DECAF trial

DECAF was a randomised trial that asked whether people with atrial fibrillation, the most common irregular heart rhythm, should give up coffee. They shouldn't need to: the irregular rhythm came back less often in those who kept drinking it than in those who quit, though the trial was small and the true size of the difference may well be smaller than it looked.

If you have atrial fibrillation and already drink coffee, giving it up is unlikely to keep the rhythm steady; quit only if you notice caffeine triggers your symptoms. The trial says nothing about starting coffee if you don't drink it, and other rhythm problems need individual advice.

What the trial did

Two hundred adults with persistent atrial fibrillation, or atrial flutter with a history of fibrillation, were enrolled at five hospitals in the US, Canada and Australia just before electrical cardioversion, the procedure that shocks the heart back into a normal rhythm. All were current coffee drinkers or had drunk coffee in the past five years. Half were asked to drink at least one cup of caffeinated coffee a day for six months; the other half to avoid coffee, decaf and all other caffeine. In practice the contrast was about one cup a day against none. The question was how many had the irregular rhythm (fibrillation or flutter) come back.^[1]

What it found

The rhythm came back in 47% of the coffee group against 64% of the abstainers — about seventeen fewer recurrences per hundred patients over six months, or a 39% lower risk (hazard ratio 0.61; 95% CI 0.42–0.89). That bracketed range is a confidence interval, the span where the true effect most plausibly lies; here it sits entirely below 1.0, so the benefit is unlikely to be chance alone. Side effects did not differ between the groups.^[2]

Why to hold the size loosely

The direction is convincing; the size much less so. A 2026 reanalysis that rebuilt the trial's patient-level data warns that a trial this small tends to overstate the effect it detects. It put the probability of a clinically meaningful benefit (a hazard ratio below 0.9) at only 88%.^[3]

The design limits what it can say, too. It was a single trial, open-label (everyone knew which group they were in), and lasted six months. Recurrence was picked up clinically rather than by continuous monitoring. And everyone was a current or recent coffee drinker, so it shows that telling a coffee drinker to quit doesn't help — not that someone with atrial fibrillation who doesn't drink coffee should start.

It confirmed the guidelines rather than overturning them

The idea that people with atrial fibrillation must give up caffeine is clinical folklore more than current advice. Before the trial reported, the 2023 American heart-rhythm guideline had already concluded that "recommending caffeine abstention to prevent AF episodes is of no benefit," while allowing that cutting caffeine may still ease symptoms in people who recognise it as a trigger.^[4]

Observational studies agree. A meta-analysis of roughly 116,000 people found caffeine was not linked to atrial fibrillation, with a modest reduction in the higher-quality studies,^[5] and a 2026 pool of nine cohorts and 704,121 adults reached the same verdict at every intake from one cup a day to five or more.^[6]

The other trial: extra beats and lost sleep

A second trial, led by DECAF's senior researcher, is the counterweight. One hundred healthy adults wore continuous heart monitors and, on randomly assigned days over two weeks, either drank caffeinated coffee or avoided caffeine. Extra beats from the upper chambers of the heart — the kind that tend to precede atrial fibrillation — were essentially unchanged, about 58 against 53 a day. But extra beats from the lower chambers (ventricles) rose by about half, people slept about 36 minutes less on coffee days, and they walked about 1,000 more steps.^[7]

The ventricular finding deserves caution: it was a secondary result, the investigators say its confidence interval should not be used to accept or reject an effect, and it was concentrated above one cup a day, with no relationship seen in people drinking a single cup. Together the two trials suggest coffee doesn't drive atrial rhythm problems, may provoke extra ventricular beats, and does cost sleep.

What to do

If you have atrial fibrillation and drink coffee, you don't need to stop to protect your rhythm. Cut back if you notice caffeine sets off your symptoms.
If you have hard-to-control arrhythmias, or an extra electrical pathway in the heart (Wolff-Parkinson-White syndrome), get individual advice: DECAF covered atrial fibrillation and flutter only.
Don't take up coffee to prevent atrial fibrillation. The trial tested quitting, not starting.

For the rest of the coffee evidence, see Coffee; for the stimulant itself, Caffeine.