Does Coffee Help You Live Longer? What 40 Studies of 3.85 Million Adults Show
What the main studies on coffee and death rates found, why decaf and a Danish genetic study complicate the story, and who should cut back.
Coffee drinkers have lower death rates than people who skip coffee, and the pattern has held across dozens of cohort studies. A 2019 analysis of 40 studies covering 3,852,651 adults found the lowest death rate at about 3.5 cups a day, 15% below non-drinkers. Drinking more than that came with no further drop, and the link held for decaf too.
What the research can't show is that coffee itself adds years. Every study here is observational, and a Danish genetic study found no sign that coffee drinking causes a longer life. The fair reading is that 2 to 4 cups a day of filtered coffee looks safe for most healthy adults and may help a little. It is not a reason to start if you don't enjoy it.
What the main studies found
The 2017 BMJ umbrella review. Researchers at the University of Southampton pulled together 201 meta-analyses of observational studies and 17 of trials. Compared with no coffee, 3 to 4 cups a day went with a 17% lower death rate from any cause and a 19% lower rate of death from cardiovascular disease. Across the outcomes they looked at, coffee was linked to benefit more often than to harm.
The 2019 meta-analysis of 40 studies. Kim, Je and Giovannucci found the curve flattened after about 3.5 cups a day. The lower death rate held across age groups, in smokers and non-smokers, at different body weights and for decaf as well as regular coffee. It was stronger in European and Asian studies than in US ones.
Ground, instant and decaf in UK Biobank. A 2022 study by Chieng and colleagues followed 449,563 adults, median age 58, for about 12.5 years. At 2 to 3 cups a day, compared with people who drank no coffee:
Ground coffee drinkers had a 27% lower death rate
Decaf drinkers had a 14% lower death rate
Instant coffee drinkers had an 11% lower death rate
All three types were also linked to fewer cases of coronary heart disease, heart failure and ischaemic stroke. Ground and instant coffee, but not decaf, were linked to fewer heart rhythm problems (arrhythmias). That runs against the common worry that coffee brings on rhythm problems, though people with a known rhythm disorder should still follow their cardiologist's advice.
Why decaf matters here
If caffeine were doing all the work, decaf drinkers would look like non-drinkers, but in these studies they had lower death rates too. A 2018 UK Biobank study of 498,134 adults, published in JAMA Internal Medicine, also found lower death rates at every level from 1 cup a day up to 8 or more. The pattern looked the same in people whose genes make them clear caffeine slowly and in those who clear it fast.
The authors took that as a sign that parts of coffee other than caffeine may matter. That is a reasonable idea, but it is still an idea, and the comparison itself has a problem. A 2021 UK Biobank analysis of up to 390,435 adults found that people with high blood pressure, angina or heart rhythm problems were more likely to drink less caffeinated coffee or switch to decaf. The authors warned that this can make coffee look better than it is when coffee drinkers are compared with non-drinkers or decaf drinkers.
What the genetic study adds
Coffee drinkers differ from non-drinkers in ways no questionnaire fully captures. To get around that, Danish researchers at Copenhagen University Hospital used gene variants that nudge people to drink more coffee, an approach called Mendelian randomization.
Among 95,366 Danes, moderate coffee drinkers had the lowest death rates. But in 112,509 Danes, the gene variants were linked to drinking 42% more coffee, yet showed no clear link with the death rate. That points away from coffee being a direct cause of a longer life, though one genetic study is not the last word.
Filtered or unfiltered
How you brew matters for your cholesterol, not just how much you drink. A meta-analysis of 14 trials found that unfiltered coffee, such as boiled coffee, raised total and LDL cholesterol. Filtered coffee raised it very little.
A Norwegian study of 508,747 adults followed for an average of 20 years fits that picture. Compared with no coffee, filtered coffee came with a 15% lower death rate in both men and women. Unfiltered coffee showed no clear difference in men and a 9% lower rate in women. Overall, unfiltered coffee was linked to higher death rates than filtered.
If your LDL cholesterol is already high, switching to filtered coffee is a simple step. Our page on limiting saturated fat covers the bigger dietary levers for LDL.
If you make filtered coffee at home, whole beans you grind yourself work well in a drip machine or pour-over.
As an Amazon Associate, Stacking Years earns from qualifying purchases. This does not affect our rankings.
A specialty-grade organic light roast, sold as whole beans for grinding at home and brewing in a paper-filter drip or pour-over. The studies compared coffee types as people reported them, not this brand.
Best for: brewing filtered coffee at home if you already drink coffee
Not fully, in one large study that checked. A 2022 UK Biobank study of 171,616 adults without heart disease or cancer, followed for a median of 7 years, split coffee by how people sweetened it.
Unsweetened coffee was linked to a lower death rate at every amount, from under 1.5 drinks to more than 4.5 drinks a day
Sugar-sweetened coffee was linked to a lower death rate at more than 1.5 and up to 3.5 drinks a day. At other amounts there was no clear difference from non-drinkers
Results for artificially sweetened coffee were less consistent
People reported their coffee habits once, at the start, so changes over the years were missed. Our page on limiting added sugar covers sugar in the wider diet.
Who should cut back
Pregnancy. This is where the umbrella review found the clearest warning signs. High coffee intake in pregnancy was linked to low birth weight, preterm birth and pregnancy loss. The European Food Safety Authority (EFSA) advises that for women who are pregnant or breastfeeding, up to 200 mg of caffeine a day from all sources raises no safety concerns for the unborn baby. A 200 ml cup of filter coffee has roughly 90 mg.
Fracture risk. The same review found a link between coffee and fractures in women but not in men. Its authors suggested leaving women at higher fracture risk out of future coffee trials. If you have low bone density, ask your doctor how much coffee suits you.
Sleep. In a small trial of 12 healthy sleepers, 400 mg of caffeine taken 6 hours before bed cut sleep by more than an hour on a sleep monitor, and the sleepers' own diaries didn't pick it up. EFSA notes that even 100 mg close to bedtime can affect sleep in some adults. Our caffeine cutoff calculator estimates how much is left in your system at bedtime.
Blood pressure. In people with high blood pressure, 200 to 300 mg of caffeine raised the top number by about 8 points within an hour, and the rise lasted at least 3 hours. Trials of drinking coffee for 2 weeks found no lasting rise. Our review of caffeine and blood pressure covers the trial data.
For healthy adults outside pregnancy, EFSA says up to 400 mg of caffeine a day raises no safety concerns, roughly four 200 ml cups of filter coffee. Cup sizes and brews vary, so check what you're actually drinking. If you have a heart rhythm problem, high blood pressure or take medicines that interact with caffeine, ask your doctor what amount suits you.
If you are cutting caffeine, decaf beans made for drip or pour-over let you keep the morning cup.
An organic decaf coffee processed with the Swiss Water method, sold as whole beans for drip or pour-over. Decaf drinkers also had lower death rates in the studies, but that finding is observational, and some people switch to decaf because of heart symptoms.
Best for: if you need to cut caffeine and still want coffee
Evidence grade: moderate for the link between 2 to 4 cups a day and a lower death rate. It is consistent across dozens of cohorts and millions of people, and it holds for decaf. But all of it is observational.
Evidence grade: limited for coffee directly adding years to your life. The Danish genetic study found no sign of a causal link. People with heart symptoms also tend to drink less coffee, which flatters coffee in these comparisons. The BMJ review's authors called for proper trials to test whether the link is causal.
FAQ
How many cups of coffee a day is linked to living longer?
In the pooled analyses, the lowest death rates were at about 3 to 4 cups a day compared with none. Above about 3.5 cups, the rate stopped falling.
Is decaf coffee as good as regular for longevity?
Decaf drinkers also had lower death rates in several studies, including UK Biobank, where 2 to 3 cups a day of decaf was linked to a 14% lower rate. Ground coffee showed a larger drop in that study. Part of the decaf picture may come from people with heart symptoms switching to decaf, so the gap is hard to read.
Should I start drinking coffee to live longer?
There's no evidence that starting coffee adds years. If you already enjoy it and sleep well, 2 to 4 cups of filtered coffee a day fits a healthy diet for most adults. If you don't drink it, there's no need to start.
Is instant coffee bad for you?
Not on the death-rate data. In UK Biobank, 2 to 3 cups a day of instant coffee was linked to an 11% lower death rate than no coffee, a smaller drop than for ground coffee.
This is general information, not medical advice. Talk to your doctor if you are pregnant, have a heart condition or high blood pressure, or take regular medicines.
In Short
Coffee drinkers had lower death rates than non-drinkers, lowest at about 3 to 4 cups a day
Decaf shows a similar pattern, so caffeine may not be the main factor
A genetic study found no sign that coffee itself causes a longer life
Filtered coffee is the better choice if your cholesterol is high
EFSA's caffeine limits are 200 mg a day in pregnancy and 400 mg a day for other healthy adults
The sleep trial's authors advise no substantial caffeine within 6 hours of bedtime
Get the free Longevity Supplement Starter Guide
Sign up for the guide and occasional Stacking Years updates. The guide is available on the confirmation screen.
Sources
Kim Y, Je Y, Giovannucci E. Coffee consumption and all-cause and cause-specific mortality: a meta-analysis by potential modifiers. European Journal of Epidemiology. 2019;34(8):731-752. doi:10.1007/s10654-019-00524-3. PMID 31055709. Meta-analysis of 40 studies, 3,852,651 participants, 450,256 deaths; non-linear dose-response. Lowest relative risk at 3.5 cups/day for all-cause mortality (RR 0.85, 95% CI 0.82 to 0.89), 2.5 cups/day for CVD mortality (RR 0.83, 0.80 to 0.87), 2 cups/day for cancer mortality (RR 0.96, 0.94 to 0.99); additional intake not associated with further lower mortality. Inverse association maintained irrespective of age, overweight status, alcohol, smoking and caffeine content; stronger in Europe and Asia than the US.
Poole R, Kennedy OJ, Roderick P, Fallowfield JA, Hayes PC, Parkes J. Coffee consumption and health: umbrella review of meta-analyses of multiple health outcomes. BMJ. 2017;359:j5024. doi:10.1136/bmj.j5024. PMID 29167102. Correction: BMJ 2018;360:k194 (PMID 29330262; corrected all-cause CI is 0.79 to 0.88, findings unchanged). 201 meta-analyses of observational research (67 outcomes) and 17 of interventional research (9 outcomes). 3 to 4 cups/day vs none: all-cause mortality RR 0.83 (0.79 to 0.88), cardiovascular mortality RR 0.81 (0.72 to 0.90), CVD RR 0.85 (0.80 to 0.90). Pregnancy, high vs low/no intake: low birth weight OR 1.31 (1.03 to 1.67), preterm birth first trimester 1.22 (1.00 to 1.49), second trimester 1.12 (1.02 to 1.22), pregnancy loss 1.46 (1.06 to 1.99). Fracture association in women, not men. Authors call for robust RCTs to test causality. Two authors report pharmaceutical fees/grants outside the submitted work.
Chieng D, Canovas R, Segan L, Sugumar H, Voskoboinik A, Prabhu S, Ling LH, Lee G, Morton JB, Kaye DM, Kalman JM, Kistler PM. The impact of coffee subtypes on incident cardiovascular disease, arrhythmias, and mortality: long-term outcomes from the UK Biobank. European Journal of Preventive Cardiology. 2022;29(17):2240-2249. doi:10.1093/eurjpc/zwac189. PMID 36162818. 449,563 participants (median 58 years, 55.3% female), follow-up 12.5 ± 0.7 years; Cox regression vs non-drinkers; outcomes from ICD codes and death records. All-cause mortality at 2 to 3 cups/day: decaffeinated HR 0.86 (0.81 to 0.91), ground HR 0.73 (0.69 to 0.78), instant HR 0.89 (0.86 to 0.93). Arrhythmia lower with ground and instant at 1 to 5 cups/day (lowest: ground 4 to 5 cups HR 0.83, 0.76 to 0.91; instant 2 to 3 cups HR 0.88, 0.85 to 0.92), not with decaf. All subtypes linked to lower incident CVD (coronary heart disease, heart failure, ischaemic stroke). Several authors report device-industry funding outside the submitted work.
Loftfield E, Cornelis MC, Caporaso N, Yu K, Sinha R, Freedman N. Association of coffee drinking with mortality by genetic variation in caffeine metabolism: findings from the UK Biobank. JAMA Internal Medicine. 2018;178(8):1086-1097. doi:10.1001/jamainternmed.2018.2425. PMID 29971434. 498,134 participants, mean age 57, about 10 years of follow-up, 14,225 deaths. vs non-drinkers: <1 cup/day HR 0.94 (0.88 to 1.01), 1 cup 0.92 (0.87 to 0.97), 2 to 3 cups 0.88 (0.84 to 0.93), 4 to 5 cups 0.88 (0.83 to 0.93), 6 to 7 cups 0.84 (0.77 to 0.92), 8+ cups 0.86 (0.77 to 0.95). Similar for instant, ground and decaffeinated coffee; no effect modification by genetic caffeine metabolism score (AHR, CYP1A2, CYP2A6, POR; P = .17 for heterogeneity).
Nordestgaard AT, Nordestgaard BG. Coffee intake, cardiovascular disease and all-cause mortality: observational and Mendelian randomization analyses in 95 000-223 000 individuals. International Journal of Epidemiology. 2016;45(6):1938-1952. doi:10.1093/ije/dyw325. PMID 28031317. Observational: U-shaped associations in 95,366 White Danes, lowest risk with medium intake. Genetic: caffeine intake allele score (rs4410790 + rs2470893) associated with 42% higher coffee intake; per-allele HR for all-cause mortality 1.01 (0.99 to 1.03) and cardiovascular mortality 1.02 (0.99 to 1.06) in 112,509 Danes; ischaemic heart disease up to 223,414 individuals with consortia. Conclusion: genetically predicted caffeine intake not associated with CVD or all-cause mortality.
Jee SH, He J, Appel LJ, Whelton PK, Suh I, Klag MJ. Coffee consumption and serum lipids: a meta-analysis of randomized controlled clinical trials. American Journal of Epidemiology. 2001;153(4):353-362. doi:10.1093/aje/153.4.353. PMID 11207153. 14 trials; dose-response relation between coffee and total and LDL cholesterol (p < 0.01); larger increases in trials of boiled or caffeinated coffee and in people with hyperlipidaemia; filtered coffee showed very little increase. Conclusion: unfiltered, but not filtered, coffee increases total and LDL cholesterol.
Tverdal A, Selmer R, Cohen JM, Thelle DS. Coffee consumption and mortality from cardiovascular diseases and total mortality: does the brewing method matter? European Journal of Preventive Cardiology. 2020;27(18):1986-1993. doi:10.1177/2047487320914443. PMID 32320635. 508,747 Norwegian men and women aged 20 to 79, mean follow-up 20 years, 46,341 deaths. All-cause mortality vs no coffee, men: filtered HR 0.85 (0.82 to 0.90), both 0.84 (0.79 to 0.89), unfiltered 0.96 (0.91 to 1.01); women: filtered 0.85 (0.81 to 0.90), both 0.79 (0.73 to 0.85), unfiltered 0.91 (0.86 to 0.96). Conclusion: unfiltered brew associated with higher mortality than filtered brew.
Liu D, Li ZH, Shen D, et al. Association of sugar-sweetened, artificially sweetened, and unsweetened coffee consumption with all-cause and cause-specific mortality: a large prospective cohort study. Annals of Internal Medicine. 2022;175(7):909-917. doi:10.7326/M21-2977. PMID 35635846. UK Biobank, 171,616 participants without CVD or cancer, mean age 55.6, median follow-up 7.0 years, 3,177 deaths. Unsweetened vs non-consumers (>0 to 1.5, >1.5 to 2.5, >2.5 to 3.5, >3.5 to 4.5, >4.5 drinks/day): HR 0.79, 0.84, 0.71, 0.71, 0.77 (all CIs below 1). Sugar-sweetened: 0.91 (0.78 to 1.07), 0.69 (0.57 to 0.84), 0.72 (0.57 to 0.91), 0.79 (0.60 to 1.06), 1.05 (0.82 to 1.36). Artificially sweetened less consistent. Intake self-reported once at baseline. Funded by Chinese national and provincial science grants.
Hyppönen E, Zhou A. Cardiovascular symptoms affect the patterns of habitual coffee consumption. American Journal of Clinical Nutrition. 2021;114(1):214-219. doi:10.1093/ajcn/nqab014. PMID 33711095. Up to 390,435 European-ancestry UK Biobank participants aged 39 to 73. People with essential hypertension, angina or arrhythmia were more likely to drink less caffeinated coffee and to be non-habitual or decaffeinated drinkers. Mendelian randomization: higher systolic and diastolic blood pressure causally linked to lower caffeinated coffee intake (MR-Egger β about 0.21 fewer cups/day per 10 mm Hg higher SBP); higher resting heart rate linked to greater odds of drinking decaf (OR 1.71 per 10 beats/min). Authors: observational coffee studies are prone to reverse causation; caution when comparing with abstainers or decaf drinkers.
Mesas AE, Leon-Muñoz LM, Rodriguez-Artalejo F, Lopez-Garcia E. The effect of coffee on blood pressure and cardiovascular disease in hypertensive individuals: a systematic review and meta-analysis. American Journal of Clinical Nutrition. 2011;94(4):1113-1126. doi:10.3945/ajcn.111.016667. PMID 21880846. In 5 trials in hypertensive people, 200 to 300 mg caffeine raised systolic BP by a mean 8.1 mm Hg (95% CI 5.7 to 10.6) and diastolic by 5.7 mm Hg, starting in the first hour and lasting at least 3 hours. In 3 studies of 2 weeks of coffee, no increase in BP vs a caffeine-free diet or decaf. 7 cohort studies found no association between habitual coffee and higher CVD risk in hypertensive people.
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;9(11):1195-1200. doi:10.5664/jcsm.3170. PMID 24235903. Randomized, double-blind, placebo-controlled crossover in 12 healthy normal sleepers; 400 mg caffeine at 0, 3 or 6 hours before habitual bedtime; all three disrupted sleep vs placebo (p < 0.05). Full text (PMC3805807): at 6 hours caffeine reduced total sleep time by more than 1 hour on the objective monitor; self-report did not detect the 6-hour effect.
European Food Safety Authority (EFSA), Panel on Dietetic Products, Nutrition and Allergies. Scientific Opinion on the safety of caffeine. EFSA Journal. 2015;13(5):4102. doi:10.2903/j.efsa.2015.4102. Summary on EFSA's caffeine topic page (https://www.efsa.europa.eu/en/topics/topic/caffeine): up to 400 mg/day raises no safety concerns for healthy adults except pregnant women; up to 200 mg/day from all sources raises no safety concerns for the foetus (pregnant/lactating women); single doses of 100 mg may affect sleep in some adults close to bedtime. Guide amounts: espresso (60 ml) 80 mg, filter coffee (200 ml) 90 mg.