Does coffee cause anxiety?

Caffeine blocks adenosine receptors, and adenosine is an inhibitory signal — a brake. Remove the brake and neurons fire faster, cortical arousal rises, and blood vessels constrict slightly, raising blood pressure. It also activates the HPA axis, increasing cortisol and adrenaline.

That is, almost item for item, the physical signature of anxiety. Faster heart, tighter chest, heightened alertness, a body primed for threat. The physiological overlap is close enough that the line between feeling energized and feeling uneasy is genuinely thin, and which side you land on depends on dose, context and how prone you are to reading bodily signals as danger.

This is the part worth sitting with: caffeine is not producing a psychological state and then expressing it physically. It's producing the physical state directly, and your mind is left to interpret it.

Most advice on this topic is binary — caffeine and anxiety don't mix, cut it out. The research is more specific, and the specificity is useful.

The studies that established caffeine as a panic trigger used high doses. A review covering more than 235 participants found over half had a panic attack after caffeine, but every dose tested was above 400mg, and 98% of those affected had experienced a panic attack before. Nobody given a placebo had one.

At the other end, a randomized placebo-controlled crossover trial gave 150mg — about a large cup — to people with diagnosed panic disorder and to healthy controls. Contrary to the researchers' own hypothesis, it did not increase subjective anxiety in either group. It did increase physiological arousal and avoidance behaviour, so it isn't nothing. But the anxiety response they expected didn't appear. Their conclusion was that abstinence advice for panic disorder should be based on higher doses and individual assessment rather than applied as a blanket rule.

The practical translation: if you have anxiety and you drink a normal cup of coffee, the evidence doesn't clearly say you're harming yourself. If you're drinking 400mg or more — one large café order can be 310mg on its own — that's the range where the risk is documented.

This is the failure mode nobody warns you about, and it's the reason people conclude that cutting caffeine "didn't help."

Withdrawal begins within about 24 hours of stopping and can persist for up to nine days, with severity scaling to how much you were drinking. The symptoms include headache, fatigue, irritability, difficulty concentrating and low mood — and, notably, increased muscle tension and restlessness.

So someone with anxiety reads that caffeine might be the culprit, quits on a Monday, and spends the next week feeling tense, foggy and irritable. The obvious conclusion is that caffeine wasn't the problem after all. The actual explanation is that they replaced one arousal state with another and never reached a clean baseline.

A gradual taper avoids nearly all of this. It's slower and less satisfying than quitting outright, and it's the difference between a useful experiment and a wasted week.

Two genes come up repeatedly in this literature, and between them they explain a lot of why identical cups produce opposite experiences.

ADORA2A governs the adenosine A2A receptor. Specific variants are associated with markedly greater anxiety in response to a caffeine challenge — and the same variants have been linked to panic disorder itself. If you carry one, the reaction you have isn't oversensitivity in any figurative sense. It's receptor biology.

CYP1A2 governs how fast your liver clears caffeine. Slow metabolisers hold a dose longer, so the same cup reaches a higher effective concentration and stays there.

You don't need a test to act on this. If small amounts have always affected you strongly, that's the answer, and no amount of habituation will change it. The useful response is to work from a lower baseline rather than to keep trying to tolerate a standard one.

Caffeine can amplify anxiety and can trigger panic in susceptible people at high doses. What it doesn't do is create an anxiety disorder in someone who wouldn't otherwise have one.

That distinction matters because caffeine is an easy thing to blame. It's visible, it's a daily habit, and cutting it feels like taking action. But if you cut back to a modest dose, hold it for a few weeks, and the anxiety is still there — the caffeine wasn't the driver. It may have been making things somewhat worse; it wasn't the cause.

Anxiety that persists on low or no caffeine, that shows up on days you've had none, or that comes with avoidance, intrusive worry or panic attacks is worth talking to a clinician about. Anxiety disorders are among the most treatable conditions in medicine, and the treatments work considerably better than dietary adjustments do. If reading this page has been a way of avoiding that conversation, that's worth noticing.

Also worth flagging: thyroid dysfunction, blood sugar instability, some medications and sleep disorders all produce anxiety-like symptoms and are all checkable.