The short answer: In large cohort studies, people who drink coffee die at modestly lower rates than people who do not. The lowest risk shows up around 2 to 4 cups a day, about 10% to 20% lower than non-drinkers, and decaf shows a similar pattern. These are associations, not proof, so coffee is a fine habit to keep but not a reason to start drinking more.
What did the big coffee studies find?
The best-known study followed 229,119 men and 173,141 women, aged 50 to 71, in the NIH-AARP Diet and Health Study, published in the New England Journal of Medicine in 2012. Over 13 years, about 52,500 of them died.
At first glance, coffee drinkers looked worse off. They were also more likely to smoke. Once the researchers adjusted for smoking and other factors, the picture flipped. Men drinking 4 to 5 cups a day had a 12% lower risk of death than non-drinkers, and women had a 16% lower risk.
Lower risk showed up for deaths from heart disease, respiratory disease, stroke, injuries, diabetes, and infections. It did not show up for cancer deaths. The authors wrote plainly that they could not tell whether the link was causal.
A second large study used the UK Biobank. It included 498,134 adults, with 14,225 deaths over 10 years. In that 2018 JAMA Internal Medicine analysis, coffee drinkers again had lower death rates at nearly every intake level, from 1 cup to 8 or more cups a day.
A 2017 umbrella review in The BMJ pooled 201 meta-analyses of observational research. Three to four cups a day, compared with none, was linked to a 17% lower risk of death from any cause. Coffee was more often linked to benefit than harm across dozens of outcomes.
How many cups is the sweet spot?
The curve is not a straight line. The benefit rises quickly over the first few cups and then flattens. More coffee does not keep buying more protection.
- Umbrella review: the largest risk reduction appeared at 3 to 4 cups a day.
- UK Biobank, 2022: a European Journal of Preventive Cardiology study of 449,563 adults found the greatest drop in deaths at 2 to 3 cups a day for ground, instant, and decaf coffee.
- NIH-AARP: risk was about the same at 2 to 3 cups, 4 to 5 cups, and 6 or more cups.
So think of it as a plateau. Most of the association appears by 2 or 3 cups, and heavier intake adds little. The UK Biobank data did not show clear harm at 6 or more cups either, but that is not a reason to aim high. Caffeine side effects climb with dose.
A note on cup size: studies count a "cup" loosely. MedlinePlus puts an 8-ounce cup of coffee at 95 to 200 mg of caffeine. A large café drink can hold two or three of those study cups.
Does decaf count, or is it the caffeine?
Decaf drinkers show much of the same pattern. In the 2022 UK Biobank analysis, 2 to 3 cups of decaf a day was linked to a 14% lower risk of death. Ground coffee at that level was linked to a 27% lower risk, and instant to an 11% lower risk.
The 2018 UK Biobank study went further. It looked at genes that control how fast the body clears caffeine. The link between coffee and lower mortality held for fast and slow metabolizers alike. The authors concluded that compounds other than caffeine likely matter. So the benefit, if real, may come from other compounds in coffee rather than the caffeine.
One difference did show up. Ground and instant coffee were linked to fewer heart rhythm problems, but decaf was not.
Do sugar and cream cancel the benefit?
A little sugar does not seem to erase it. A 2022 Annals of Internal Medicine study followed 171,616 UK Biobank adults for a median of 7 years. Compared with non-drinkers, people who drank unsweetened coffee had a 16% to 29% lower risk of death across intake levels.
People who drank sugar-sweetened coffee also had lower risk at moderate intake, about 28% to 31% lower at 1.5 to 3.5 cups a day. The benefit faded at the high end. At more than 4.5 sweetened cups a day, risk was no lower than for non-drinkers. Results for artificially sweetened coffee were inconsistent.
The study did not look at cream separately. It also cannot tell you how a syrup-heavy café drink compares. A plain cup and a 500-calorie blended drink are not the same thing, and the research only really covers the first. For more on how drinks fit a long-life diet, see our longevity nutrition guide.
Why can't these studies prove coffee helps?
Every study above is observational. Researchers asked people how much coffee they drank, then counted deaths years later. That design has three known weak spots.
- Confounding. Coffee drinkers differ from non-drinkers in income, smoking, diet, and activity. The NIH-AARP study shows how strong this is: before adjusting for smoking, coffee looked harmful. Adjustment helps, but no model captures every difference.
- Reverse causation. People who feel unwell, or who have been told by a doctor to cut back, often quit coffee. That can make the non-drinker group look sicker than it would otherwise be.
- One-time measurement. Most of these studies measured intake once, at the start. Habits change over a decade or more.
The BMJ reviewers said the same thing: randomized trials are needed to show whether the links are causal. Those trials do not exist for mortality, and they probably never will. The same caution applies to other "drinkers live longer" findings, including older claims that moderate drinking of alcohol helps you live longer.
What the data can say with more confidence is reassuring. For most adults, moderate coffee drinking is not linked to higher mortality.
Who should limit caffeine?
The FDA cites 400 mg of caffeine a day, about two to three 12-ounce cups of coffee, as an amount not generally linked to negative effects in most adults. People vary widely in how sensitive they are and how fast they clear caffeine.
- Pregnancy. ACOG says intake under 200 mg a day does not appear to be a major factor in miscarriage or preterm birth. ACOG reaffirmed that guidance in 2026. The BMJ review also linked higher coffee intake in pregnancy to low birth weight and pregnancy loss.
- Breastfeeding. A small amount of caffeine passes into breast milk, so check with your clinician.
- Sleep problems and anxiety. MedlinePlus lists insomnia, anxiety, and restlessness among the effects of too much caffeine. If sleep is the issue, insomnia.md covers it in depth.
- Heart rhythm problems, GERD, ulcers, and migraines. MedlinePlus advises asking your provider about limits if you have any of these.
- Women at risk of fracture. The BMJ review found an association between coffee and fracture risk in women but not in men.
- Children and teens. Medical experts advise against energy drinks for this age group, per the FDA.
The bottom line
Coffee drinkers do tend to live a bit longer in large studies, with the benefit leveling off around 2 to 4 cups a day. Decaf shows a similar pattern, and moderate sugar use does not appear to cancel it. The studies cannot prove coffee causes the benefit. If you enjoy coffee and tolerate it well, you have no reason to quit. If you don't drink it, starting for longevity is not supported by the evidence. Sleep, activity, and not smoking matter far more, as our review of longevity habits explains.
Last updated: September 2026. This article is for informational purposes only and does not constitute medical advice. Talk with your clinician about caffeine limits if you are pregnant, breastfeeding, or have a heart, sleep, or digestive condition.