Evidence. A meta-analysis of 21 trials in runners found better time to exhaustion and slightly faster time trials[1].

  2. 2.More caffeine always means a bigger improvement.

    Show the answer

    Myth. Very high amounts, such as 9 mg per kg, brought more side effects and don't seem to be needed[2].

  3. 3.Everyone responds to caffeine in the same way.

    Show the answer

    Myth. In one trial, the effect depended on a gene that affects caffeine metabolism[3].

  4. 4.Caffeine is banned in athletics.

    Show the answer

    Myth. It is not on the WADA Prohibited List; it is monitored[43][2].

  5. 5.Caffeine can disrupt your sleep.

    Show the answer

    Evidence. Disrupted sleep is one of the side effects the ISSN describes[2].

Have you tried caffeine?

An anonymous reader poll. We store only the page, your answer and the time.

Have you tried caffeine?

What readers have said so far

No answers yet. We show the results once 30 readers have answered.

A poll of readers who chose to answer, not a scientific survey.

Frequently asked questions

Does caffeine improve running performance?
Research suggests it does, modestly: in a meta-analysis of 21 trials in runners, time to exhaustion improved by about 17% and time trials got about 0.7% faster[1].
How much caffeine did studies use for running?
Most used 3 to 6 mg per kg of body mass[2]; the running trials ranged from 3 to 9 mg per kg, or 200 to 300 mg in gum, strips or powder[1].
How long before a run do studies give caffeine?
Most often 60 minutes before exercise; caffeinated gum may act faster[2].
Is coffee as good as caffeine tablets before a run?
Coffee can work but is harder to dose: coffee providing 3 to 8.1 mg per kg improved endurance in 5 of 9 studies[33], and caffeine content varies widely between coffees[2].
Do caffeine gels or gum work faster?
Gum does: caffeine from gum reached peak blood levels sooner than from capsules[16]. Gels have been studied much less, with mixed results[2].
Do I need to cut out coffee before a race for caffeine to work?
Studies suggest not: habitual intake didn't change the benefit in a meta-analysis of 60 studies[24], and 3 mg per kg helped cyclists with or without 4 days off caffeine[26].
Why does caffeine work for some runners and not others?
Genes may play a part: in one trial, caffeine helped athletes with one version of the CYP1A2 gene and not others[3], though a review found such differences small and inconsistent[32].
Can caffeine affect sleep?
Yes. Across 24 studies it cut total sleep by 45 minutes on average[39].
What are the side effects of caffeine for runners?
A fast heartbeat or palpitations and trouble falling asleep were the most common, and side effects became more frequent at higher doses[37].
How much caffeine is safe in a day?
For healthy adults who aren't pregnant, EFSA concluded that up to 400 mg a day, and single doses up to 200 mg, raise no safety concerns[28]. In pregnancy, the NHS says no more than 200 mg a day[4].
Is caffeine safe for teenage runners?
EFSA proposed a safe level of 3 mg per kg a day for children and adolescents[28]. The running trials were in adults, so may not apply to teenagers[1].
Does caffeine help in a marathon or ultra?
Research hasn't answered this yet: the running meta-analysis, which searched the literature to October 2022, included no marathon or ultra trials[1]. In other endurance events, the benefit grew with the length of the event[23].
Does a caffeine mouth rinse work?
The evidence is mixed: 3 of 11 studies found a benefit[13], and rinsing during a 1-hour cycling time trial didn't help or raise blood caffeine[14].
Does caffeine work for women runners?
Research suggests so, though most studies were in men: 220 of 254 people in the running meta-analysis[1]. Aerobic benefits were similar in both sexes[30] and across the menstrual cycle[25].
Is caffeine banned in athletics?
No. It is not on the WADA Prohibited List and has been monitored rather than controlled since 2004[43][2].

References

  1. [1]

    Wang Z, Qiu B, Gao J, Del Coso J Effects of caffeine intake on endurance running performance and time to exhaustion: a systematic review and meta-analysis. Nutrients, 2022.

    Meta-analysis21 randomised trials in runners (254 people, 220 men): caffeine improved time to exhaustion (medium effect) and time-trial times (small effect).

    doi.org/10.3390/nu15010148
  2. [2]

    Guest NS, VanDusseldorp TA, Nelson MT, et al. International society of sports nutrition position stand: caffeine and exercise performance. Journal of the International Society of Sports Nutrition, 2021.

    Consensus statementMost consistent benefits for aerobic endurance; 3-6 mg/kg improved performance; very high amounts brought side effects; responses vary with genetics and habitual intake.

    doi.org/10.1186/s12970-020-00383-4
  3. [3]

    Guest N, Corey P, Vescovi J, El-Sohemy A Caffeine, CYP1A2 genotype, and endurance performance in athletes. Medicine and Science in Sports and Exercise, 2018.

    Randomised controlled trial101 athletes, 10 km cycling: caffeine helped those with the AA version of the CYP1A2 gene, did nothing for AC and slowed CC at the higher amount.

    doi.org/10.1249/MSS.0000000000001596
  4. [4]

    NHS Foods to avoid in pregnancy. NHS, 2026.

    RegulatorNo more than 200 mg of caffeine a day in pregnancy; too much has been linked to miscarriage and stillbirth. Lists amounts such as 100 mg in a cup of instant coffee.

    www.nhs.uk/pregnancy/keeping-well/foods-to-avoid/
  5. [5]

    Australian Institute of Sport AIS Sports Supplement Framework: Group A and linked fact sheets. Australian Sports Commission, 2026.

    Consensus statementLists caffeine in Group A: strong scientific evidence for use in specific situations in sport, using evidence-based protocols.

    www.ausport.gov.au/ais/nutrition/supplements/group_a
  6. [6]

    Maughan RJ, Burke LM, Dvorak J, et al. IOC consensus statement: dietary supplements and the high-performance athlete. British Journal of Sports Medicine, 2018.

    Consensus statementCaffeine is among the few performance supplements with good evidence. Contamination is a known anti-doping risk of supplement products.

    doi.org/10.1136/bjsports-2018-099027
  7. [7]

    Burke LM, Castell LM, Casa DJ, et al. International Association of Athletics Federations consensus statement 2019: nutrition for athletics. International Journal of Sport Nutrition and Exercise Metabolism, 2019.

    Consensus statementEvidence-based supplements for athletics include caffeine, bicarbonate, beta-alanine, nitrate and creatine, with value specific to each event.

    doi.org/10.1123/ijsnem.2019-0065
  8. [8]

    Del Coso J, Muñoz G, Muñoz-Guerra J Prevalence of caffeine use in elite athletes following its removal from the World Anti-Doping Agency list of banned substances. Applied Physiology, Nutrition, and Metabolism, 2011.

    Observational study20,686 post-competition urine samples, 2004-2008: 73.8% contained caffeine, with the highest levels in endurance sports.

    doi.org/10.1139/h11-052
  9. [9]

    Aguilar-Navarro M, Muñoz G, Salinero JJ, et al. Urine caffeine concentration in doping control samples from 2004 to 2015. Nutrients, 2019.

    Observational study7,488 samples: detectable caffeine rose from about 70% (2004-2008) to 75.7% (2015); highest urine levels in cycling, athletics and rowing.

    doi.org/10.3390/nu11020286
  10. [10]

    Doherty M, Smith PM Effects of caffeine ingestion on rating of perceived exertion during and after exercise: a meta-analysis. Scandinavian Journal of Medicine and Science in Sports, 2005.

    Meta-analysis21 studies: caffeine cut perceived exertion during exercise by 5.6%, explaining about 29% of the variation in performance gains.

    doi.org/10.1111/j.1600-0838.2005.00445.x
  11. [11]

    Spriet LL Exercise and sport performance with low doses of caffeine. Sports Medicine, 2014.

    Expert reviewLow doses (under 3 mg/kg, about 200 mg) can be ergogenic and improve alertness and mood, with few if any side effects.

    doi.org/10.1007/s40279-014-0257-8
  12. [12]

    Rogers PJ, Hohoff C, Heatherley SV, et al. Association of the anxiogenic and alerting effects of caffeine with ADORA2A and ADORA1 polymorphisms and habitual level of caffeine consumption. Neuropsychopharmacology, 2010.

    Randomised controlled trial379 people after 16 hours without caffeine: the ADORA2A TT genotype was linked to more caffeine-induced anxiety. Caffeine didn't raise alertness in low consumers; in regular ones it returned alertness to baseline.

    doi.org/10.1038/npp.2010.71
  13. [13]

    Ehlert AM, Twiddy HM, Wilson PB The effects of caffeine mouth rinsing on exercise performance: a systematic review. International Journal of Sport Nutrition and Exercise Metabolism, 2020.

    Systematic review11 crossover studies: caffeine mouth rinsing improved performance in 3; 8 found no or only suggestive benefits.

    doi.org/10.1123/ijsnem.2020-0083
  14. [14]

    Doering TM, Fell JW, Leveritt MD, et al. The effect of a caffeinated mouth-rinse on endurance cycling time-trial performance. International Journal of Sport Nutrition and Exercise Metabolism, 2014.

    Randomised controlled trial10 trained male cyclists: caffeine mouth rinses during a 1-hour time trial didn't improve performance or raise blood caffeine.

    doi.org/10.1123/ijsnem.2013-0103
  15. [15]

    Wickham KA, Spriet LL Administration of caffeine in alternate forms. Sports Medicine, 2018.

    Expert reviewGum is absorbed faster than capsules; mouth rinses mostly haven't helped aerobic exercise; energy-drink studies rarely isolate caffeine.

    doi.org/10.1007/s40279-017-0848-2
  16. [16]

    Kamimori GH, Karyekar CS, Otterstetter R, et al. The rate of absorption and relative bioavailability of caffeine administered in chewing gum versus capsules to normal healthy volunteers. International Journal of Pharmaceutics, 2002.

    Randomised controlled trial84 healthy men: caffeine from gum peaked in the blood sooner than from capsules, with comparable amounts absorbed at 100 and 200 mg.

    doi.org/10.1016/s0378-5173(01)00958-9
  17. [17]

    Southward K, Rutherfurd-Markwick KJ, Ali A The effect of acute caffeine ingestion on endurance performance: a systematic review and meta-analysis. Sports Medicine, 2018.

    Meta-analysis46 studies of endurance time trials: overall, caffeine improved completion time by about 2.2% and mean power by about 3.0%; a small effect at 3-6 mg/kg.

    doi.org/10.1007/s40279-018-0939-8
  18. [18]

    Grgic J, Grgic I, Pickering C, et al. Wake up and smell the coffee: caffeine supplementation and exercise performance, an umbrella review of 21 published meta-analyses. British Journal of Sports Medicine, 2020.

    Meta-analysisAcross 21 meta-analyses, caffeine improved aerobic endurance, strength and power, with moderate-quality evidence; most studies were in young men.

    doi.org/10.1136/bjsports-2018-100278
  19. [19]

    O'Rourke MP, O'Brien BJ, Knez WL, Paton CD Caffeine has a small effect on 5-km running performance of well-trained and recreational runners. Journal of Science and Medicine in Sport, 2008.

    Randomised controlled trial30 runners: 5 km time trials were about 1% faster after 5 mg/kg of caffeine than after a placebo, in both well-trained and recreational runners.

    doi.org/10.1016/j.jsams.2006.12.118
  20. [20]

    Prins PJ, Goss FL, Nagle EF, et al. Energy drinks improve five-kilometer running performance in recreational endurance runners. Journal of Strength and Conditioning Research, 2016.

    Randomised controlled trial18 recreational runners: a caffeinated energy drink 60 minutes before made a treadmill 5 km faster than a placebo (1,413 v 1,444 seconds).

    doi.org/10.1519/JSC.0000000000001391
  21. [21]

    Cohen BS, Nelson AG, Prevost MC, et al. Effects of caffeine ingestion on endurance racing in heat and humidity. European Journal of Applied Physiology and Occupational Physiology, 1996.

    Randomised controlled trial7 trained runners, three 21 km road races in heat and humidity: 5 or 9 mg/kg of caffeine did not change race times significantly.

    doi.org/10.1007/BF02425499
  22. [22]

    Potgieter S, Wright HH, Smith C Caffeine improves triathlon performance: a field study in males and females. International Journal of Sport Nutrition and Exercise Metabolism, 2018.

    Randomised controlled trial26 triathletes (14 men, 12 women): 6 mg/kg before an Olympic-distance triathlon cut total time by 1.3% compared with placebo.

    doi.org/10.1123/ijsnem.2017-0165
  23. [23]

    Shen JG, Brooks MB, Cincotta J, Manjourides JD Establishing a relationship between the effect of caffeine and duration of endurance athletic time trial events: a systematic review and meta-analysis. Journal of Science and Medicine in Sport, 2019.

    Meta-analysis56 comparisons in endurance time trials: caffeine's effect grew with event duration, by 0.15 (Cohen's d) per extra 30 minutes.

    doi.org/10.1016/j.jsams.2018.07.022
  24. [24]

    Carvalho A, Marticorena FM, Grecco BH, et al. Can I have my coffee and drink it? A systematic review and meta-analysis to determine whether habitual caffeine consumption affects the ergogenic effect of caffeine. Sports Medicine, 2022.

    Meta-analysis60 studies: habitual intake didn't change caffeine's acute benefit. Benefits were significant below 3 and at 3-6 mg/kg, not above 6 mg/kg.

    doi.org/10.1007/s40279-022-01685-0
  25. [25]

    Lara B, Gutiérrez-Hellín J, García-Bataller A, et al. Ergogenic effects of caffeine on peak aerobic cycling power during the menstrual cycle. European Journal of Nutrition, 2020.

    Randomised controlled trial13 female triathletes: 3 mg/kg raised peak cycling power by 2.7-3.6%, similarly in three phases of the menstrual cycle.

    doi.org/10.1007/s00394-019-02100-7
  26. [26]

    Irwin C, Desbrow B, Ellis A, et al. Caffeine withdrawal and high-intensity endurance cycling performance. Journal of Sports Sciences, 2011.

    Randomised controlled trial12 trained male cyclists: 3 mg/kg improved a 1-hour time trial by about 3% with or without 4 days' prior caffeine withdrawal.

    doi.org/10.1080/02640414.2010.541480
  27. [27]

    Lara B, Ruiz-Moreno C, Salinero JJ, Del Coso J Time course of tolerance to the performance benefits of caffeine. PLoS One, 2019.

    Randomised controlled trial11 active people taking 3 mg/kg daily for 20 days: the gain in peak cycling power shrank progressively, suggesting tolerance.

    doi.org/10.1371/journal.pone.0210275
  28. [28]

    EFSA Panel on Dietetic Products, Nutrition and Allergies Scientific opinion on the safety of caffeine. EFSA Journal, 2015.

    RegulatorHealthy adults: single doses up to 200 mg, even before intense exercise, and 400 mg a day raise no safety concerns; 200 mg a day in pregnancy; 3 mg/kg a day for children and adolescents.

    doi.org/10.2903/j.efsa.2015.4102
  29. [29]

    UK Government Banning the sale of high-caffeine energy drinks to children: consultation outcome. GOV.UK, 2026.

    RegulatorSales to under-16s in England of drinks other than tea or coffee with over 150 mg of caffeine per litre to be banned, expected from April 2027.

    www.gov.uk/government/consultations/banning-the-sale-of-high-caffeine-energy-drinks-to-children/outcome/banning-the-sale-of-high-caffeine-energy-drinks-to-children-consultation-outcome
  30. [30]

    Mielgo-Ayuso J, Marques-Jiménez D, Refoyo I, et al. Effect of caffeine supplementation on sports performance based on differences between sexes: a systematic review. Nutrients, 2019.

    Systematic review10 studies comparing the sexes: similar aerobic benefit in men and women; 4 of 7 found larger anaerobic gains in men.

    doi.org/10.3390/nu11102313
  31. [31]

    Pataky MW, Womack CJ, Saunders MJ, et al. Caffeine and 3-km cycling performance: effects of mouth rinsing, genotype, and time of day. Scandinavian Journal of Medicine and Science in Sports, 2016.

    Randomised controlled trial38 recreational cyclists: 6 mg/kg improved 3 km performance only in CYP1A2 AC carriers; adding a mouth rinse helped both genotypes.

    doi.org/10.1111/sms.12501
  32. [32]

    Grgic J, Pickering C, Del Coso J, et al. CYP1A2 genotype and acute ergogenic effects of caffeine intake on exercise performance: a systematic review. European Journal of Nutrition, 2021.

    Systematic review17 studies of CYP1A2 genotype: differences in response were small, inconsistent or limited to specific tests.

    doi.org/10.1007/s00394-020-02427-6
  33. [33]

    Higgins S, Straight CR, Lewis RD The effects of preexercise caffeinated coffee ingestion on endurance performance: an evidence-based review. International Journal of Sport Nutrition and Exercise Metabolism, 2016.

    Expert review9 studies: coffee providing 3-8.1 mg/kg of caffeine improved endurance in 5, with moderate evidence for cycling and running.

    doi.org/10.1123/ijsnem.2015-0147
  34. [34]

    Souza DB, Del Coso J, Casonatto J, Polito MD Acute effects of caffeine-containing energy drinks on physical performance: a systematic review and meta-analysis. European Journal of Nutrition, 2017.

    Meta-analysis34 studies: caffeinated energy drinks improved endurance, strength and jumping; gains tracked taurine dose, not caffeine dose.

    doi.org/10.1007/s00394-016-1331-9
  35. [35]

    Conger SA, Warren GL, Hardy MA, Millard-Stafford ML Does caffeine added to carbohydrate provide additional ergogenic benefit for endurance?. International Journal of Sport Nutrition and Exercise Metabolism, 2011.

    Meta-analysis21 studies: caffeine added to carbohydrate gave a small extra endurance benefit, about half that of caffeine v placebo.

    doi.org/10.1123/ijsnem.21.1.71
  36. [36]

    Juliano LM, Griffiths RR A critical review of caffeine withdrawal: empirical validation of symptoms and signs, incidence, severity, and associated features. Psychopharmacology, 2004.

    Expert reviewWithdrawal symptoms such as headache and fatigue typically start 12-24 hours after stopping caffeine and last 2-9 days.

    doi.org/10.1007/s00213-004-2000-x
  37. [37]

    de Souza JG, Del Coso J, Fonseca FS, et al. Risk or benefit? Side effects of caffeine supplementation in sport: a systematic review. European Journal of Nutrition, 2022.

    Systematic review25 studies, 421 athletes: individual side effects affected 6-34% at low doses and 8-83% at high doses; palpitations and poor sleep onset were most common.

    doi.org/10.1007/s00394-022-02874-3
  38. [38]

    Pallarés JG, Fernández-Elías VE, Ortega JF, et al. Neuromuscular responses to incremental caffeine doses: performance and side effects. Medicine and Science in Sports and Exercise, 2013.

    Randomised controlled trial13 resistance-trained men: 3 mg/kg improved fast, light-load movements; 9 mg/kg helped heavy loads but raised side effects to 15-62%.

    doi.org/10.1249/MSS.0b013e31829a6672
  39. [39]

    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.

    Meta-analysis24 studies: caffeine cut total sleep by 45 minutes and sleep efficiency by 7%; coffee would need drinking 8.8 hours before bed to avoid lost sleep.

    doi.org/10.1016/j.smrv.2023.101764
  40. [40]

    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.

    Randomised controlled trial400 mg of caffeine at bedtime, 3 hours before or 6 hours before each disrupted sleep compared with placebo.

    doi.org/10.5664/jcsm.3170
  41. [41]

    Dunican IC, Higgins CC, Jones MJ, et al. Caffeine use in a Super Rugby game and its relationship to post-game sleep. European Journal of Sport Science, 2018.

    Observational study20 elite rugby players: after an evening game with their usual caffeine, they went to bed about 3 hours later and slept 1.5 hours less.

    doi.org/10.1080/17461391.2018.1433238
  42. [42]

    NHS Heart palpitations. NHS, 2026.

    RegulatorLists caffeine among possible triggers of palpitations; call 999 or go to A&E if they persist or come with chest pain, breathlessness or fainting.

    www.nhs.uk/symptoms/heart-palpitations/
  43. [43]

    World Anti-Doping Agency The Prohibited List. World Anti-Doping Agency, 2026.

    RegulatorThe substances and methods prohibited in sport. Caffeine is not on it.

    www.wada-ama.org/en/prohibited-list

Who wrote and reviewed this

Written by Grant MacMillan, Editor, RunV. Not yet reviewed by an expert.

Last checked against its references . Next review due . How pages are reviewed · How we grade research

Final split / Start now

Your first week is on us.

Download RunV, answer a few questions about your goal, and your adaptive plan is ready before your shoes are laced. Premium coaching free for your first week. Free to download, cancel anytime.

App StoreGoogle Play

7-day free trial · Monthly / Quarterly / Annual · cancel anytime