Compound library

Understand what people are taking.

Browse medicines, supplements, herbs, and health foods. Each profile brings community reports and linked research into one clear view.

Caffeine UK: Evidence & User Reports

supplement

Caffeine

Also known as Caffeine anhydrous, 1,3,7-Trimethylxanthine

Stimulant that blocks the brain's adenosine ('tiredness') receptors, found in coffee, tea, energy drinks and pre-workout products. Meta-analyses of placebo-controlled trials show small but consistent gains in endurance (time trials about 2-3% faster at 3-6 mg per kg of body weight), in strength and power, and in attention and reaction time, with the biggest effects after sleep loss. The costs are just as well documented: it shortens and lightens sleep, raises blood pressure, and can trigger anxiety or panic in people prone to it. EFSA considers up to 400 mg a day from all sources, and no more than 200 mg at once, safe for healthy adults.

Before you take caffeine

Sleep: caffeine shortens and lightens sleep. A typical pre-workout serving needed about 13 hours before bed to avoid lost sleep.
Blood pressure: regular caffeine raises it. Ask a GP or pharmacist first if you have high blood pressure or a heart condition.
Anxiety: it can raise anxiety, and high doses triggered panic attacks in about half of people with panic disorder.
Stacking: count every source together: coffee, tea, energy drinks, pre-workouts and 'fat burners'. Caffeine is often stacked with creatine or L-theanine, but the trials on this page tested caffeine alone.
EFSA's safe-use guide for healthy adults: up to 400 mg a day from all sources and no more than 200 mg at once; up to 200 mg a day in pregnancy. Not for children.

General information, not medical advice.

3linked conditions

0linked studies

On this pageWhat studies found Evidence Availability Outcomes & conditions →

Evidence

What the research actually shows

We link to relevant studies and leave interpretation to you and your healthcare professional.

What studies found

The headline results, condition by condition

The large figures are what the studies reported for each thing they measured. Open a card for every result.

Cognitive Function

19 of 20 results reported a benefit

5 studies, including a review of trials

Reaction timeModerate effectmoderately better than the comparison groupAverage of 2 results · 2 reviews of trials5 of 5 results reported a benefit2025 study 1 ↗2020 study 2 ↗
Attention accuracySlight effectslightly better than the comparison group1 review of trials2 of 2 results reported a benefit2025 study ↗

Outcomes & conditions

Why people look into Caffeine — and what they report

Associated conditions and community-reported outcomes, wrapped together on one page. Aggregates appear only once enough reports arrive.

Outcomes & conditions →

Important: This page reflects observational, aggregate-only community reporting and independent published research. It is not medical advice, and it does not claim any compound treats, cures, or is effective for any condition. Always consult a qualified healthcare professional before starting, stopping, or changing any supplement or treatment.

Availability

Where to get Caffeine (UK)

How this compound is sold in the UK, and where our readers typically find it.

Retailer search links — we may earn commission on purchases at no cost to you. Commission never affects our evidence ratings.

Search Caffeine on Amazon UK →Search Caffeine on Holland & Barrett →Search Caffeine on iHerb →

Suggested uses (unproven)

Where people report using it and what trials suggest

Strong evidenceEndurance performance (3-6 mg per kg about an hour before)
Strong evidenceAttention and reaction time, especially after sleep loss
Moderate evidenceMuscle strength and power (small effect)

Where the evidence is negative

Common claims the trials do not support

Low doses (1-3 mg per kg) did not significantly improve cycling time trials in a 2024 meta-analysis of 15 studies.
Improved marksmen's reaction time but not their shooting accuracy (2019 review of 17 studies).
Lower-body maximal strength gains were not significant in a 2018 meta-analysis; the upper-body effect was small.
Responses vary: in a 46-study review, two studies found slower time trials with caffeine.

The world's most-used stimulant and one of the few sports supplements with consistent meta-analysis support; its costs to sleep, blood pressure and anxiety are as well documented as its benefits.

Safety notes: Interaction cautions. Sleep: in a 2023 meta-analysis of 24 studies, caffeine cut total sleep by about 45 minutes and reduced deep sleep; a typical pre-workout serving (about 217 mg) needed roughly 13 hours before bed to avoid lost sleep. Blood pressure: regular caffeine raises it (about 2 mmHg systolic across trials, about 4 mmHg in trials of pure caffeine rather than coffee); get medical advice first if you have high blood pressure or a heart condition. Anxiety: caffeine raises anxiety in healthy people, and challenge doses of 400-750 mg triggered panic attacks in about half of people with panic disorder. Stacking: count every source, including coffee, tea, energy drinks, pre-workouts, 'fat burners' and some pain or cold remedies, and check pre-workouts for other stimulants. Pairing with L-theanine or creatine is common, but those combinations are separate studies, not this page's evidence. EFSA's safe-use guide for healthy adults: up to 400 mg a day from all sources and no more than 200 mg at once (about 3 mg per kg); up to 200 mg a day in pregnancy. Not for children. General information, not medical advice.

Evidence grades reflect study quality and consistency, not marketing claims.

No hand-picked studies have been added to this profile yet.

Search PubMed ↗Find clinical trials ↗

What trials found, in numbers

Summaries of published trial results for this compound, grouped by condition. In these tables every number is exactly as the researchers published it. The headline figures above show a simple average where several studies measured the same thing in the same unit, and say so; nothing else is calculated or converted.

How to read this table

Benefit: What the trial saw happen: whether the measured outcome moved in the helpful direction in that one study (the tag on each row: “Benefit seen”, “No clear benefit”, “Possible harm” or “Unclear”). It is a study result — not advice, and not a promise it will work for you.
Evidence: How strong the research behind that result is. A meta-analysis pools many studies (strongest); a systematic review collates them; an RCT tests against a control (placebo); observational studies only watch — they can suggest, not prove.
Effect size: How large the measured change was, exactly as the researchers published it. “No number reported” means the abstract gave no effect size — we never invent one.

In short: benefit is what was observed; evidence is how much you should trust the observation. A “Benefit seen” tag on a weak study counts for far less than the same tag on a meta-analysis.

Cognitive Function (20 outcomes)

Outcome measured	What trials found	Effect size	People studied	Duration	Type of evidence	Study
Attention, accuracy and speed during exercise (pooled, 5 studies)	Caffeine (drinks, capsules, bars or gum) improved Attention, accuracy and speed during exercise (pooled, 5 studies) compared with placebo.Benefit seen	no number reported	—	—	Meta-analysis · 2021	PMID 33800853
Self-reported energy, mood and cognitive function during exercise	Low to moderate dose caffeine improved Self-reported energy, mood and cognitive function during exercise compared with placebo.Benefit seen	no number reported	—	—	Meta-analysis · 2021	PMID 33800853
Accuracy on attention tests	Caffeine improved Accuracy on attention tests (g = 0.68; 95 % CI: 0.48-0.88) with no comparator reported. (95 % CI: 0.48-0.88)Benefit seen	g = 0.68	—	—	Meta-analysis · 2020	PMID 31837359
Accuracy on information processing tasks	Caffeine improved Accuracy on information processing tasks (g = 0.43; 95 % CI: 0.30-0.55) with no comparator reported. (95 % CI: 0.30-0.55)Benefit seen	g = 0.43	—	—	Meta-analysis · 2020	PMID 31837359
Crystallized intelligence	Caffeine improved Crystallized intelligence with no comparator reported.Benefit seen	no number reported	—	—	Meta-analysis · 2020	PMID 31837359
Executive function	Caffeine improved Executive function (g = 0.35; 95 % CI: 0.15-0.55) with no comparator reported. (95 % CI: 0.15-0.55)Benefit seen	g = 0.35	—	—	Meta-analysis · 2020	PMID 31837359
Lateral vehicular control on driving tests	Caffeine improved Lateral vehicular control on driving tests (g = 1.67; 95 % CI: 1.32-2.02) with no comparator reported. (95 % CI: 1.32-2.02)Benefit seen	g = 1.67	—	—	Meta-analysis · 2020	PMID 31837359
Longitudinal vehicular control on driving tests	Caffeine improved Longitudinal vehicular control on driving tests (g = 1.60; 95 % CI: 1.16-2.03) with no comparator reported. (95 % CI: 1.16-2.03)Benefit seen	g = 1.60	—	—	Meta-analysis · 2020	PMID 31837359
Memory	Caffeine improved Memory with no comparator reported.Benefit seen	no number reported	—	—	Meta-analysis · 2020	PMID 31837359
Occupational performance	Caffeine improved Occupational performance with no comparator reported.Benefit seen	no number reported	—	—	Meta-analysis · 2020	PMID 31837359
Reaction time	Caffeine improved Reaction time (g = 1.11; 95 % CI: 0.75-1.47) with no comparator reported. (95 % CI: 0.75-1.47)Benefit seen	g = 1.11	—	—	Meta-analysis · 2020	PMID 31837359
Response time on attention tests	Caffeine improved Response time on attention tests with no comparator reported.Benefit seen	no number reported	—	—	Meta-analysis · 2020	PMID 31837359
Response time on information processing tasks	Caffeine improved Response time on information processing tasks (g = 1.95; 95 % CI: 1.39-2.52) with no comparator reported. (95 % CI: 1.39-2.52)Benefit seen	g = 1.95	—	—	Meta-analysis · 2020	PMID 31837359
accuracy	Caffeine improved accuracy (0.27 SMD) compared with placebo in 1455.Benefit seen	0.27 SMD	1455	—	Systematic review · 2025	PMID 40335666
reaction time	Caffeine improved reaction time (0.28 SMD) compared with placebo in 1455.Benefit seen	0.28 SMD	1455	—	Systematic review · 2025	PMID 40335666
Marksman reaction time	Caffeine improved Marksman reaction time with no comparator reported.Benefit seen	no number reported	—	—	Systematic review · 2019	PMID 31045484
Marksmanship (shooting) accuracy	Caffeine did not significantly improve Marksmanship (shooting) accuracy with no comparator reported.No clear benefit	no number reported	—	—	Systematic review · 2019	PMID 31045484
Attention and vigilance in sleep-deprived adults	Caffeine improved Attention and vigilance in sleep-deprived adults with no comparator reported.Benefit seen	no number reported	—	—	Systematic review · 2017	PMID 28969341
Complex reaction time in sleep-deprived adults	Caffeine improved Complex reaction time in sleep-deprived adults with no comparator reported.Benefit seen	no number reported	—	—	Systematic review · 2017	PMID 28969341
Problem solving and reasoning in sleep-deprived adults	Caffeine improved Problem solving and reasoning in sleep-deprived adults with no comparator reported.Benefit seen	no number reported	—	—	Systematic review · 2017	PMID 28969341

Endurance (10 outcomes)

Outcome measured	What trials found	Effect size	People studied	Duration	Type of evidence	Study
time to complete endurance running time trials	Caffeine improved time to complete endurance running time trials (−0.101; 95% CI = −0.190 to −0.012, p = 0.026) compared with placebo in 254. (95% CI = −0.190 to −0.012, p = 0.026)Benefit seen	−0.101	254	—	Meta-analysis · 2022	PMID 36615805
time to exhaustion in running tests	Caffeine improved time to exhaustion in running tests (0.392; 95% CI = 0.214 to 0.571; p < 0.001) compared with placebo in 254. (95% CI = 0.214 to 0.571; p < 0.001)Benefit seen	0.392	254	—	Meta-analysis · 2022	PMID 36615805
Endurance event performance (pooled standardised mean difference, 56 comparisons)	Caffeine improved Endurance event performance (pooled standardised mean difference, 56 comparisons) (0.33; 95% CI=0.21, 0.45; p=1.00) compared with placebo. (95% CI=0.21, 0.45; p=1.00)Benefit seen	0.33	—	—	Meta-analysis · 2019	PMID 30170953
mean power output	Caffeine improved mean power output (3.03 ± 3.07%; 0.23 ± 0.15) compared with placebo.Benefit seen	3.03 ± 3.07%	—	—	Meta-analysis · 2018	PMID 29876876
time-trial completion time	Caffeine improved time-trial completion time (2.22 ± 2.59%; 0.41 ± 0.2) compared with placebo.Benefit seen	2.22 ± 2.59%	—	—	Meta-analysis · 2018

Strength (17 outcomes)

Outcome measured	What trials found	Effect size	People studied	Duration	Type of evidence	Study
Mean velocity	Caffeine supplementation improved Mean velocity (0.80 SMD; 95% CI: 0.48-1.12; p < 0.001) with no comparator reported. (95% CI: 0.48-1.12; p < 0.001)Benefit seen	0.80 SMD	—	—	Meta-analysis · 2020	PMID 31643020
Movement velocity (overall)	Caffeine supplementation improved Movement velocity (overall) (0.62 SMD; 95% confidence interval [CI]: 0.39-0.84; p < 0.001) with no comparator reported. (95% confidence interval [CI]: 0.39-0.84; p < 0.001)Benefit seen	0.62 SMD	—	—	Meta-analysis · 2020	PMID 31643020
Movement velocity in lower-body exercises	Caffeine supplementation improved Movement velocity in lower-body exercises (0.82 SMD; 95% CI: 0.42-1.23; p < 0.001) with no comparator reported. (95% CI: 0.42-1.23; p < 0.001)Benefit seen	0.82 SMD	—	—	Meta-analysis · 2020	PMID 31643020
Movement velocity in upper-body exercises	Caffeine supplementation improved Movement velocity in upper-body exercises (0.59 SMD; 95% CI: 0.37-0.82; p < 0.001) with no comparator reported. (95% CI: 0.37-0.82; p < 0.001)Benefit seen	0.59 SMD	—	—	Meta-analysis · 2020	PMID 31643020
Movement velocity with high loads	Caffeine supplementation improved Movement velocity with high loads (0.70 SMD; 95% CI: 0.33-1.07; p < 0.001) with no comparator reported. (95% CI: 0.33-1.07; p < 0.001)