Coffee for sensitive stomachs: what to drink and how to brew it

Researchers have looked at coffee and a sensitive stomach in two ways: what coffee does to acid-producing cells in a lab, and what is in the cup itself. Neither line of work has tested which coffee gives a particular person fewer symptoms.

What the lab work found

In 2010, Veronika Somoza of the University of Vienna and Thomas Hofmann of the Technical University of Munich exposed human stomach cells to several coffee preparations, including regular, dark roast and decaf. They reported to the American Chemical Society that caffeine, catechols and compounds called N-alkanoyl-5-hydroxytryptamides pushed the cells to secrete more acid (ScienceDaily summary). A fourth compound, N-methylpyridinium (NMP), did the opposite and blocked acid production.

NMP forms during roasting and isn't present in raw beans. Dark-roasted coffee can contain up to twice as much as light-roasted coffee. The researchers ran this on cell cultures and planned a trial in people next. Treat it as a reason to try a darker roast, and stop short of calling it proof.

Roast level and growing elevation

Roasting breaks down chlorogenic acids. In a 2009 study in the Journal of Agricultural and Food Chemistry, Joon-Kwan Moon, Hyui Sun Yoo and Takayuki Shibamoto measured 61 to 86 mg/g of total chlorogenic acids in green beans and 2 to 7 mg/g after roasting. Beans roasted at 230°C for 12 minutes kept about half. Beans roasted at 250°C for 21 minutes kept almost none. The authors found a general correlation between these acids and the coffee's pH.

Where the coffee grew matters too. In a study of Central American farms presented at the 2004 International Conference on Coffee Science, Edgardo Alpizar and Benoît Bertrand found that chlorogenic acid content rose with elevation. A darker roast of a lower-grown coffee changes two variables at once. Dark roasts taste heavier and more bitter, and some drinkers won't like the trade.

Cold brew

A 2018 study in Scientific Reports by Niny Rao and Megan Fuller at Thomas Jefferson University compared cold brew and hot brew from six light-roast coffees. The pH of the two came out about the same, between 4.85 and 5.13. Hot brew had higher total titratable acidity, a measure of all the acid in the cup. Cold brew also had lower antioxidant activity.

Lower titratable acidity is a real difference, but the study didn't test whether it changes heartburn. It's still one of the easier changes to make at home.

A basic batch: stir about 100 g of coffee ground coarse into 4 cups of cold water, refrigerate it for 14 to 16 hours, and strain it through a paper filter. Drink it over ice, or cut it 1:1 with hot water for a hot cup.

Caffeine and timing

Caffeine was one of the compounds that raised acid secretion in the 2010 cell study. Half-caf or decaf lowers the dose without changing the drink. Decaf still has some caffeine: a 2006 University of Florida study measured 8.6 to 13.9 mg in 16-ounce cups of chain-store decaf.

The American College of Gastroenterology's GERD guideline does not recommend cutting out every common trigger food as a default. It lists weight loss, raising the head of the bed and not eating 2 to 3 hours before bedtime as changes that may help. If coffee sets off your reflux every time, raise it with a doctor.

Low-acid brands

Some brands sell coffee marketed as low acid. Before you pay extra, ask how the brand measured the claim. The 2018 Jefferson study shows why the method matters: cold brew and hot brew had the same pH but different titratable acidity, so a coffee can look low-acid on one measure and not on another.

A two-week test

Change one thing per week so you can tell which change mattered.

Week one: switch to a dark roast and keep your usual brew method, cup size and time of day. After each cup, write down the coffee and how your stomach felt an hour later.

Week two: keep the dark roast and switch to cold brew. If week two goes better than week one, the brew method mattered. If both weeks beat your usual coffee by the same margin, the roast did the work. Then try half-caf the same way.