Does coffee cause anxiety?

Caffeine does, and how much depends on the dose. A 2024 meta-analysis of 8 studies in about 546 adults, seven of them randomized, found caffeine raised anxiety scores by a large standardized amount (SMD 0.94, 95% CI 0.28 to 1.60), though the studies disagreed sharply with each other and mostly tested caffeine doses rather than coffee as people drink it. Below 400 mg the rise was moderate. At 400 mg or more it was large. The people most affected have panic disorder: about half of them had a panic attack after a big dose. Everyday coffee drinking looks different: in UK Biobank the lowest anxiety risk was at about 2 to 3 cups a day, and an Iranian study found no link at all.

Established. In blinded placebo-controlled challenges, 51.1 percent of 237 people with panic disorder had a panic attack after 400 to 750 mg of caffeine and none did after placebo. Among healthy controls in the same kind of test, 1.7 percent did. That is a meta-analysis of controlled studies, and it is the firmest finding on this page.

What the trials found

The largest body of trials is a 2026 systematic review of 27 randomized trials in 7,015 healthy people. Doses of 200 mg of caffeine or more consistently raised anxiety symptoms, and the rise was bigger in people who do not usually consume caffeine. The authors could not pool the trials into one number because their methods differed too much. The review gives the direction and the threshold, and leaves the size open.

The 2024 meta-analysis is the one that does give a size. Its subgroups split at 400 mg: an SMD of 0.61 below that line (95% CI 0.42 to 0.79) and 2.86 at or above it (95% CI 2.50 to 3.22). The authors called the first a moderate increase and the second an extremely significant one. Both subgroups were formed after studies were dropped to cut disagreement between them, which makes the exact numbers softer than they look.

The same panic disorder review concludes that caffeine at roughly the amount in 5 cups of coffee also raises anxiety in healthy adults. It could not compare caffeine against placebo for anxiety itself, because only 2 of its 10 studies reported anxiety on placebo.

Why the same cup hits people differently

In a double-blind study of 102 healthy people who rarely used caffeine, 50 mg did not raise anxiety ratings and 450 mg raised them in most participants. At 150 mg, whether anxiety rose depended on variants of the ADORA2A gene. The study's abstract does not say whether the dose order was randomized. A small 2003 study gave coffee with 3 to 300 mg of caffeine and found a dose-dependent rise in anxiety in men and none in women. It did not report how many people took part, so the sex difference is a lead to test.

What everyday coffee drinking looks like

Trials give one measured dose in a lab. Habit is a different question, and only observational data answers it. In UK Biobank, 146,566 adults reported their coffee intake between 2006 and 2010 and were screened for anxiety in 2016. The link was J-shaped, with the lowest risk at about 2 to 3 cups a day. The abstract does not report how large the higher risks at either end were, and anxiety was measured only once. In a 2026 Iranian study of 1,994 adults, drinking 1 cup a day or more was not linked to anxiety scores in any model, although only 348 people drank that much.

Unsettled. For a habitual drinker having 1 to 3 cups a day, nobody has run the trial that would settle it. The randomized evidence mostly uses caffeine capsules or drinks at fixed doses, often in people who rarely use caffeine, and the population studies cannot separate coffee from everything else about the people who drink it.

Who should be careful

People with panic disorder have the clearest reason to watch caffeine. Beyond the 51.1 percent figure, a 1985 challenge in 21 patients and 17 healthy volunteers found 71 percent of the patients said caffeine felt like their own panic attacks. That dose was 10 mg per kg of body weight, about 700 mg for someone weighing 154 lb (70 kg). Little is known about what smaller doses do in this group.

Children and teens are the second group. The FDA warns that too much caffeine in them can cause anxiety, a faster heart rate, palpitations and sleep problems, and its consumer page gives no numeric limit for this age group. Caffeine in pregnancy has its own page, on coffee and pregnancy.

Not for everyone. If you have panic disorder, the challenge studies show that high doses of caffeine can trigger panic attacks. We found no trials in other anxiety disorders. Whether cutting back helps you in particular is a question for whoever treats you.

What expert bodies say

The US Food and Drug Administration cites 400 mg of caffeine a day, about two to three 12-fluid-ounce cups of coffee, as an amount not generally associated with negative effects for most adults. That figure is not specific to anxiety disorders, and the trials above show anxiety rising from 200 mg, more so in people unused to caffeine. The European Food Safety Authority's 2015 caffeine opinion would be the other reference, and we could not retrieve its wording on anxiety.

How we searched

What would change this answer

• A pooled analysis at low risk of bias of coffee itself, at the 1 to 3 cups people usually drink, in habitual drinkers. The current trials mostly test caffeine at fixed doses.
• Trials in people with generalized anxiety disorder, who are the people most likely to ask this question. We found no registered trial.
• Smaller doses in panic disorder. The challenge studies used 400 to 750 mg, and nobody knows where below that the risk starts.

Everything else about coffee, from its caffeine content to the other evidence on health, is on the coffee page.

Sources

1. ev-cfanx-01 · Meta-analysis or systematic review · systematic review and meta-analysis (7 RCTs, 1 cross-sectional study) · n = 546
   Meta-analysis of random effects models revealed that caffeine consumption significantly increased the risk of anxiety compared with the control group [SMD = 0.94, 95%Cl = (0.28, 1.60), p < 0.05; heterogeneity: I2 = 94.7% >50%, p < 0.001] (Figure 4).
   
   Who: adults in 7 RCTs and 1 cross-sectional study, anxiety measured by BAI, POMS, STAI or SAS
   Effect: SMD 0.94 (95% CI 0.28 to 1.60); I2 = 94.7%
   Certainty: Very high heterogeneity; seven RCTs at moderate risk of bias; mostly caffeine doses, not coffee as usually drunk.
   Frontiers in Psychology, 2024, 15:1270246 (Liu et al.) · checked 2026-10-07 · we read the fulltext
2. ev-cfanx-02 · Meta-analysis or systematic review · systematic review and meta-analysis, dose subgroups (fixed effects after sensitivity exclusions) · n = 546
   The results of the subgroup analysis of the fixed effects model showed a moderate increase in anxiety scores in the test group in the low-dose subgroup [SMD = 0.61, 95%Cl = (0.42, 0.79), p < 0.05; heterogeneity: I2 = 0% <50, p > 0.1]. In the high-dose subgroup, the anxiety scores of the test group increased extremely significantly [SMD = 2.86, 95%Cl = (2.50, 3.22), p < 0.05; heterogeneity: I2 = 45.9% <50, p > 0.1].
   
   Who: adults in the included trials, split by caffeine dose below 400 mg or 400 mg and above
   Effect: low dose (<400 mg) SMD 0.61 (0.42 to 0.79), I2 0%; high dose (>=400 mg) SMD 2.86 (2.50 to 3.22), I2 45.9%
   Certainty: Subgroups built after excluding studies to remove heterogeneity; treat sizes as approximate.
   Frontiers in Psychology, 2024, 15:1270246 (Liu et al.) · checked 2026-10-07 · we read the fulltext
3. ev-cfanx-03 · Meta-analysis or systematic review · systematic review of randomized controlled trials (narrative synthesis) · n = 7015
   It was identified that doses equal to or greater than 200 mg of caffeine were consistently associated with a significant increase in anxiety symptoms. This effect was more pronounced in non‐habitual caffeine consumers.
   
   Who: healthy individuals of any age and sex in RCTs of caffeine capsules or caffeinated drinks
   Effect: anxiety rose consistently at doses of 200 mg or more; effect plateaued above 400 mg; stronger in non-habitual consumers
   Certainty: No pooled estimate; many trials at high or moderate risk of bias; dose range 12.5 to 720 mg.
   Stress and Health, 2026, 42(1):e70139 (Nascimento et al.) · checked 2026-10-07 · we read the fulltext
4. ev-cfanx-04 · Meta-analysis or systematic review · systematic review of randomized controlled trials (narrative synthesis) · n = 7015
   Furthermore, a limitation of this review is that the studies are very heterogeneous, due to methodological differences, a fact that prevented the inclusion of articles in a quantitative synthesis.
   
   Who: healthy individuals in RCTs of caffeine
   Effect: no quantitative synthesis possible; standard trial methodology needed
   Certainty: Authors' stated limitation.
   Stress and Health, 2026, 42(1):e70139 (Nascimento et al.) · checked 2026-10-07 · we read the fulltext
5. ev-cfanx-05 · Meta-analysis or systematic review · systematic review and meta-analysis of blinded placebo-controlled challenge studies · n = 237
   The 9 studies investigating panic attacks included 237 patients, of which 51.1% had a panic attack following caffeine, but none after placebo. Six of these studies compared 128 patients with 115 healthy controls (HC), finding that patients (53.9%) were more vulnerable than HC (1.7%) for panic attacks following caffeine (log RR: 3.47; 95% CI 2.06-4.87).
   
   Who: adults with panic disorder and healthy controls in blinded placebo-controlled caffeine challenge studies
   Effect: panic attacks 51.1% after caffeine vs 0% after placebo; patients 53.9% vs healthy controls 1.7%, log RR 3.47 (2.06 to 4.87)
   Certainty: Doses 400 to 750 mg, mostly 480 mg; little is known about smaller doses.
   General Hospital Psychiatry, 2022, 74:22-31 (Klevebrant and Frick) · checked 2026-10-07 · we read the abstract
6. ev-cfanx-06 · Meta-analysis or systematic review · systematic review and meta-analysis of blinded placebo-controlled challenge studies · n = 232
   The results confirm that caffeine at doses roughly equivalent to 5 cups of coffee induces panic attacks in a large proportion of PD patients and highly discriminates this population from healthy adults. Caffeine also increases anxiety in PD patients as well as among healthy adults at these doses although the exact relationship between caffeine-induced anxiety and panic attacks remains uncertain.
   
   Who: adults with panic disorder and healthy controls
   Effect: anxiety higher in patients than controls after caffeine, Hedges g 1.02 (0.09 to 1.96); doses 400 to 750 mg
   Certainty: Only 2 of 10 studies reported anxiety for placebo, so caffeine vs placebo anxiety could not be pooled.
   General Hospital Psychiatry, 2022, 74:22-31 (Klevebrant and Frick) · checked 2026-10-07 · we read the abstract
7. ev-cfanx-07 · Observational data · placebo-controlled caffeine challenge (10 mg/kg), randomization not stated · n = 38
   Seventy-one percent of the patients reported that the behavioral effects of caffeine were similar to those experienced during panic attacks.
   
   Who: 17 healthy subjects and 21 patients with agoraphobia with panic attacks or panic disorder
   Effect: 71% of patients said caffeine effects resembled their panic attacks; symptoms correlated with plasma caffeine in patients only
   Certainty: Old, small study; dose 10 mg/kg (about 700 mg for 154 lb (70 kg)).
   Arch Gen Psychiatry, 1985 · checked 2026-10-07 · we read the abstract
8. ev-cfanx-08 · Observational data · prospective cohort study · n = 146,566
   We found J-shaped associations between coffee consumption and both incident depression and anxiety, with the lowest risk of the mental disorders occurring at around 2-3 cups per day.
   
   Who: UK Biobank participants, anxiety measured in 2016 with GAD-7
   Effect: J-shaped association; lowest risk of incident anxiety at around 2-3 cups/day
   Certainty: Observational; odds ratios not in the abstract; anxiety measured once at follow-up.
   Psychiatry Research, 2023, 326:115307 (Min et al.) · checked 2026-10-07 · we read the abstract
9. ev-cfanx-09 · Observational data · cross-sectional analysis of a multicentric study · n = 1994
   No significant associations were found between coffee intake and mean anxiety scores in any of the models (p > 0.05).
   
   Who: adults in the multicentric LIPOKAP study, Iran
   Effect: no significant difference in mean anxiety scores between <1 and >=1 cup/day (p > 0.05)
   Certainty: Low coffee intake overall (348 drank 1 cup/day or more).
   Scientific Reports, 2026, 16:24883 (Mahjourian et al.) · checked 2026-10-07 · we read the fulltext
10. ev-cfanx-10 · Observational data · double-blind placebo-controlled within-subject dose challenge (0, 50, 150, 450 mg) · n = 102
    The lowest dose of caffeine did not increase ratings of anxiety while the highest dose increased anxiety in the majority of subjects.
    
    Who: healthy men and women consuming less than 300 mg caffeine per week
    Effect: ADORA2A rs5751876 associated with caffeine-induced anxiety at 150 mg; 50 mg no rise; 450 mg rise in the majority
    Certainty: Randomization of session order not stated in the abstract; full text not extracted.
    Neuropsychopharmacology, 2008, 33(12):2791-2800 (Childs et al.) · checked 2026-10-07 · we read the abstract
11. ev-cfanx-11 · Observational data · controlled dose study with coffee (3, 75, 150, 300 mg caffeine), randomization not stated · n = —
    The beverage increased state anxiety, in a dose-dependent manner, in males but not in females.
    
    Who: healthy male and female volunteers
    Effect: dose-dependent rise in STAI state anxiety in males; none in females
    Certainty: Sample size not given in the abstract; full text not in PMC.
    Human Psychopharmacology, 2003, 18(2):141-143 (Botella and Parra) · checked 2026-10-07 · we read the abstract
12. ev-cfanx-12 · Position of an expert body · government consumer guidance · n = —
    For most adults, the FDA has cited 400 milligrams* a day — that's about two to three 12-fluid-ounce cups of coffee — as an amount not generally associated with negative effects.
    
    Who: most healthy adults, US
    Effect: 400 mg/day reference amount; anxiety and jitters listed among signs of too much caffeine
    Certainty: Not specific to anxiety disorders.
    US Food and Drug Administration, Spilling the Beans: How Much Caffeine is Too Much? (consumer update) · checked 2026-10-07 · we read the section
13. ev-cfanx-13 · Position of an expert body · government consumer guidance · n = —
    Too much caffeine in children and teens can cause increased heart rate, heart palpitations, high blood pressure, anxiety, and lead to sleep problems, digestive problems and dehydration.
    
    Who: children and teens, US
    Effect: anxiety among listed effects of too much caffeine; no numeric limit for youth on this page
    Certainty: No dose threshold given for children.
    US Food and Drug Administration, Spilling the Beans: How Much Caffeine is Too Much? (consumer update) · checked 2026-10-07 · we read the section