26Evidence A

Drink three to four cups of coffee a day, keeping total caffeine under 400 mg, without sugar or sweet foam toppings; pregnant women, women at high risk of fractures, and people who get palpitations, anxiety, or insomnia from coffee are excluded

People who drink three to four cups of coffee a day have about a 17% lower risk of death than non-drinkers. On days when people drink coffee, extra heartbeats from the lower chambers (premature ventricular contractions) increase, and blood pressure in people with hypertension rises for several hours after drinking. Over the years, however, regular coffee drinkers do not develop more new arrhythmias. Caffeine content varies widely from cup to cup, so keep total intake under 400 mg a day.

Cost

Home-brewed coffee costs CNY 1–2 a day. Brewing takes a fe…

Benefit

An umbrella review synthesized 201 meta-analyses whose underlying primary studies were all observational witho…

Cost

Home-brewed coffee costs CNY 1–2 a day. Brewing takes a few minutes. With large cups, pay attention to how many milligrams of caffeine you consume in a day.

Benefit

An umbrella review synthesized 201 meta-analyses whose underlying primary studies were all observational without random assignment. For all-cause mortality, the umbrella review cited a dose-response meta-analysis: compared with non-drinkers, people drinking 3–4 cups a day had a 17% lower risk of death (relative risk 0.83, 95% CI 0.79–0.88). At 7 cups a day, the reduction narrowed to 10% (relative risk 0.90, 95% CI 0.85–0.96), with no rebound in risk observed. A study from the US National Health and Nutrition Examination Survey (NHANES) tracked 40,725 adults for a median follow-up of 9.8 years, also observational. The study divided coffee drinkers into two patterns. The morning pattern (drinking concentrated between 4:00 and 11:59, with almost none in the afternoon or evening) accounted for 36% of participants. Compared with non-drinkers, their all-cause mortality risk was 12% lower (hazard ratio 0.88, 95% CI 0.81–0.96); after further adjusting for short sleep and sleep difficulties, it was 16% lower (hazard ratio 0.84, 95% CI 0.74–0.95). Cardiovascular mortality risk was 19% lower (hazard ratio 0.81, 95% CI 0.70–0.94); after adjusting for sleep, it was 31% lower (hazard ratio 0.69, 95% CI 0.55–0.87). The all-day pattern (drinking across morning, afternoon, and evening) accounted for 14%, and showed no clear difference in all-cause mortality (hazard ratio 0.99, 95% CI 0.90–1.10). Neither group showed a clear difference in cancer mortality. Daily caffeine limits have official backing. The European Food Safety Authority (EFSA) and Health Canada both set the upper limit for healthy adults at 400 mg per day across all sources combined, including tea, cola, and energy drinks. EFSA also recommends spacing intake throughout the day, with no single dose exceeding 200 mg. Caffeine content per cup varies widely. In EFSA tables, a 200 mL cup of filter coffee contains about 90 mg. In Health Canada tables, an 8 oz (237 mL) cup of brewed coffee averages 135 mg. In US Food and Drug Administration (FDA) tables, a 12 oz (about 355 mL) cup of brewed coffee contains 113–247 mg. Thus, the FDA equates 400 mg to roughly 2 to 3 12-oz cups, while the review equates it to roughly 4 cups. The heart and blood pressure do react in the hours after consumption. A randomized trial had 100 adults wear electrocardiogram monitors for 14 days, randomly assigning them daily via text message to drink caffeinated coffee or avoid caffeine. On coffee days, premature ventricular contractions averaged 154 per day versus 102 on caffeine-free days (rate ratio 1.51, 95% CI 1.18–1.94). Premature atrial contractions were 58 versus 53, showing no clear difference. On coffee days, participants slept 36 minutes less at night. Another meta-analysis looked specifically at patients with hypertension. A single dose of 200–300 mg caffeine raised systolic blood pressure by an average of 8.1 mmHg and diastolic blood pressure by 5.7 mmHg. The increase appeared within one hour and lasted at least 3 hours. In 3 studies where coffee was consumed daily for two weeks, blood pressure was no higher than when abstaining. Long-term, the UK Biobank tracked 386,258 people for an average of 4.5 years, also observationally. Each additional daily cup of coffee was associated with a 3% lower risk of new-onset arrhythmia (hazard ratio 0.97, 95% CI 0.96–0.98). An American Heart Association 2026 scientific statement summarized randomized trials in habitual coffee drinkers, finding that caffeinated coffee reduced atrial fibrillation episodes while increasing premature ventricular contractions.

Original sources

Poole R, Kennedy OJ, Roderick P, et al. (2017). Coffee consumption and health: umbrella review of meta-analyses of multiple health outcomes. BMJ, 359, j5024. https://doi.org/10.1136/bmj.j5024,含更正 BMJ (2018), 360, k194. https://doi.org/10.1136/bmj.k194(摘要里总死亡率的可信范围由 0.83–0.88 改为 0.79–0.88);Wang X, Ma H, Sun Q, et al. (2025). Coffee drinking timing and mortality in US adults. European Heart Journal, 46(8), 749-759. https://doi.org/10.1093/eurheartj/ehae871;Yang R, Lei Q, Liu Z, et al. (2024). Relationship between timing of coffee and tea consumption with mortality (total, cardiovascular disease and diabetes) in people with diabetes: the U.S. National Health and Nutrition Examination Survey, 2003–2014. BMC Medicine, 22, 526. https://doi.org/10.1186/s12916-024-03736-x;EFSA Panel on Dietetic Products, Nutrition and Allergies (2015). Scientific Opinion on the safety of caffeine. EFSA Journal, 13(5), 4102. https://doi.org/10.2903/j.efsa.2015.4102;EFSA 咖啡因专题页(每份含量表)https://www.efsa.europa.eu/en/topics/topic/caffeine;FDA. Spilling the Beans: How Much Caffeine Is Too Much? https://www.fda.gov/consumers/consumer-updates/spilling-beans-how-much-caffeine-too-much;Health Canada. Caffeine in foods. https://www.canada.ca/en/health-canada/services/food-nutrition/food-safety/food-additives/caffeine-foods.html;加糖:Liu D, Li ZH, Shen D, et al. (2022). Association of Sugar-Sweetened, Artificially Sweetened, and Unsweetened Coffee Consumption With All-Cause and Cause-Specific Mortality. Annals of Internal Medicine, 175(7), 909-917. https://doi.org/10.7326/M21-2977;Treskes RW, Clausen J, Marott JL, et al. (2023). Use of sugar in coffee and tea and long-term risk of mortality in older adult Danish men. PLOS ONE, 18(10), e0292882. https://doi.org/10.1371/journal.pone.0292882;房颤:Wong CX, Cheung CC, Montenegro G, et al. (2026). Caffeinated Coffee Consumption or Abstinence to Reduce Atrial Fibrillation. JAMA, 335(4), 317-325. https://doi.org/10.1001/jama.2025.21056;Joglar JA, Chung MK, Armbruster AL, et al. (2024). 2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation. Circulation, 149(1). https://doi.org/10.1161/CIR.0000000000001193;心律与血压:Marcus GM, Rosenthal DG, Nah G, et al. (2023). Acute Effects of Coffee Consumption on Health among Ambulatory Adults. New England Journal of Medicine, 388(12), 1092-1100. https://doi.org/10.1056/NEJMoa2204737;Mesas AE, Leon-Muñoz LM, Rodriguez-Artalejo F, Lopez-Garcia E. (2011). The effect of coffee on blood pressure and cardiovascular disease in hypertensive individuals: a systematic review and meta-analysis. American Journal of Clinical Nutrition, 94(4), 1113-1126. https://doi.org/10.3945/ajcn.111.016667;Kim EJ, Hoffmann TJ, Nah G, et al. (2021). Coffee Consumption and Incident Tachyarrhythmias: Reported Behavior, Mendelian Randomization, and Their Interactions. JAMA Internal Medicine, 181(9), 1185-1193. https://doi.org/10.1001/jamainternmed.2021.3616;Marcus GM, Hu FB, van Dam RM, et al. (2026). Caffeine and Cardiovascular Disease: A Scientific Statement From the American Heart Association. Circulation, 154(9), e345-e355. https://doi.org/10.1161/CIR.0000000000001454;Hyppönen E, Zhou A. (2021). Cardiovascular symptoms affect the patterns of habitual coffee consumption. American Journal of Clinical Nutrition, 114(1), 214-219. https://doi.org/10.1093/ajcn/nqab014

Open source link
Book note

The main caveat is that the review authors noted most evidence comes from observational studies; reliable randomized controlled trials are needed to confirm causality. A UK Biobank study found that people with hypertension, angina, or arrhythmia were already less likely to drink caffeinated coffee. Genetic analyses also support this: people with higher blood pressure drink less, and those with a higher resting heart rate are more likely to choose decaf. The authors of that study cautioned that comparing coffee drinkers to non-drinkers might mistake these baseline differences for benefits of coffee. The morning-timing benefit comes from only one US study, which is why it was not included in the item heading. Its authors noted the finding cannot be interpreted as causal; morning coffee drinkers might simply lead healthier lives overall. Another US study focused solely on people with diabetes found opposite results: those drinking coffee from dawn to mid-morning had a 25% higher mortality risk, while those drinking from mid-morning to noon had a 20% lower risk. Neither study compared whether noon or 2:00 PM is a better cutoff. Chapter 3, Item 4 (no caffeine after 2:00 PM) addresses sleep and should be calculated backward from your personal bedtime, which does not conflict with this recommendation. The 400 mg threshold is a safety upper limit, not a required target dose. Decaffeinated coffee offers similar, slightly smaller benefits, so there is no need to drink extra coffee just to reach that number. Pregnant women should not exceed 200 mg per day (EFSA) or 300 mg per day (Health Canada). Women at high risk of fractures are excluded from this advice. People with anxiety or insomnia are also excluded; this is the book authors' clinical judgment, as the umbrella review drew no conclusions on these groups. Those who experience palpitations should cut back or switch to decaf, which is likewise this book's judgment. People with atrial fibrillation do not need to quit coffee universally. In one trial of patients after electrical cardioversion for persistent atrial fibrillation, those drinking about 1 cup of coffee a day had fewer recurrences than abstainers (47% versus 64%). The 2023 US atrial fibrillation guidelines also do not recommend caffeine cessation to prevent episodes, though individuals whose symptoms are triggered by coffee should reduce intake. For other arrhythmias, evidence is lacking, so follow your doctor's advice. Adding sugar did not appear to negate the mortality benefit, but drinks with syrups and sweet foam toppings have not been studied and carry heavy sugar loads, so avoiding them is still recommended.

My note