Coffee’s biological activity comes from three main compound classes. Caffeine, a methylxanthine, blocks adenosine receptors in the brain, counteracting the neurochemical signal for sleepiness and elevating alertness, dopamine signaling, and reaction time. Chlorogenic acids, polyphenolic antioxidants, slow glucose absorption, improve insulin sensitivity, and reduce oxidative stress in liver and vascular tissue. Diterpenes (cafestol and kahweol, found primarily in unfiltered coffee) have both lipid-raising effects and emerging anti-cancer properties. Filtered coffee removes most diterpenes while preserving caffeine and chlorogenic acids.
Caffeine is the most widely used psychoactive substance globally. At doses of 100–400 mg (roughly 1–4 cups of brewed coffee) its effects onset within 30–45 minutes and last 4–6 hours depending on individual metabolic rate, and regular users develop tolerance to some of them.
What this page does not claim: none of the sources cited here measured alertness, attention, vigilance, reaction time, or mood as an outcome. The umbrella review below covers disease and mortality endpoints, the two cohort studies cover mortality and life expectancy, and the Examine overview is a general summary that could not be opened in its cited form. This page therefore carries no claim about acute cognitive performance, and will carry none until evidence that measured it is curated.
The BMJ umbrella review found coffee consumption consistently associated with a lower risk of type 2 diabetes: 30% lower for high versus low consumption (relative risk 0.70, 95% CI 0.65 to 0.75) and 6% lower per extra cup a day, with risk still falling between one and six cups. Decaffeinated coffee carried an association of similar magnitude, pointing to chlorogenic acids rather than caffeine as the primary driver. Liver outcomes carried the largest effect sizes of anything in the review: compared with no coffee, consumption was associated with a 29% lower risk of non-alcoholic fatty liver disease (0.71, 0.60 to 0.85), a 27% lower risk of liver fibrosis (odds ratio 0.73, 0.56 to 0.94), and a 39% lower risk of cirrhosis (0.61, 0.45 to 0.84), alongside lower risk of chronic liver disease and liver cancer.
The umbrella review reports a non-linear association between coffee consumption and lower risk of incident cardiovascular disease (relative risk 0.85, 95% CI 0.80 to 0.90), coronary heart disease (0.90, 0.84 to 0.97), and stroke (0.80, 0.75 to 0.86), with the largest benefit at three to five cups a day for those specific outcomes and no modification by sex; the review’s headline summary places the largest reduction across cardiovascular outcomes at three to four cups a day versus none. Cardiovascular mortality was 19% lower at three cups a day (0.81, 0.72 to 0.90). Intake above that level was not associated with harm; the magnitude of the benefit was smaller and the estimates at the highest intakes did not reach significance. All-cause mortality is a separate endpoint in the same review (relative risk 0.83 at three cups a day) and this claim deliberately does not fold it into the cardiovascular one. The evidence is pooled observational data that the review grades as low or very low quality, and the review notes that Mendelian randomisation found no genetic evidence for a causal relation with cardiovascular and all-cause mortality. Unfiltered coffee (French press, espresso) raises LDL cholesterol due to diterpenes; filtered coffee does not.
Coffee consumption was consistently associated with a lower risk of Parkinson’s disease in the umbrella review, across every exposure category it examined (high versus low, any versus none, one extra cup a day) and after adjustment for smoking; decaffeinated coffee pointed the same way but did not reach significance. Consumption was also associated with lower risk of cognitive disorders, most clearly for Alzheimer’s disease (relative risk 0.73, 95% CI 0.55 to 0.97) in meta-analyses of cohort studies. These are pooled observational associations, and consistency across exposure categories is not the same as demonstrated causation: the review’s own conclusion is that randomised trials are needed before the associations can be read as causal. Proposed mechanisms include caffeine’s adenosine blockade, chlorogenic acid’s antioxidant activity, and modulation of neuroinflammatory pathways.
A 2025 prospective cohort analysis of 43,114 NHANES participants aged 20 and over, followed for a median of 8.7 years, estimated life expectancy at age 50 across intake levels: 30.06 years at no coffee, 30.82 at up to 1 cup a day, 32.08 at 1 to 2 cups, 31.24 at 2 to 3 cups, and 31.45 above 3 cups. Moderate intake of 1 to 2 cups a day therefore corresponded to a gain of 2.02 years (95% CI 1.17 to 2.85) over non-drinkers, which the authors attribute largely to a 0.61-year reduction in cardiovascular deaths. Higher intakes did not add to that gain, and no intake group fell below non-drinkers. The caffeinated subgroup showed the same pattern (2.22 years at 1 to 2 cups a day), while the decaffeinated subgroup reached significance only above 3 cups a day (3.76 years, on a far smaller sample and with a correspondingly wide interval). This is a life expectancy estimate modeled from an observational cohort, not a measured effect of taking up coffee; the authors call for randomised trials to determine whether the relationship is causal.
A 2025 European Heart Journal study of 40,725 adults identified a meaningful interaction between coffee timing and mortality. Participants who concentrated their coffee intake in the morning had a 16% lower all-cause mortality risk (hazard ratio 0.84, 95% CI 0.74 to 0.95) and a 31% lower cardiovascular mortality risk (0.69, 0.55 to 0.87) compared to non-drinkers. The all-day drinking pattern showed no significant association with either. This finding is consistent with caffeine’s effects on cortisol and circadian rhythm: morning consumption aligns with the body’s natural cortisol peak and avoids disrupting sleep architecture via the 5–6 hour half-life of caffeine.
3–4 cups of filtered coffee per day represents the dose most associated with benefit in epidemiological research, with the strongest longevity signal in people who drink coffee primarily in the morning. Individuals sensitive to caffeine (anxiety, palpitations, poor sleep) should limit intake or shift consumption to early morning given caffeine’s 5–6 hour half-life. Pregnant women are advised to limit caffeine to under 200 mg/day. Specialty preparations (lattes, cappuccinos) may add significant sugar and saturated fat calories that dilute the health profile described here for plain brewed coffee.