Decaf is the obvious answer for anyone who loves coffee but needs to avoid caffeine, whether because of sleep, anxiety, pregnancy, a heart rhythm problem or a medicine. The name suggests the caffeine is gone. It is not, quite, and the search suggestions reflect that uncertainty: people want to know whether decaf is really caffeine free, how much caffeine it contains, and whether it will still keep them awake. Laboratory measurements answer those questions quite precisely, and the answers are reassuring for most people and worth knowing for a few.
How decaffeination works
Caffeine is removed from green, unroasted coffee beans before they are roasted. A review of decaffeination methods by Ramalakshmi and Raghavan, in Critical Reviews in Food Science and Nutrition[1], describes the main approaches as using organic solvents, water, or supercritical carbon dioxide, a form of carbon dioxide held under pressure so that it behaves partly like a liquid and dissolves caffeine. Each method has trade-offs in cost, how selectively it removes caffeine, and how much it affects other compounds and flavour. None removes every molecule of caffeine, which is why decaf is decaffeinated rather than caffeine free.
How much caffeine is left in decaf?
The most direct measurements come from McCusker and colleagues, in the Journal of Analytical Toxicology[2]. In the first phase, they collected ten decaffeinated coffee samples from different coffee establishments. Caffeine ranged from 0 to 13.9 mg per 16-ounce serving. In the second phase, they bought the same decaf drinks repeatedly from one outlet to check consistency: decaf espresso shots contained 3.0 to 15.8 mg of caffeine per shot, and brewed decaf contained 12.0 to 13.4 mg per 16-ounce serving.
An earlier study by the same group[3] tested specialty coffees and found that coffees sold as decaffeinated contained less than 17.7 mg of caffeine per serving. The same study found regular caffeinated specialty coffees contained 58 to 259 mg per serving, with a 16-ounce brewed coffee averaging 188 mg.
Putting those figures side by side shows what decaffeination achieves. A large brewed decaf with around 13 mg of caffeine contains well under a tenth of the caffeine in the average large regular coffee from the same study. For most people that is a meaningful reduction. But it is not zero, and the amounts vary from cup to cup.
Why several decafs can add up
The researchers who measured decaf were explicit about why it matters. McCusker and colleagues[2] noted that people on caffeine-restricted diets for medical reasons, or taking medicines that interact with caffeine, may switch to decaf believing they have eliminated caffeine. They advised that patients vulnerable to caffeine's effects should be told caffeine may be present in coffees described as decaffeinated, and they called for more research into the possible effects of this residual caffeine and the possibility of physical dependence at such low doses.
The variation is as important as the average. Even the same decaf espresso from the same shop ranged roughly fivefold in caffeine from one shot to the next in that study, from 3.0 to 15.8 mg, so a person who drinks decaf every day will not get a consistent dose. That unpredictability is not a safety problem for most people, but it helps explain why a very caffeine-sensitive person might feel one decaf and not the next.
In practical terms, one or two cups of decaf add only a small amount of caffeine. Five or six large decafs a day, or several decaf espresso drinks with multiple shots, could add up to an amount similar to a small regular coffee. For most people that is irrelevant. For someone told to avoid caffeine strictly, or taking a medicine that dramatically slows caffeine clearance, it is worth knowing.
Can such a small amount of caffeine have any effect?
It is tempting to assume that 10 or 15 mg of caffeine is too little to notice. A study by Smit and Rogers, in Psychopharmacology[4], suggests otherwise, at least under some conditions. In a double-blind study, 23 adults with low or moderate-to-high caffeine habits, after overnight caffeine abstinence, received 0, 12.5, 25, 50 or 100 mg of caffeine. All doses, including 12.5 mg, significantly affected cognitive performance, and the dose-response relationship was rather flat. Effects on performance were more marked in people with higher habitual caffeine intake.
That does not mean a decaf will keep you awake at night. The study measured performance and mood after abstinence, not sleep, and it used caffeine rather than decaf coffee. But it shows that small doses of caffeine are not pharmacologically meaningless, which is one reason people who are very sensitive to caffeine sometimes notice decaf.
Decaf keeps most of coffee's other compounds
Removing caffeine does not turn coffee into flavoured water. Farah and colleagues, in the Journal of Agricultural and Food Chemistry[5], examined how water decaffeination affects chlorogenic acids, the main family of polyphenols in coffee, and related compounds called chlorogenic acid lactones. In green beans, decaffeination actually raised total chlorogenic acid levels on a dry-weight basis by 16% on average. After roasting, chlorogenic acid levels in decaffeinated coffee were 3% to 9% lower than in regular coffee, while lactone levels were 5.5% to 18% higher. The authors described the differences as relatively small, though possibly enough to affect flavour and biological properties.
A survey of 65 capsule coffees from five major brands in Italy by Angelino and colleagues[6] likewise found that decaffeinated capsules still contained chlorogenic acids and other compounds, while regular espresso capsules had higher levels of trigonelline, N-methylpyridinium and niacin than lungo and decaf capsules. Caffeine in the regular capsules ranged from 54 to 208 mg per serving.
Why decaf shows up in the health research
Because decaf keeps most of coffee's other compounds, it is a useful tool for researchers trying to work out whether caffeine or something else explains coffee's health associations. In the large UK Biobank analysis by Loftfield and colleagues[7], the inverse association between coffee drinking and mortality looked similar for decaffeinated coffee as well as for ground and instant coffee. In the dose-response meta-analysis of type 2 diabetes by Ding and colleagues[8], decaffeinated coffee was also associated with lower risk, only slightly less strongly than caffeinated coffee.
Those findings are observational and do not prove decaf is protective, but they do suggest that much of what coffee is associated with is not about caffeine. We cover them in our article on coffee and lifespan and our article on coffee and diabetes.
What decaf does and does not do, across the evidence
Because this batch of articles looked at many everyday coffee questions, decaf turns up again and again as a comparison, and the pattern is instructive. For blood pressure, a review by Mesas and colleagues[9] found that caffeine raised blood pressure for several hours in people with hypertension, but that two weeks of coffee did not raise blood pressure compared with a caffeine-free diet or with decaf. Decaf avoids the acute caffeine spike, which is exactly what someone worried about blood pressure readings would want.
For gout, the long-term study of men by Choi, Willett and Curhan[10] found that decaf was also associated with lower risk: one to three cups a day had a relative risk of 0.67 compared with none. For weight, the three-cohort analysis by Henn and colleagues[11] found each extra daily cup of unsweetened decaf associated with 0.12 kg less weight gain over four years, the same as unsweetened caffeinated coffee. For kidney stones, Ferraro and colleagues[12] found a 16% lower risk with decaf, compared with 26% with caffeinated coffee.
For the bowel, decaf behaved more like coffee than like water. In the colonic pressure study by Rao and colleagues[13], decaf stimulated the colon more than water, though caffeinated coffee was about 23% stronger. So anyone hoping decaf will stop coffee sending them to the bathroom may be disappointed.
Taken together, these findings suggest a rough rule. Effects that are clearly driven by caffeine, such as the blood pressure spike, jitteriness and sleep disruption, are largely removed by decaf. Effects and associations that seem to come from coffee's other compounds, such as those for gout, weight, kidney stones, the bowel and possibly stomach acid, largely remain. All the long-term associations here are observational, so this is a pattern in the evidence rather than a proven set of effects.
Decaf still affects the stomach
People sometimes switch to decaf expecting it to be gentler on the stomach, and the evidence is mixed on that. The experiment by Cohen and Booth[14] found decaf stimulated stomach acid almost as much as regular coffee, 16.5 compared with 20.9 mEq per hour at maximum. On the other hand, a pH-probe study by Wendl and colleagues[15] found that decaffeinating coffee significantly reduced the acid reflux it caused, even though adding caffeine to water did not increase reflux. So decaf may help with reflux for some people while doing little to reduce stomach acid production. The details are in our article on coffee and acid reflux.
Decaf and caffeine withdrawal
One surprising finding concerns what decaf does for regular coffee drinkers who are trying to cut down. Mills, Boakes and Colagiuri[16] studied regular coffee drinkers who had abstained from caffeine for 24 hours and then gave all of them decaf. Half were told it was regular coffee and half were told it was decaf. Those who believed they had caffeine reported significantly greater reductions in cravings, fatigue, lack of alertness and flu-like feelings.
A follow-up by Mills and colleagues[17] went further, with 61 heavy coffee drinkers after 24 hours of abstinence. One group was given decaf and told it was caffeinated, one was given decaf and told honestly it was decaf, and a control group was given water. The open-label decaf group, who knew they were drinking decaf, still had a significant 9.5-point reduction in caffeine withdrawal symptoms, greater than the water group, though less than the group deceived into thinking it was regular coffee. The authors concluded that decaf can reduce caffeine withdrawal symptoms even when people know it is decaf.
That is useful for anyone cutting back. The ritual, taste and smell of coffee appear to carry part of the relief, so swapping some cups for decaf may make reducing caffeine easier than stopping coffee altogether. Part of that effect may also be the small amount of residual caffeine in decaf itself.
Decaf during pregnancy and for sensitive people
For people who need to limit caffeine, such as during pregnancy, decaf is a practical way to keep drinking coffee while dramatically reducing caffeine. The measurements above suggest a typical cup of decaf contains a small fraction of the caffeine in regular coffee. But because decaf is not caffeine free, many cups a day still add up, and the variation between cups means the actual amount is uncertain. Guidance on caffeine in pregnancy is covered in our article on caffeine during pregnancy.
The limits of this evidence
The caffeine measurements come from specific outlets and brands in the United States and Italy at particular points in time, so they illustrate the range rather than guarantee what a particular decaf contains. The decaffeination chemistry study looked at water-processed Arabica. The low-dose caffeine study tested caffeine after abstinence rather than decaf coffee in everyday use. The health associations for decaf are observational. And we found no study measuring decaf caffeine levels in coffees sold in the Philippines.
The practical answer
Decaf coffee is not caffeine free, but it is very low in caffeine. Laboratory tests found about 0 to 14 mg in a large brewed decaf and up to about 16 mg in a decaf espresso shot, compared with 58 to 259 mg in regular specialty coffees. For most people, a couple of decafs is a negligible amount of caffeine. Very sensitive people, those on strict caffeine limits, and anyone taking a medicine that sharply slows caffeine clearance should know that several decafs a day can add up.
Decaf also keeps most of coffee's other compounds, which is why it appears in health research alongside regular coffee, and it can ease caffeine withdrawal even when you know it is decaf. For how decaf is made, see our guide to decaffeination, and for medicines that interact with caffeine, our article on caffeine and medication. Nothing here is medical advice.