Reference: JAMA. 2026 Jan 27;335(4):317-325
Practice Point: A cup of coffee a day appears to be safe for patients with atrial fibrillation, and it might even reduce recurrence after cardioversion.
EBM Pearl: Pragmatic trials are likely the future of clinical research, offering lower costs, shorter timelines, and greater applicability to everyday practice.
A patient comes in to see their primary care doc complaining of palpitations. Nine times out of 10, “How much coffee do you drink?” is one of the first questions asked. (True story: Once I asked that question, and the patient replied, “Six.” I said, “Wow, six cups is a lot.” She replied, “No, six pots.”) Anyway, caffeinated coffee has long been thought to be proarrhythmic, leading our left-brained predecessors to assume it should be avoided for those with heart problems. Makes sense. But since then, several studies have failed to demonstrate harm from coffee in patients with atrial fibrillation (AF), although the observational nature of those findings has kept most clinicians from giving the go-ahead when it comes to drinking caffeinated beverages. Recently, a group of investigators designed a pragmatic trial to answer the question, “Does eliminating coffee avoid fibrillation?”—the aptly named DECAF trial.
Investigators enrolled adults with AF who regularly drank coffee, had a planned cardioversion, and were willing to either continue or abstain from caffeine for 6 months after the procedure. Immediately following successful cardioversion, 200 patients were randomized 1:1 to continue consuming coffee plus other caffeinated foods or beverages as desired or to abstain completely from all caffeine for 6 months. Patients randomized to coffee consumption drank an average of about one cup per day, with negligible other sources of caffeine, which, for what it’s worth, is below the reported mean intake in the United States of 14 ounces of coffee per day.
This was a pragmatic trial, meaning patients were recruited from routine clinical care, and cardioversion and subsequent AF follow-up were planned and paid for by entities independent of the study team. The investigators contacted patients at 1, 3, and 6 months to assess self-reported caffeine intake and to determine whether AF had recurred. AF was diagnosed by the patient’s usual cardiologist and verified by investigators using either standard ECG, wearable ECG monitor (like an Apple Watch), or implantable cardiac device. A majority of patients had continuous cardiac monitoring.
In addition to cost and time savings, pragmatic trials offer greater real-world applicability and facilitate quicker translation into routine clinical practice. But that comes at the risk of sacrificing some level of internal validity because of all the added flexibility. That said, the results of this trial were a little surprising. At 6 months, AF had recurred in 47% of those in the coffee arm and 64% of those randomized to abstain (hazard ratio 0.61, 95% CI 0.42-0.89). There were also numerically fewer adverse events, including ED visits, hospitalizations, and death, in the coffee continuation group. In an effort to explain these findings away, some have looked to baseline differences, with the abstinence group having numerically more patients with coronary artery disease and history of ablation, two factors that increase risk of recurrence. But the authors did sensitivity analyses that accounted for these differences and got similar results. Others note that only 69% of those in the abstinence group actually abstained, but that can be flipped around to suggest we might actually be missing some of the magnitude of difference between the groups The authors cite coffee-induced blockade of adenosine receptors, anti-inflammatory properties of coffee, and increased energy to exercise as potential explanations for the reduction in AF recurrence in the coffee consumption group.
From a clinical standpoint, these trial results suggest pretty strongly that we should stop routinely advising patients to avoid reasonable amounts of coffee out of concern that it will trigger or worsen AF. Of course, this does not apply to patients who clearly experience AF symptoms after caffeine intake or to those drinking 6 cups (or pots) a day.
More broadly, this trial is a good reminder that what we expect based on physiology or pharmacology does not always match what happens when interventions are tested in the real world. That mismatch is exactly why hypotheses need to be tested in trials rather than assumed to be true and incorporated into practice. Pragmatic trials like this one make testing faster and more feasible by leveraging routine clinical care, requiring fewer resources, and generating answers that can be applied almost immediately. In other words, sometimes the best way to practice evidence-based medicine is to stop guessing and run the trial.
For more information, see the topic Atrial Fibrillation in DynaMed.
DynaMed EBM Focus Editorial Team
This EBM Focus was written by Katharine DeGeorge, MD, MS, Executive Editor at DynaMed and Associate Professor of Family Medicine at the University of Virginia. Edited by Alan Ehrlich, MD, FAAFP, Executive Editor at DynaMed and Associate Professor in Family Medicine at the University of Massachusetts Medical School; McKenzie Ferguson, PharmD, BCPS, Senior Clinical Writer at DynaMed; Rich Lamkin, MPH, MPAS, PA-C, Senior Clinical Writer at DynaMed; Claire Symanski, PhD, Medical Editor and Team Lead for ENT at DynaMed; Michael Butler, PhD, Medical Writer at DynaMed; Matthew Lavoie, BA, Senior Medical Copyeditor at DynaMed; Hannah Ekeh, MA, Senior Associate Editor II at DynaMed; and Jennifer Wallace, BA, Senior Associate Editor at DynaMed.