Caffeine

Studies in people

• Open question

  Exercise performance²

  Caffeine can boost endurance, strength and focus during sport, but effects vary between people.

• Concern

  Pregnancy risks³

  In studies of pregnant women, caffeine was linked to negative pregnancy outcomes, but the evidence is mostly observational.

• Open question

  Appetite and eating⁴

  Caffeine before a meal may slightly cut how much you eat, but it is unclear if it truly curbs appetite.

• Open question

  Muscle strength and sprinting⁵

  Caffeine may improve endurance exercise, but results are mixed for strength and sprinting, and it may blunt creatine's benefits.

• Benefit

  Liver scarring risk⁷

  In US adults, caffeine intake of at least 78 mg a day was linked to a lower risk of liver fibrosis.

• Open question

  Physical and mental performance⁹

  Caffeine can boost physical and mental performance, but people respond differently depending on how their body breaks it down.

• No harm found

  Body fluid balance¹⁰

  Caffeine in tea, coffee and soft drinks does not cause extra fluid loss in regular drinkers.

• Benefit

  Kidney stones¹¹

  People who drink more coffee or take in more caffeine seem to get fewer kidney stones.

• Open question

  Brain and sleep effects¹²

  Caffeine can boost alertness and mood, though it may disturb sleep in sensitive people.

• Concern

  Sleep and heart effects¹³

  In athletes, caffeine supplements caused heart palpitations and trouble falling asleep, more so at higher doses.

• Benefit

  Kidney disease risk¹⁶

  People who consumed more caffeine were less likely to have chronic kidney disease, though this link is not proof of cause.

• Benefit

  Asthma symptoms¹⁷

  People who drank caffeine in coffee regularly were less likely to have current asthma symptoms than those who did not.

• Concern

  Stress hormone response¹⁹

  In people, caffeine raises the stress hormone cortisol, but daily caffeine makes that response smaller.

• Open question

  Inner ear vertigo²⁰

  People with Menière's disease drank more caffeine than people without it, but scientists do not yet know whether caffeine is the cause.

• Open question

  Leg muscle strength²¹

  In strength-trained men, caffeine capsules and caffeine gum improved leg strength and jumping.

• Concern

  Sleep quality in young women²²

  Young women who had sugary caffeine drinks, like energy drinks or high-calorie coffee drinks, reported poorer sleep.

• Open question

  Sleep quality in adults²⁴

  Adults who slept poorly drank more caffeine than good sleepers, and more caffeine went with less time in bed.

Animal and lab studies only

• Benefit

  Fatty liver disease²⁶

  In mice, caffeine reduced liver fat, inflammation and early scarring from a fatty liver diet.

Single cases of misuse or overdose

One person or one incident, such as an accidental overdose or a mistake in a kitchen or factory. They are real, but a single case does not show what happens at normal amounts, so the signal does not count them.

• Heart rhythm problems¹⁵

  A woman who accidentally took a huge dose of caffeine in a preworkout supplement developed a dangerous irregular heartbeat.

How it’s made and used

• Running endurance⁶

  In runners, caffeine helped them run longer before tiring and shaved a little time off races.

• Boxing power and speed⁸

  In eight Olympic male boxers, caffeine improved jumping and cycling power, but the study did not prove it helps punching.

• Soccer endurance and effort¹⁴

  In a small soccer study, caffeine plus a carb meal did not clearly boost performance, though players felt the effort was easier.

• Jumping and squat performance¹⁸

  In recreationally trained people, caffeine changed jump and squat force readings used to judge athlete readiness.

• Running performance²³

  Caffeine, alone or with beetroot juice, made no difference to runners' speed or effort in this study.

• Kidney and water balance²⁵

  Caffeine makes the kidneys filter more and hold onto less salt and water, so you pee more.

The signal at the top counts the findings in people and in animal and lab studies. A concern in studies of people outranks animal and lab studies, and “No concerns” needs at least one finding in people. When most findings in people point one way, the page leans that way (“Mostly”), and every finding is still shown. Findings whose study type is not stated are shown but not counted. This page is a reference, not medical advice. It describes what regulators and published research say. Talk to a doctor or dietitian about your own diet.

Expert panels that weigh all the research. Shown for context: the signal at the top does not count them. IARC rates whether something can cause cancer at all, not how likely it is at the amounts in food.

• WHO cancer agency (IARC), 1990

  Not enough evidence to say whether it causes cancer in people (Group 3).