According to research reviewed by Gram, caffeine did not clearly change depression symptoms across 15 randomized trials with 668 participants, but anxiety scores were higher on average after caffeine. The anxiety increase appeared mainly at doses above 200 mg, and the overall evidence is rated very low in certainty.
Caffeine is one of the most popular drinks in the world, but its effects on mood have been unclear. A new review, summarized in this Gram Research analysis, combined 15 randomized trials with 668 participants to look for answers. The review found no clear change in depression symptoms overall. For anxiety, people who had caffeine reported higher anxiety scores on average than people who did not. However, the results varied widely between studies, and the overall evidence was rated very low in certainty. A separate look at doses suggested that anxiety scores rose mainly at doses above 200 mg, roughly the amount in two cups of brewed coffee. These findings are exploratory, so they should not be treated as firm rules. Researchers still need larger, longer studies before drawing clear conclusions about caffeine and mood.
Key Statistics
A 2026 systematic review and meta-analysis of 15 randomized controlled trials with 668 participants found no clear overall difference in depression symptoms between caffeine and control conditions.
Across those trials, anxiety scores were higher on average after caffeine, with a standardized difference of 0.51 (95% CI 0.12 to 0.89), although the evidence was rated very low in certainty.
In an exploratory analysis, caffeine doses above 200 mg were associated with higher anxiety scores (standardized difference of 0.97), while doses of 200 mg or less showed no clear difference.
The Quick Take
- What they studied: Does caffeine change depression or anxiety symptoms? The researchers looked only at controlled experiments where some people got caffeine and others got a placebo or no caffeine.
- Who participated: The review combined 15 randomized trials that included 668 participants in total. Participants included people with and without mental health conditions, and the studies used a range of caffeine amounts.
- Key finding: Caffeine did not show a clear effect on depression symptoms overall. Anxiety scores were higher on average after caffeine (standardized difference of 0.51, 95% CI 0.12 to 0.89), but the results differed a lot between studies and the evidence was very low in certainty.
- What it means for you: If you notice more nervousness after strong caffeine doses, the research gives some support to paying attention to that pattern. Because the evidence is limited, it does not prove caffeine causes anxiety for everyone, and people with anxiety or depression should talk with a doctor about their caffeine use.
The Research Details
This review is a systematic review and meta-analysis. In plain terms, the researchers searched many medical and scientific databases, including PubMed, Embase, and Chinese databases, for every randomized controlled trial that tested caffeine and measured mood. A randomized controlled trial is an experiment where people are randomly assigned to receive caffeine or a comparison treatment, which helps reduce bias. The search ran from the start of each database through May 16, 2026, and the review was registered in advance in PROSPERO (CRD42024564399), which means the plan was public before the analysis began.
To combine the results, the team used a statistical method that pools findings from separate studies. Mood scores were converted into a common scale called a standardized mean difference, which allows studies using different questionnaires to be compared. Because the studies differed in people, doses, and methods, the team used a random-effects model, which assumes that the true effect may vary from study to study. They also calculated a prediction interval, which estimates the range of results a future study might find.
The researchers also grouped trials in several ways, such as by psychiatric status, whether caffeine was taken once or over time, and by dose. The dose and subgroup analyses were labeled exploratory, meaning they were used to generate ideas for future research rather than to prove a cause.
A meta-analysis of randomized trials is one of the strongest ways to answer a question like this, because it combines many controlled experiments instead of relying on one small study. Randomization helps make sure the caffeine group and the comparison group start out similar, so any difference in mood is more likely to come from caffeine itself. Pooling results also makes it possible to see patterns that single studies might miss.
The review followed the PRISMA 2020 reporting guidelines, a widely used checklist for systematic reviews, and it was pre-registered. These steps make the work more transparent. However, the studies were quite different from one another, and the statistical measures showed very high variation between trials (I squared of about 90% or more). The authors rated the overall certainty of the evidence as very low for both outcomes, and they found signs of small-study effects for anxiety, which can mean that smaller studies with positive results are more likely to be published. Readers should treat these findings as early and tentative.
What the Results Show
For depression, the overall analysis did not show a clear difference between caffeine and control conditions. The average effect was a standardized difference of -0.49, with a 95% confidence interval of -1.33 to 0.34. Because this interval includes zero, the data are consistent with no effect, a small benefit, or a small harm. The prediction interval was even wider, from -2.87 to 1.88, which suggests that the effect of caffeine on depression could differ a lot from one setting to another.
For anxiety, the average result pointed in a consistent direction. People who had caffeine had higher anxiety scores on average, with a standardized difference of 0.51 (95% CI 0.12 to 0.89). However, the prediction interval (-0.95 to 1.96) crossed zero, so a new study could plausibly find no effect or even the opposite. This means the anxiety finding is suggestive but not settled.
When the researchers looked only at acute, single-dose studies in people without psychiatric conditions, the results were similar. The depression estimate was -0.15 (95% CI -2.12 to 1.81), and the anxiety estimate was 0.38 (95% CI -0.02 to 0.79). The anxiety result came close to statistical significance but still included zero at its lower edge.
In the exploratory dose analysis, trials using more than 200 mg of caffeine showed higher anxiety scores (standardized difference of 0.97, 95% CI 0.55 to 1.39). Trials using 200 mg or less showed essentially no difference (standardized difference of -0.05, 95% CI -0.41 to 0.30). The authors stress that this pattern does not establish a safe dose or a clear cutoff point.
The review found that the strength of the evidence was limited across the board. Very low certainty ratings mean that future research is likely to change the estimates. The signs of small-study effects for anxiety suggest that some published positive results may be exaggerated. The authors also emphasized that the findings do not show a steady dose-response relationship, meaning anxiety did not clearly rise in a straight line as caffeine amounts increased. For practical purposes, the most useful message is that caffeine’s effects on mood appear to depend on the person, the dose, and the setting.
Earlier observational studies, which track what people drink in daily life, have sometimes linked heavy coffee intake with lower depression risk and have linked very high caffeine intake with more anxiety or jitteriness. This review adds an important point: when researchers test caffeine directly in controlled experiments, the picture is less clear and more variable. The result fits with the broader idea that caffeine can increase feelings of nervousness in some people, especially at higher doses, while showing no consistent link to depression in short-term trials.
The biggest limitation is the very high variation between studies, which makes it hard to know what the true average effect is. The total number of participants (668) is modest, and the trials were short, so they cannot tell us what happens to mood over months or years of regular caffeine use. Many studies used different caffeine sources, such as coffee, tablets, or energy drinks, and different mood questionnaires, which adds noise. The dose analysis was exploratory and based on relatively few trials. Finally, there were signs of publication bias for anxiety, and the overall certainty of evidence was rated very low.
The Bottom Line
Based on current evidence, there is no strong reason to believe that moderate caffeine intake harms or helps depression in the short term; confidence in this is low because the data are sparse and inconsistent. There is a suggestion that caffeine doses above 200 mg may increase anxiety in some people; confidence in this is low to very low, so it should be treated as a possible pattern to watch for rather than a settled fact. People who notice jitteriness or worry after caffeine should consider cutting back and seeing how they feel.
People who feel nervous, restless, or anxious after caffeine, people who take high-dose caffeine products such as some energy drinks or supplements, and people managing anxiety or depression should pay attention to these findings. People who drink moderate amounts of coffee or tea and feel fine are unlikely to need to change anything based on this review. Pregnant people, people with heart conditions, and people taking medications that interact with caffeine should follow their doctor’s guidance rather than this summary.
If you decide to cut back on caffeine to see whether your anxiety improves, many people notice changes within one to two weeks, though caffeine withdrawal headaches or fatigue can appear in the first few days. Measurable changes in mood symptoms may take several weeks to show up clearly. Because this review found limited evidence on long-term effects, expectations should stay realistic, and any changes should be tracked over at least a few weeks.
Frequently Asked Questions
Does caffeine make anxiety worse?
Caffeine may increase anxiety on average. In a 2026 review of 15 randomized trials with 668 participants, anxiety scores were higher after caffeine, but results varied widely and the evidence is very low in certainty. Sensitivity differs from person to person.
Does caffeine affect depression symptoms?
Short-term controlled trials did not show a clear effect of caffeine on depression symptoms overall. The pooled estimate had a wide range that included no effect, so researchers cannot rule out small benefits or harms. More long-term research is needed.
How much caffeine is too much for anxiety?
An exploratory analysis found higher anxiety scores at doses above 200 mg, while doses of 200 mg or less showed no clear difference. This is not a established safe limit, and individual tolerance varies, so people with anxiety should consult a doctor.
Should I quit coffee if I feel anxious?
You do not need to quit entirely based on this research. Tracking your daily caffeine and anxiety levels for two to four weeks can show whether the link applies to you. Cutting back gradually and talking with a healthcare provider is a reasonable next step.
Want to Apply This Research?
- Log your total daily caffeine in milligrams, including coffee, tea, energy drinks, and supplements, and rate your anxiety on a simple 0 to 10 scale at the same time each day. Flag any day when caffeine exceeds 200 mg so you can compare anxiety ratings on high-caffeine and low-caffeine days.
- Try a two-week experiment where you keep caffeine at or below 200 mg per day, then compare your anxiety ratings with your baseline. Shift larger caffeine doses earlier in the day and avoid them late in the afternoon to protect sleep, which itself affects mood.
- Review your caffeine and anxiety logs monthly to look for patterns across weeks rather than single days. Share the trend summary with a healthcare provider if anxiety symptoms are persistent, worsening, or interfering with daily life, and treat app data as a conversation starter rather than a diagnosis.
This article summarizes a systematic review and meta-analysis for general education and is not medical advice. The review found very low certainty evidence and exploratory dose findings, so results may not apply to you. Caffeine affects people differently, and anyone with anxiety, depression, heart problems, pregnancy, or who takes medication should talk with a qualified healthcare professional before changing caffeine intake.
This research translation is published by Gram Research, the science division of Gram, an AI-powered nutrition tracking app.