Wellness Goals · Mental health condition

Anxiety Disorders: Symptoms, Evaluation and Care

Q: What is the difference between normal worry and an anxiety disorder?

NIMH says feeling anxious is a normal part of life. With an anxiety disorder the anxiety does not go away, is felt in many situations and can get worse over time. If it starts to interfere with school, work or relationships, it is time to seek help.

Q: What types of anxiety disorders are there?

NIMH lists generalized anxiety disorder, panic disorder, social anxiety disorder and phobia-related disorders. WHO also lists agoraphobia, separation anxiety disorder, specific phobias and selective mutism. People can have more than one.

Q: How are anxiety disorders evaluated?

A provider asks when symptoms began, how long they have lasted, how often they occur and how they affect your life, and may do a physical exam to rule out an unrelated physical cause. For generalized anxiety disorder NIMH says worry must be hard to control on most days for at least 6 months, with at least three of six listed symptoms.

Q: What does established care look like?

Talking therapy, medicines, or both. CBT is described by NIMH as the gold standard talking therapy. SSRI and SNRI antidepressants are among the medicines named. The NHS says a GP will usually advise talking therapies before medicines. A clinician can explain choices and side effects.

Q: Do exercise, mindfulness or yoga help anxiety?

Reviews found a medium drop in anxiety with exercise, and one large trial found an 8-week mindfulness course as good as an antidepressant in adults with anxiety disorders, though it had no placebo group. Kundalini yoga helped but not as much as CBT. They can support care, not replace it.

Q: Do lavender, chamomile or ashwagandha help?

A few small trials of lavender oil capsules and chamomile extract found lower anxiety than placebo, and a review of ashwagandha trials found lower anxiety with low certainty. NCCIH says conclusions are not firm, and products differ. Talk to a doctor or pharmacist first, especially if you take medicines or are pregnant.

Q: Is kava safe for anxiety?

Not automatically. The largest trial in generalized anxiety disorder found no benefit over placebo, and NCCIH says kava has been linked to rare but serious liver injury, and should not be combined with alcohol or sedatives. Ask a doctor before considering it.

Q: Does caffeine make anxiety worse?

In people with panic disorder, large amounts of caffeine brought on panic attacks in about half of those tested and almost never in healthy adults. The NHS and NIMH advise cutting back on caffeine. Smaller everyday amounts were not studied in that review.

Q: Can I stop my medicine once I feel better?

Do not stop or change a prescribed medicine on your own. Some are linked to withdrawal effects when stopped, so talk with the prescriber about any change. A herb or supplement should never replace it.

Anxiety disorders (generalized anxiety disorder, panic disorder, social anxiety disorder and phobia-related disorders) · also called Anxiety, chronic worry, nerves, GAD, panic disorder, social anxiety, trastornos de ansiedad, ansiedad

Anxiety disorders are common mental health conditions in which fear or worry does not go away, is felt in many situations and can interfere with daily life. They differ from everyday worry in how long they last and how much they get in the way. A clinician evaluates symptoms and history, and established care is talking therapy such as CBT, medicines, or both. Exercise, mindfulness and some herbs have been studied as support, but none replaces care, and kava and some other herbs carry real safety concerns.

General information, not medical advice. Last substantive update October 31, 2026 · Not medically reviewed. A named clinician has not signed off on this page. How this page is made

About Anxiety Disorders: Symptoms, Evaluation and Care

Anxiety disorders are common conditions in which fear or worry does not go away, is felt in many situations and can get worse over time. They differ from everyday worry in how long they last and how much they interfere with life. Established care is talking therapy such as CBT, medicines, or both, and complementary approaches have been studied as support, not as a replacement.

What anxiety disorders are, and how they differ from everyday worry

NIMH says feeling anxious is a normal part of life. Many people worry about health, money, school, work or family. Anxiety disorders involve more than occasional worry or fear: for people with these disorders, anxiety does not go away, is felt in many situations and can get worse over time.

NIMH describes stress as a response to an outside cause, and anxiety as the body's reaction to stress that can occur even when there is no current threat. Occasional stress and anxiety are normal, but they should not control your life. If anxiety does not go away and interferes with daily activities, it may be an anxiety disorder.

Anxiety disorders are common. NIMH says about a third of US adolescents and adults experience one at some point in their lives, and WHO estimates 470 million people worldwide had one in 2023, with only about 1 in 4 people in need receiving treatment. This page is for general information and cannot diagnose you.

^{31 32 33 23}

Symptoms and types of anxiety disorders

NIMH and WHO name several types: generalized anxiety disorder (persistent, excessive worry about everyday matters), panic disorder, social anxiety disorder and phobia-related disorders. WHO also lists agoraphobia, separation anxiety disorder, specific phobias and selective mutism. People can have more than one at the same time.

For generalized anxiety disorder, NIMH lists excessive worry about everyday things, trouble controlling worry, feeling on edge or irritable, restlessness, trouble concentrating, sleep problems or tiredness, pain such as headaches, muscle aches or stomach aches, trembling, sweating, and trouble swallowing. The NHS adds palpitations, dizziness and low mood. NHS says a panic attack involves sudden intense fear with symptoms such as a racing heartbeat and shortness of breath, usually lasts 5 to 20 minutes, and is frightening but not dangerous.

These symptoms overlap with ordinary stress and with physical problems, so a list cannot tell you whether you have a disorder. What matters is how long it lasts and how much it interferes with your life.

^{31 32 28 27 33}

How anxiety disorders are evaluated

NIMH says a provider, such as a primary care doctor, psychiatrist, psychologist or clinical social worker, asks when symptoms began, how long they have lasted, how often they occur and how they affect your life. A physical exam may be done to make sure an unrelated physical problem is not causing the symptoms.

For generalized anxiety disorder, NIMH says diagnosis needs worry that is hard to control on most days for at least 6 months, plus at least three of these: feeling restless or on edge, fatigue, trouble concentrating, irritability, muscle tension or sleep problems. The NHS describes similar signs, including worry about a wide range of things for at least 6 months that affects daily life. NICE advises a comprehensive assessment that looks at distress and daily impairment, not only symptom counts, and at other conditions, substance misuse and past treatments.

NIMH says anxiety disorders often occur together with depression, other anxiety disorders, post-traumatic stress disorder, chronic pain, heart conditions, and with alcohol and substance misuse. These need to be treated as a whole.

^{32 28 30}

What established care looks like

NIMH says treatment typically involves psychotherapy, medication or both, and that finding the right plan may take trial and error. Cognitive behavioural therapy (CBT) is described by NIMH as well studied and the gold standard choice for psychotherapy. Acceptance and commitment therapy is another option for generalized anxiety disorder. WHO says the psychological treatments with the most evidence are CBT-based, including exposure therapy. A 2018 analysis of 41 placebo-controlled trials found moderate effects of CBT across anxiety-related disorders and large effects in generalized anxiety disorder.

Medicines named by NIMH include SSRI and SNRI antidepressants, which may take several weeks to work, benzodiazepines, which are used only briefly because of tolerance and dependence, and buspirone. NICE advises starting with the least intrusive, most effective option and, if drug treatment is chosen, an SSRI first. The NHS says a GP will usually advise talking therapies before medicines and that adults can refer themselves to NHS talking therapies. A 2019 network analysis found several medicines across classes were better than placebo, and that some were poorly tolerated.

This page does not recommend any medicine or amount. A clinician can explain choices, side effects and what to expect.

^{32 28 30 33 5 19}

What complementary approaches have and have not shown

NIMH and the NHS both say healthy habits can help alongside care: regular exercise, enough sleep, cutting back on caffeine and not using alcohol, cigarettes, gambling or drugs to relieve anxiety. NIMH says stress management such as exercise, mindfulness and meditation can reduce anxiety symptoms and enhance the effects of psychotherapy. Reviews found a medium effect of exercise on anxiety, and one large trial found an 8-week mindfulness course as good as an antidepressant, though without a placebo group.

Kundalini yoga helped generalized anxiety disorder more than stress education but not as much as CBT. Relaxation training lowered anxiety in a review of 27 studies, although CBT may do better for some disorders. In people with insomnia, online CBT for insomnia gave a small drop in anxiety.

For herbs and supplements the picture is weaker. Lavender oil capsules, chamomile extract and omega-3 supplements each have a few small positive trials. Kava and ashwagandha have mixed or low-certainty results and have been linked to liver injury. L-theanine did not help in one small trial, magnesium evidence is poor, and acupuncture, valerian and passionflower are supported only by limited evidence according to NCCIH. See the Benefits and Safety tabs.

^{32 28 20 18 8 17 12 11 6 21 16 1 15 4 23}

How strong is the research?

Most herb and supplement trials read were small and short. Mind-body trials cannot hide the treatment from participants. Some studies compare an approach with no treatment, which is an easy bar to beat, and the trial of mindfulness against an antidepressant had no placebo arm.

NCCIH says psychological and physical approaches generally have good safety records but are not risk free, and that dietary supplements can have side effects and interactions. It also reminds readers that natural does not always mean safe, and that a product labelled standardised is not guaranteed to be consistent. Supplement trials usually test one named product, so results do not carry over to other brands.

^{23 8 6 1}

At a glance

Traditional uses history, not proof

Studied for research

Regular exercise, moderate certainty
Mindfulness-based stress reduction, moderate certainty
Kundalini yoga, mixed
Relaxation training, low certainty
Online CBT for insomnia, low certainty for anxiety
Cutting back on caffeine, low certainty
Lavender oil capsules, low certainty
Chamomile extract, low certainty
Kava, mixed, with liver safety concerns
Ashwagandha, low certainty
Omega-3 fatty acids, low certainty
L-theanine, insufficient
Magnesium, very low certainty

Main compounds

Exercise
Mindfulness
Yoga
Relaxation training
Sleep and caffeine habits
Lavender oil
Chamomile
Kava
Ashwagandha
Omega-3
L-theanine
Magnesium

Common forms

Group courses and classes
Online programmes
Standardised capsule and tablet extracts
Lifestyle changes

Research links: NIMH: Generalized Anxiety Disorder: What You Need to Know · NHS: Generalised anxiety disorder (GAD) · NICE guideline CG113: generalised anxiety disorder and panic disorder in adults · WHO: Anxiety disorders fact sheet · NCCIH: Anxiety and Complementary Health Approaches · 2023 trial: MBSR versus escitalopram for anxiety disorders · 2018 analysis: CBT for anxiety and related disorders (41 trials) · Research Library · Genes & Biomarkers Library

What it is and its history

Accepted scientific name: Anxiety disorders (a group of mental health conditions)
Common names: Anxiety disorders (English); Anxiety, chronic worry, nerves (English); Trastornos de ansiedad (Spanish); Ansiedad, angustia (Spanish)
Look-alikes and adulterants: Name: Everyday stress or worry; Note: NIMH says stress is a response to an external cause, while anxiety can occur even without a current threat; occasional anxiety is normal, but it should not control your life.; Name: Depression; Note: The two often occur together. See the stress, anxiety, burnout and depression guide.; Name: Physical problems with similar symptoms; Note: NIMH says a physical exam may be done to make sure an unrelated physical problem is not causing the symptoms.

What is inside, and how it may work

What it contains and how those compounds act in cells and animals. Laboratory and animal mechanisms are not proven effects in people.

Main compounds

Compound	Family	Activity studied	Level of evidence
Cognitive behavioural therapy (CBT)	psychological therapy (established care)	First-line talking therapy for anxiety disorders. A 2018 analysis of 41 placebo-controlled trials (2,843 people) found moderate effects of CBT on the main symptoms across anxiety-related disorders, with large effects for generalized anxiety disorder. ^{5 32 30 33}	Human data
Acceptance and commitment therapy (ACT)	psychological therapy (established care)	NIMH describes it as another option for GAD with a growing body of supporting research. ³²	Agency summary
SSRI and SNRI antidepressants	medicines (established care, for a clinician to discuss)	Commonly used for anxiety disorders and may take several weeks to work (NIMH). A 2019 network analysis of 89 trials found that several medicines across drug classes were more effective than placebo with relatively good tolerability, while some were poorly tolerated. ^{32 30 19}	Human data
Benzodiazepines and buspirone	medicines (established care, for a clinician to discuss)	NIMH says benzodiazepines can reduce anxiety rapidly but cause tolerance and dependence in some people, so they are prescribed only briefly; buspirone is not a sedative and has less potential to be addictive. NICE and WHO also limit benzodiazepines. ^{32 30 33}	Agency summary
Exercise, sleep, caffeine and alcohol habits	lifestyle	Studied as support alongside care. ^{20 18 11 10}	Human data
Mindfulness, yoga and relaxation	mind-body	Studied alone or alongside care. ^{8 17 12}	Human data
Lavender oil, chamomile, kava, ashwagandha	herbal extracts	Studied as capsule or tablet extracts. ^{6 13 16 1}	Human data
Omega-3, L-theanine, magnesium	supplements	Studied as supplements. ^{21 15 4}	Human data

Mechanisms, receptors and pathways

Target or pathway	Gene or receptor	What was found	Level
Why anxiety disorders develop		NIMH says the exact cause of GAD is unknown and that research suggests a mix of genetics, brain chemistry, biology and environment. External causes, such as a traumatic event or a stressful environment, may raise risk. The NHS lists other anxiety disorders, relatives with anxiety or depression, stressful or traumatic experiences, a painful long-term health condition and a history of drug or alcohol misuse as things that increase the chance of GAD. WHO adds that people who have lived through abuse, severe losses or other adverse experiences are more likely to develop an anxiety disorder. ^{32 28 33}	Agency summary
How complementary approaches might work		The reviews read measured symptoms, not mechanisms. Suggested explanations such as calmer body arousal, better sleep or changes in brain signalling are proposals, not shown. ^{18 6}	Not established

Isolated compounds versus whole extracts

The herb trials used specific capsule or tablet extracts (a standardised lavender oil product, a chamomile extract, a water-based kava root extract). Tea, essential oil on skin or a different brand cannot be assumed to act the same.

16benefits and uses looked at

33studies and sources checked

Moderate certaintybest evidence rating found

3lab markers tracked in trials

2 moderate certainty2 mixed findings8 low certainty1 very low certainty3 not enough evidence

What was looked at

The 16 benefits and uses checked for Anxiety Disorders: Symptoms, Evaluation and Care, each with how strong the evidence is. Tap one to open it.

Regular exercise and physical activityModerate certainty
Mindfulness-based stress reduction (MBSR) and mindfulness programmesModerate certainty
Yoga (Kundalini yoga in the main trial)Mixed findings
Relaxation training (progressive relaxation, autogenic training, applied relaxation)Low certainty
Better sleep (digital cognitive behavioural therapy for insomnia)Low certainty
Cutting back on caffeineLow certainty
Cutting back on alcoholNot enough evidence
Lavender oil capsules (the Silexan preparation) and lavender aromatherapyLow certainty
Chamomile extract (supplement)Low certainty
Kava (Piper methysticum)Mixed findings
Ashwagandha (Withania somnifera) root extractLow certainty
Omega-3 fatty acids (fish oil type supplements)Low certainty
L-theanine (an amino acid from tea)Not enough evidence
Magnesium supplementsVery low certainty
AcupunctureLow certainty
Valerian and passionflowerNot enough evidence

Lab markers tracked in the studies

Anxiety symptom scores (clinician-rated and self-rated)Panic attacks after a caffeine challengeHamilton Anxiety Rating Scale (HAM-A) total score

What is Anxiety Disorders: Symptoms, Evaluation and Care good for?

The complementary approaches that have been studied for anxiety, and how strong the human evidence is for each. These describe what studies found in the groups they studied. They are not advice to use any approach, they are not a replacement for care, and each result is for the product or programme tested.

Regular exercise and physical activityModerate certaintyReviews found lower anxiety than with usual care, a medium effect. The six trials in people with a diagnosed anxiety disorder were small. Mindfulness-based stress reduction (MBSR)Moderate certaintyAn 8-week mindfulness course was as good as an anxiety medicine in a trial of 276 adults with anxiety disorders, but there was no placebo group. Yoga (Kundalini yoga)Mixed findingsBetter than stress education but not as good as CBT in generalized anxiety disorder; NCCIH calls the wider evidence inconclusive. Relaxation trainingLow certaintyA review of 27 studies found a medium drop in anxiety, but included studies without control groups; CBT may work better for some disorders. Better sleep (online CBT for insomnia)Low certaintyA small drop in anxiety symptoms in adults with insomnia, not in people with anxiety disorders. Cutting back on caffeineLow certaintyLarge caffeine amounts brought on panic attacks in about half of people with panic disorder. Cutting back as a treatment was not tested in the studies read. Cutting back on alcoholNot enough evidenceAgencies advise against alcohol as a way to cope with anxiety. No trial was read. Lavender oil capsules (Silexan)Low certaintySeveral trials of one oral product found lower anxiety than placebo; NCCIH says small studies and limited independent funding prevent firm conclusions. Chamomile extractLow certaintyTwo small trials in generalized anxiety disorder were encouraging, but the main outcome of the longer one was not significant. KavaMixed findingsSmall short trials leaned positive, the largest trial found no benefit in generalized anxiety disorder, and kava has been linked to serious liver injury. AshwagandhaLow certaintyA meta-analysis of 12 trials found lower anxiety and stress, but with very inconsistent results and low certainty. Rare liver injury has been reported. Omega-3 fatty acidsLow certaintyA review of 19 trials found a small drop in anxiety symptoms, mainly in people with a clinical diagnosis and at higher daily amounts. L-theanineNot enough evidenceOne small trial added to antidepressants found no benefit for anxiety. MagnesiumVery low certaintySome small trials suggest less anxiety in people prone to it, but the reviewers call the evidence poor. AcupunctureLow certaintyNCCIH says there is limited evidence, and overall study quality is low. Valerian and passionflowerNot enough evidenceNCCIH says there is not enough evidence for valerian, and only a little research for passionflower.

Benefits, one by one

Each card is a research question, not a promise. Tap a goal to filter. Open a card to see the studies, doses and limits.

Regular exercise and physical activity

Moderate certainty

Reviews of trials found lower anxiety symptoms with exercise than with usual care or no exercise, a medium-sized effect. Most of the reviews were of low quality, and only six small trials looked at people with a diagnosed anxiety disorder.

5 studiesThe trials used many kinds of exercise and programmes. The NHS names …

Studies, doses and limits

What the trials found

A 2017 meta-analysis of 6 randomised trials (262 adults) in people with an anxiety or stress-related disorder, and a 2023 overview of 97 reviews (1,039 trials, 128,119 adults) covering healthy people, people with mental health conditions and people with long-term illness.

Results

In the 2017 analysis exercise lowered anxiety symptoms more than control conditions, a moderate effect. In the 2023 overview physical activity had a medium effect on anxiety compared with usual care, and higher-intensity activity was linked to greater improvement. The NHS, NIMH and WHO pages all list regular exercise as a helpful habit.

Effect sizes as reported

2017: standardised mean difference -0.58, p = 0.02 (the confidence interval printed in the abstract does not contain the estimate, so it is not repeated). 2023: median effect size for anxiety -0.42 (interquartile range -0.66 to -0.26).

Why this rating

Compiler's rating: moderate. Two reviews agree on a medium effect and agencies recommend it as support, but the small trial set in diagnosed anxiety disorders is thin and 77 of 97 reviews in the overview scored critically low on quality.

Form and dose studied

The trials used many kinds of exercise and programmes. The NHS names running, walking, swimming and yoga as examples. These describe what was studied, not what anyone should do.

Who was studied

Adults with anxiety or stress-related disorders (2017); adults in general, with and without health conditions (2023).

Limits

Few trials in people with diagnosed anxiety disorders, people knew they were exercising, symptoms were self-reported in many trials, and the benefit shrank in longer programmes in the overview.

Why researchers looked

Exercise is free, widely available and good for general health, so researchers asked whether it also lowers anxiety.

Proposed mechanism Hypothesis

Proposed: exercise may change stress hormones, sleep and mood. The reviews read measured symptoms, not these mechanisms.

A proposed mechanism does not show that the effect happens in people.

Exercise carries a risk of injury and is not right for everyone. People with heart, joint or other health conditions can ask a doctor first.

^{20 18 28 32 33}

Mindfulness-based stress reduction (MBSR) and mindfulness programmes

Moderate certainty

One large trial found an 8-week mindfulness course as good as a commonly prescribed anxiety medicine in adults with anxiety disorders. Pooled studies also found lower anxiety, but many were not compared with a control group.

3 studiesThe trial used a weekly group course lasting 8 weeks, compared with a…

Studies, doses and limits

What the trials found

A 2023 randomised trial of 276 adults with a diagnosed anxiety disorder compared an 8-week MBSR course with escitalopram, an antidepressant. A 2010 meta-analysis of 39 studies (1,140 people) pooled mindfulness-based therapy across many conditions. NCCIH also lists a 2021 trial in social anxiety and a 2021 review of 23 studies in anxiety disorders, which were not opened here.

Results

In the 2023 trial the main anxiety score fell by 1.35 points with MBSR and 1.43 with escitalopram, and MBSR was judged noninferior. Study-related side effects were reported by 15.4 percent on MBSR and 78.6 percent on escitalopram. In the 2010 analysis anxiety scores fell from before to after treatment (Hedges's g 0.63 overall, 0.97 in people with anxiety and mood disorders) and stayed lower at follow-up.

Effect sizes as reported

2023 trial: group difference -0.07 (95% CI -0.38 to 0.23) against a noninferiority margin of -0.495. 2010 analysis: g 0.63 overall and 0.97 in anxiety and mood disorders.

Why this rating

Compiler's rating: moderate. A well-run randomised trial with a blinded rater supports it, and NCCIH describes the evidence as promising. The trial had no placebo or waiting-list arm, one in four did not finish, and the 2010 effect sizes are before-and-after changes, not comparisons with a control group.

Form and dose studied

The trial used a weekly group course lasting 8 weeks, compared with a medicine taken daily. These describe what was studied, not what anyone should do.

Who was studied

Adults with a diagnosed anxiety disorder, mean age 33, 75 percent women (2023); varied clinical groups (2010).

Limits

No placebo arm, people knew which treatment they had, and the 2010 analysis included uncontrolled studies. Long-term follow-up is limited.

Why researchers looked

Mindfulness teaches a way of noticing worry without being carried away by it, so researchers tested whether that lowers anxiety.

Proposed mechanism Hypothesis

Proposed: less reacting to worrying thoughts and a calmer body stress response. The studies read measured symptoms, not these mechanisms.

A proposed mechanism does not show that the effect happens in people.

In the 2023 trial no one on MBSR left because of side effects. Some people have difficult experiences with meditation; see the Safety tab.

^{8 7 23}

Yoga (Kundalini yoga in the main trial)

Mixed findings

In one large trial for generalized anxiety disorder Kundalini yoga worked better than stress education, but not as well as cognitive behavioural therapy (CBT). NCCIH says the evidence for diagnosed anxiety disorders is inconclusive.

2 studiesTwelve weekly group sessions of 120 minutes plus 20 minutes of daily …

Studies, doses and limits

What the trials found

A 2021 randomised trial of 226 adults with generalized anxiety disorder (3 groups: Kundalini yoga, CBT or stress education). NCCIH also cites a 2019 review of 38 studies of yoga for anxiety symptoms and a 2021 review in people with anxiety disorders, which were not opened here.

Results

After 12 weeks 54.2 percent responded on yoga and 70.8 percent on CBT, compared with 33.0 percent on stress education. Yoga was not shown to be as effective as CBT. NCCIH says that although some reviews found yoga helpful for anxiety symptoms, evidence is insufficient to recommend routine yoga as the main treatment for anxiety disorders.

Effect sizes as reported

Response yoga vs stress education: odds ratio 2.46 (95% CI 1.12 to 5.42). CBT vs stress education: odds ratio 5.00 (95% CI 2.12 to 11.82). Yoga did not meet the test for being as good as CBT (difference 16.6 percent, p = 0.42 for noninferiority).

Why this rating

Compiler's rating: mixed. The trial found a benefit over stress education but not equal to CBT, about one in three did not complete it, and NCCIH calls the wider evidence inconclusive.

Form and dose studied

Twelve weekly group sessions of 120 minutes plus 20 minutes of daily homework, led by two instructors. These describe what was studied, not what anyone should do.

Who was studied

Adults with generalized anxiety disorder, mean age 33, about 70 percent women.

Limits

Stress education is a weak comparison, the trial was single-blind, and 31 percent did not complete it. One style of yoga was tested.

Why researchers looked

Yoga combines gentle movement, breathing and relaxation, so researchers tested it for anxiety and compared it with a first-line treatment.

Proposed mechanism Hypothesis

Proposed: relaxation, breathing control and gentle activity. Not tested in the trial read.

A proposed mechanism does not show that the effect happens in people.

As with any physical activity, yoga can cause injuries, most commonly sprains and strains, according to NCCIH.

^{17 23}

Relaxation training (progressive relaxation, autogenic training, applied relaxation)

Low certainty

A review of 27 studies found relaxation training lowered anxiety scores by a medium amount. NCCIH says CBT may work better for some anxiety disorders, and the review included studies without a control group.

3 studiesMethods included Jacobson's progressive relaxation, autogenic trainin…

Studies, doses and limits

What the trials found

A 2008 review of 27 studies (1997 to 2007) of relaxation methods, including randomised trials, observational studies and studies without a control group. NCCIH cites a 2018 review of 50 studies in which relaxation was less effective than CBT for post-traumatic stress and obsessive-compulsive disorders; that review was not opened here.

Results

Anxiety fell by a medium amount (Cohen's d 0.57 within groups and 0.51 between groups). Benefit was larger for meditation, for volunteers and for longer treatments. WHO lists relaxation skills among stress-management skills that can help reduce anxiety symptoms.

Effect sizes as reported

d 0.57 (95% CI 0.52 to 0.68) within groups; d 0.51 (95% CI 0.46 to 0.634) between groups.

Why this rating

Compiler's rating: low. The review pooled randomised and non-randomised studies from 1997 to 2007, benefits were bigger in volunteers than in patients, and NCCIH says CBT may be more helpful for at least some anxiety disorders.

Form and dose studied

Methods included Jacobson's progressive relaxation, autogenic training, applied relaxation and meditation. These describe what was studied, not what anyone should do.

Who was studied

People with anxiety problems and disorders, and volunteers (2008 review).

Limits

Mixed study designs, older searches, and no way to separate the effect of the method from attention and expectation.

Why researchers looked

Relaxation skills are simple and low-risk, so they are among the oldest non-drug approaches tested for anxiety.

Proposed mechanism Hypothesis

Proposed: lowering muscle tension and arousal. Not tested in the review read.

A proposed mechanism does not show that the effect happens in people.

NCCIH says relaxation techniques are generally safe, but some people report more anxiety, intrusive thoughts or fear of losing control, and there are rare reports of problems in people with epilepsy, some psychiatric conditions or a history of trauma.

^{12 23 33}

Better sleep (digital cognitive behavioural therapy for insomnia)

Low certainty

In adults with insomnia, online CBT for insomnia gave a small drop in anxiety symptoms as well as better sleep. These trials were in people with insomnia, not people with an anxiety disorder.

3 studiesOnline CBT for insomnia programmes, with or without a therapist. Thes…

Studies, doses and limits

What the trials found

A 2023 meta-analysis of 22 randomised trials of digital CBT for insomnia in adults with insomnia who also reported depression or anxiety measures. NIMH and the NHS also advise enough sleep.

Results

At the end of treatment the digital CBT-I group had a small drop in anxiety symptoms and a large improvement in sleep. Results were stronger when people stuck with the programme, and fully automated programmes also worked.

Effect sizes as reported

Anxiety symptoms: standardised mean difference -0.29 (95% CI -0.40 to -0.19; 18 trials). Sleep: -0.76 (95% CI -0.95 to -0.57; 22 trials).

Why this rating

Compiler's rating: low for anxiety. The effect on anxiety was small, anxiety was a secondary outcome, and the trials enrolled people with insomnia rather than people with anxiety disorders.

Form and dose studied

Online CBT for insomnia programmes, with or without a therapist. These describe what was studied, not what anyone should do.

Who was studied

Adults with insomnia.

Limits

Not a study of anxiety disorders. Programmes differ, and the abstract does not give follow-up length.

Why researchers looked

Anxiety and poor sleep feed each other. NIMH says getting enough sleep can reduce anxiety symptoms when paired with standard care.

Proposed mechanism Hypothesis

Proposed: better sleep may lower daytime worry and arousal. Not tested in the review read.

A proposed mechanism does not show that the effect happens in people.

Sleep problems can also be a sign of depression or another condition; see a clinician if they last.

^{11 32 28}

Cutting back on caffeine

Low certainty

In people with panic disorder, large caffeine doses brought on panic attacks in about half of those tested, and almost never in healthy adults. Whether cutting back treats anxiety in general has not been tested in the studies read.

3 studiesCaffeine amounts in the studies were between 400 and 750 mg in one go…

Studies, doses and limits

What the trials found

A 2022 meta-analysis of 10 blinded, placebo-controlled caffeine challenge studies in adults with panic disorder. NIMH says reducing caffeine can lower anxiety symptoms when paired with standard care, and the NHS advises against lots of coffee, tea, cola or energy drinks.

Results

In 9 studies with 237 patients, 51.1 percent had a panic attack after caffeine and none after placebo. In 6 studies comparing patients with healthy adults, panic attacks followed caffeine in 53.9 percent of patients and 1.7 percent of healthy adults. Patients were also more sensitive to caffeine's anxiety-raising effect.

Effect sizes as reported

Panic attacks, patients vs healthy adults: log risk ratio 3.47 (95% CI 2.06 to 4.87). Subjective anxiety: Hedges's g 1.02 (95% CI 0.09 to 1.96).

Why this rating

Compiler's rating: low for cutting back as a treatment. The caffeine challenge studies are blinded and consistent, but they tested one-off large amounts, roughly five cups of coffee as the authors put it. No trial of cutting down on everyday caffeine as a way to treat anxiety was read.

Form and dose studied

Caffeine amounts in the studies were between 400 and 750 mg in one go, as reported. The review could not say anything about smaller amounts. These describe what was studied, not what anyone should do.

Who was studied

Adults with panic disorder, and healthy adult comparison groups.

Limits

Few studies, only two reported anxiety data for the placebo condition, one-off challenges rather than long-term habits, and a narrow range of amounts.

Why researchers looked

Caffeine is a stimulant, and the NHS and NIMH both advise people with anxiety to reduce it.

Proposed mechanism Hypothesis

Proposed: caffeine raises alertness and heart rate, which can feel like anxiety. Not measured in the review read.

A proposed mechanism does not show that the effect happens in people.

People with panic disorder can tell their clinician about their caffeine intake. Caffeine is also found in tea, cola, energy drinks and some medicines.

^{10 32 28}

Cutting back on alcohol

Not enough evidence

Health agencies advise against using alcohol to relieve anxiety because it can make anxiety worse and is linked to poor mental health. No trial of cutting back alcohol as an anxiety treatment was read for this page.

4 studiesNo amounts are given in the sources read.

Studies, doses and limits

What the trials found

No trial or review of alcohol reduction for anxiety was read. The statements below come from agency pages.

Results

The NHS advises people not to use alcohol, cigarettes, gambling or drugs to relieve anxiety because these can contribute to poor mental health. WHO lists avoiding or cutting down on alcohol, and not using illicit drugs, as self-care because they can make anxiety worse. NIMH says anxiety disorders commonly occur together with alcohol and substance misuse.

Effect sizes as reported

No effect sizes: agency advice, not trial results.

Why this rating

Compiler's rating: insufficient, meaning no trial evidence was read. The agency advice is consistent, but it is advice, not trial results.

Form and dose studied

No amounts are given in the sources read.

Who was studied

People who drink alcohol, and people with anxiety disorders.

Limits

This card rests on agency pages only.

Why researchers looked

Many people use alcohol to take the edge off anxiety, and anxiety disorders often occur together with alcohol and substance misuse.

Proposed mechanism Hypothesis

Not established in the sources read.

A proposed mechanism does not show that the effect happens in people.

People who find it hard to cut down on alcohol can ask a clinician for support.

^{28 27 33 32}

Lavender oil capsules (the Silexan preparation) and lavender aromatherapy

Low certainty

Several trials of one oral lavender oil capsule product found lower anxiety than placebo and, in one trial, a similar fall to a benzodiazepine medicine. NCCIH says the studies are small and lack independent funding, so no firm conclusion can be drawn.

5 studiesThe oral trials used a standardised lavender oil capsule product (cal…

Studies, doses and limits

What the trials found

A 2019 review of 65 randomised trials (7,993 people) and 25 non-randomised studies of lavender in any form; a 2010 trial of 221 adults with anxiety disorder not otherwise specified; and a 2010 trial comparing the same capsule product with lorazepam in adults with generalized anxiety disorder.

Results

In the 221-adult trial the anxiety score fell by 16.0 points (59.3 percent) on lavender oil and 9.5 points (35.4 percent) on placebo over 10 weeks, and 76.9 percent responded versus 49.1 percent. In the lorazepam trial both groups fell by about 11 points over 6 weeks. The 2019 review found lower anxiety with the oral product and with lavender inhalation, and with lavender oil massage, though it could not tell whether massage benefit came from the oil or the massage. Inhalation studies were mostly in anxiety-inducing situations, not anxiety disorders.

Effect sizes as reported

2010 placebo trial: HAMA fall 16.0 vs 9.5 points (one-sided p < 0.01). 2010 lorazepam trial: fall of 11.3 points on lavender oil, 11.6 on lorazepam. 2019 review: oral product Hamilton Anxiety mean difference -2.90 (95% CI -4.86 to -0.95); inhalation Hedges's g -0.73 (95% CI -1.00 to -0.46).

Why this rating

Compiler's rating: low. Results are positive across several trials, but NCCIH cites small studies, lack of independent funding and testing and lack of diverse participants; the 2019 reviewers describe most trials as high risk of bias; and the lorazepam trial had no placebo arm.

Form and dose studied

The oral trials used a standardised lavender oil capsule product (called Silexan), at 80 mg a day, as reported, for 6 to 10 weeks. Other lavender products differ and cannot be assumed to act the same. These describe what was studied, not what anyone should do.

Who was studied

Adults with generalized anxiety disorder or with anxiety disorder not otherwise specified (oral trials); varied groups (2019 review).

Limits

A single branded preparation, short trials, NCCIH's note about lack of independent funding and testing, and a high risk of bias in most trials in the 2019 review.

Why researchers looked

Lavender has a long traditional reputation for calming, and one standardised oral preparation has been tested in clinical trials.

Proposed mechanism Hypothesis

Proposed: calming effects on brain signalling. Not tested in the trials read.

A proposed mechanism does not show that the effect happens in people.

NCCIH says oral lavender oil can cause diarrhoea, headache, nausea or burping, may theoretically interact with sedatives, and has little safety information for pregnancy and breastfeeding. See the Safety tab.

^{9 22 6 25 23}

Chamomile extract (supplement)

Low certainty

Two small trials of a chamomile extract in generalized anxiety disorder found lower anxiety, but the second one did not show a significant drop in relapse. NCCIH calls the findings preliminary.

3 studiesBoth trials used a standardised chamomile extract in capsules (Matric…

Studies, doses and limits

What the trials found

A 2009 randomised trial of 57 adults with mild to moderate GAD, and a 2016 two-phase trial that enrolled 179 adults with moderate to severe GAD.

Results

In the 2009 trial the anxiety score fell more on chamomile extract than on placebo over 8 weeks (p = 0.047). In the 2016 trial, 93 of 179 people responded to 12 weeks of open-label chamomile, then were randomised to continue chamomile or switch to placebo for 26 weeks. Fewer relapsed on chamomile (15.2 percent versus 25.5 percent), but that difference was not statistically significant, while symptoms stayed lower on chamomile. NCCIH cites a 2019 review of three trials that found no significant drop in situational anxiety.

Effect sizes as reported

2009: p = 0.047 for the larger fall in total anxiety score. 2016: relapse hazard ratio 0.52 (95% CI 0.20 to 1.33; p = 0.16); symptoms lower on chamomile during follow-up (p = 0.0032).

Why this rating

Compiler's rating: low. Both trials were small, the 2009 result was borderline, the main 2016 outcome (relapse) was not significant, and the first phase of that trial had no placebo group.

Form and dose studied

Both trials used a standardised chamomile extract in capsules (Matricaria), not tea. The 2016 trial gave 1,500 mg a day in three capsules in its open-label phase, as reported. These describe what was studied, not what anyone should do.

Who was studied

Adults with generalized anxiety disorder.

Limits

Small trials and an open-label first phase without placebo.

Why researchers looked

Chamomile is a traditional calming herb, usually drunk as a tea, and capsule extracts have been tested for anxiety.

Proposed mechanism Hypothesis

Proposed: mild calming effects on brain signalling. Not tested in the trials read.

A proposed mechanism does not show that the effect happens in people.

Adverse events were similar to placebo in both trials. NCCIH lists headache, dizziness and digestive problems as reported side effects.

^{2 13 23}

Kava (Piper methysticum)

Mixed findings

Kava has shown a small benefit for anxiety in some short trials, but a 16-week trial in generalized anxiety disorder found no difference from placebo. Kava has also been linked to rare but serious liver injury, so it is a herb to discuss with a doctor first.

4 studiesThe 2020 trial used a water-based extract of dried kava root in table…

Studies, doses and limits

What the trials found

A 2018 review of 12 articles on kava for GAD, and a 2020 trial of 171 adults with generalized anxiety disorder who were not taking medicines. NCCIH summarises the wider research and the liver safety record.

Results

In the 2018 review two placebo-controlled trials and one reference-controlled trial supported kava and one placebo-controlled trial did not; pooling three placebo-controlled trials (130 people) favoured kava but not significantly. In the 2020 trial the anxiety score difference favoured placebo by 1.37 points (not significant), and 17.4 percent of kava users and 23.8 percent of placebo users were in remission. Memory problems and tremor were reported more often with kava, and liver test abnormalities were more frequent in the kava group, though no one met criteria for herb-induced liver injury. NCCIH says kava may help anxiety if taken for several weeks, but does not appear to help generalized anxiety disorder.

Effect sizes as reported

2018 review: effect sizes 0.59 to 0.99, not statistically significant. 2020 trial: difference favouring placebo of 1.37 points (p = 0.25); remission 17.4% vs 23.8% (p = 0.46).

Why this rating

Compiler's rating: mixed. Earlier small trials leaned positive, the largest and longest trial was negative, reviews call the evidence insufficient, and liver safety is a serious concern.

Form and dose studied

The 2020 trial used a water-based extract of dried kava root in tablets, standardised for kavalactones, twice a day for 16 weeks, as reported. Kava products differ by plant variety, plant part and preparation. These describe what was studied, not what anyone should do.

Who was studied

Adults with generalized anxiety disorder (reviews and the 2020 trial).

Limits

Small earlier trials, products differ, and the most recent large trial was negative. The 2020 trial reports many participants had liver test changes.

Why researchers looked

Kava is a Pacific Island plant drink, and kava products are promoted for anxiety.

Proposed mechanism Hypothesis

Proposed: calming effects on brain signalling. Not tested in the trials read.

A proposed mechanism does not show that the effect happens in people.

Kava has been linked to rare but serious liver injury, including deaths. See the Safety tab before considering it.

^{14 16 24 23}

Ashwagandha (Withania somnifera) root extract

Low certainty

A meta-analysis of 12 trials found large drops in anxiety and stress scores with ashwagandha extracts, but results varied a lot and the authors rated the certainty low. Rare cases of liver injury have been reported.

2 studiesThe trials used root extracts of different strengths and lengths. The…

Studies, doses and limits

What the trials found

A 2022 meta-analysis of 12 randomised trials (1,002 participants, ages 25 to 48) of ashwagandha extract against placebo, measuring anxiety and stress.

Results

Anxiety and stress scores fell more with ashwagandha than with placebo. The results differed greatly from trial to trial, and the reviewers rated the certainty of the evidence low for both outcomes. The abstract does not say how many participants had a diagnosed anxiety disorder.

Effect sizes as reported

Anxiety: standardised mean difference -1.55 (95% CI -2.37 to -0.74; I-squared 93.8 percent). Stress: -1.75 (95% CI -2.29 to -1.22).

Why this rating

The reviewers' own rating is low certainty for both outcomes. Compiler agrees: very large and inconsistent effects across small trials suggest bias or product differences.

Form and dose studied

The trials used root extracts of different strengths and lengths. These describe what was studied, not what anyone should do.

Who was studied

Adults aged 25 to 48 in trials measuring anxiety or stress; whether they had diagnosed anxiety disorders is not stated in the abstract.

Limits

Very high heterogeneity, small trials, and extract differences. The abstract does not report on safety.

Why researchers looked

Ashwagandha is an Ayurvedic herb sold for stress, and many small trials have measured anxiety.

Proposed mechanism Hypothesis

Proposed: effects on stress hormones and brain signalling. Not tested in the review read.

A proposed mechanism does not show that the effect happens in people.

Case reports describe liver injury linked to ashwagandha supplements, including one case that ended in a liver transplant. See the Safety tab.

^{1 3}

Omega-3 fatty acids (fish oil type supplements)

Low certainty

A meta-analysis of 19 trials found a small drop in anxiety symptoms with omega-3 supplements, mainly in people with a clinical diagnosis and at higher daily amounts. Anxiety was not the main target in many of the trials.

1 studiesThe review pooled trials of different omega-3 products and amounts. B…

Studies, doses and limits

What the trials found

A 2018 meta-analysis of 19 clinical trials (1,203 people on omega-3 and 1,037 not).

Results

Anxiety symptoms were lower with omega-3 treatment than with control arms, and the link was stronger in groups with a specific clinical diagnosis. The benefit was significant only in the subgroups with higher daily amounts, not lower amounts.

Effect sizes as reported

Hedges's g 0.374 (95% CI 0.081 to 0.666; p = 0.01). Average amount in the trials 1,605.7 mg a day, as reported.

Why this rating

Compiler's rating: low. The authors say omega-3 might help and that studies in people whose main problem is anxiety are needed. The trials were mixed in populations and design.

Form and dose studied

The review pooled trials of different omega-3 products and amounts. Benefit appeared only in trials using at least 2,000 mg a day, as reported. These describe what was studied, not what anyone should do.

Who was studied

Adults in varied groups, 55 percent women, mean age about 44 and 41 in the two arms.

Limits

Mixed populations, anxiety often a secondary outcome, and no information on safety in the abstract.

Why researchers looked

Omega-3 fats are part of brain cells, and researchers asked whether supplements calm anxiety.

Proposed mechanism Hypothesis

Proposed: effects on inflammation and brain cell membranes. Not tested in the review read.

A proposed mechanism does not show that the effect happens in people.

The sources read for this page do not cover side effects or interactions of omega-3 supplements, so ask a doctor or pharmacist, especially if you take regular medicines.

²¹

L-theanine (an amino acid from tea)

Not enough evidence

One small trial added L-theanine to the antidepressant treatment of 46 adults with generalized anxiety disorder and found no benefit for anxiety, though sleep satisfaction was better.

1 studiesThe trial used 450 to 900 mg a day as an add-on to stable antidepress…

Studies, doses and limits

What the trials found

A 2019 randomised trial of 46 adults with GAD who took L-theanine or matching placebo with their current stable antidepressant for 8 weeks.

Results

L-theanine did not beat placebo for anxiety or for insomnia severity. Self-reported sleep satisfaction was better on L-theanine, and in people with little insomnia the sleep score separated in its favour.

Effect sizes as reported

Anxiety (HAMA) p = 0.73; insomnia severity p = 0.35; sleep satisfaction item p = 0.015.

Why this rating

Compiler's rating: insufficient. One small preliminary trial was read; it did not support benefit for anxiety, and no larger trial in anxiety disorders was read.

Form and dose studied

The trial used 450 to 900 mg a day as an add-on to stable antidepressant treatment, as reported. These describe what was studied, not what anyone should do.

Who was studied

Adults with generalized anxiety disorder already on a stable antidepressant.

Limits

Very small trial, short, and sleep results are secondary and preliminary.

Why researchers looked

L-theanine is found in tea leaves and is sold for calm and sleep.

Proposed mechanism Hypothesis

Proposed: effects on brain signalling. Not tested in the trial read.

A proposed mechanism does not show that the effect happens in people.

The abstract does not report side effects. Ask a clinician before adding any supplement to prescribed medicines.

¹⁵

Magnesium supplements

Very low certainty

A systematic review found some trials suggesting less anxiety with magnesium in people prone to anxiety, but the quality of the evidence is poor.

1 studiesForms and amounts varied by study and are not given in the abstract. …

Studies, doses and limits

What the trials found

A 2017 systematic review of 18 studies of magnesium alone or with up to five other ingredients in people vulnerable to anxiety (mildly anxious, premenstrual syndrome, postpartum, high blood pressure).

Results

Four of eight studies in anxious people, four of seven in premenstrual syndrome and one of two in high blood pressure reported positive effects. Magnesium had no effect on postpartum anxiety. No study used a validated stress measure.

Effect sizes as reported

No pooled effect size in the abstract.

Why this rating

Compiler's rating: very low. The reviewers call the quality of the evidence poor. Some studies mixed magnesium with other ingredients.

Form and dose studied

Forms and amounts varied by study and are not given in the abstract. These describe what was studied, not what anyone should do.

Who was studied

People mildly anxious, with premenstrual syndrome, postpartum, or with high blood pressure.

Limits

Poor-quality trials, combination products, and no pooled analysis.

Why researchers looked

Low magnesium has been linked to anxiety, so researchers asked whether supplements help.

Proposed mechanism Hypothesis

Proposed: magnesium's role in nerve signalling. Not tested in the review read.

A proposed mechanism does not show that the effect happens in people.

The review abstract does not cover safety or interactions. Ask a doctor or pharmacist before taking any supplement, especially if you have a health condition or take medicines.

⁴

Acupuncture

Low certainty

NCCIH says there is limited evidence that acupuncture may reduce anxiety symptoms, but the overall quality of the studies is low.

1 studiesNot reported in the NCCIH summary.

Studies, doses and limits

What the trials found

NCCIH summarises a 2022 review of 27 randomised trials (1,782 people) in generalized anxiety disorder and a 2021 analysis of 12 studies (916 patients) of acupuncture before surgery. Neither review was opened here.

Results

NCCIH reports that the 2022 review concluded acupuncture relieved anxiety symptoms better than the comparison treatments, mostly medicines, with fewer side effects. The authors called for more high-quality trials.

Effect sizes as reported

No figures were read.

Why this rating

Compiler's rating: low, following NCCIH's statement that overall study quality is low. This card rests on the NCCIH summary only.

Form and dose studied

Not reported in the NCCIH summary.

Who was studied

Adults with generalized anxiety disorder, and people having surgery.

Limits

Read second-hand through NCCIH; study quality low.

Why researchers looked

Acupuncture is a common complementary approach, and several reviews have examined it for anxiety.

Proposed mechanism Hypothesis

Not established in the sources read.

A proposed mechanism does not show that the effect happens in people.

NCCIH says serious complications are rare but can follow non-sterile needles or improper treatment, including infection, punctured organs and nervous system injury.

²³

Valerian and passionflower

Not enough evidence

NCCIH says there is not enough evidence to say whether valerian helps anxiety, and only a small amount of research suggests passionflower might help, with no definite conclusions.

1 studiesNot reported in the NCCIH summary.

Studies, doses and limits

What the trials found

No trial of either herb was opened here. The statements come from NCCIH.

Results

NCCIH says passionflower may help reduce anxiety and anxiety before a dental procedure in a small amount of research, and that the evidence for valerian is not enough for any conclusions.

Effect sizes as reported

No figures were read.

Why this rating

Compiler's rating: insufficient, following NCCIH.

Form and dose studied

Not reported in the NCCIH summary.

Who was studied

Adults; details not in the summary.

Limits

Read second-hand through NCCIH; no trial opened.

Why researchers looked

Both are traditional calming herbs sold for nerves and sleep.

Proposed mechanism Hypothesis

Not established in the sources read.

A proposed mechanism does not show that the effect happens in people.

NCCIH says valerian is generally safe for short-term use by most adults and that long-term safety is unknown, and that passionflower is considered safe when taken by mouth.

²³

Connections to the rest of health

How Anxiety Disorders: Symptoms, Evaluation and Care connects to the rest of our health knowledge graph. Every link says what kind of connection it is. A safety link (for example to the liver) is a warning, not a benefit. A study that merely mentions a topic never creates a benefit link.

Explore every connection: goals, body systems, organs, symptoms, genes

Kind of connection Connected to Search

A. Wellness goals

Connected to	Evidence status	Context	Studies
Stress Anxiety	Moderate certainty	Regular exercise and physical activity: Reviews of trials found lower anxiety symptoms with exercise than with usual care or no exercise, a medium-sized effect. Most of the reviews were of low quality, and only six small trials looked at people with a diagnosed anxiety disorder. Mindfulness-based stress reduction (MBSR) and mindfulness programmes: One large trial found an 8-week mindfulness course as good as a commonly prescribed anxiety medicine in adults with anxiety disorders. Pooled studies also found lower anxiety, but many were not compared with a control group.	^{20 18 28 32 33 8 7 23 17 12 11 10 9 22 6 25 14 16 24 1 3 21}
Stress Anxiety	Not enough evidence	Cutting back on alcohol: Health agencies advise against using alcohol to relieve anxiety because it can make anxiety worse and is linked to poor mental health. No trial of cutting back alcohol as an anxiety treatment was read for this page. Acupuncture: NCCIH says there is limited evidence that acupuncture may reduce anxiety symptoms, but the overall quality of the studies is low.	^{28 27 33 32 23}
Stress Anxiety	Low certainty	Chamomile extract (supplement): Two small trials of a chamomile extract in generalized anxiety disorder found lower anxiety, but the second one did not show a significant drop in relapse. NCCIH calls the findings preliminary. L-theanine (an amino acid from tea): One small trial added L-theanine to the antidepressant treatment of 46 adults with generalized anxiety disorder and found no benefit for anxiety, though sleep satisfaction was better.	^{2 13 23 15}
Stress Anxiety	Very low certainty	Magnesium supplements: A systematic review found some trials suggesting less anxiety with magnesium in people prone to anxiety, but the quality of the evidence is poor.	⁴

B. Body systems

Connected to	Kind of connection	Evidence status	Context	Studies
Nervous System	Biological association	Moderate certainty	Studied for: Regular exercise and physical activity Studied for: Mindfulness-based stress reduction (MBSR) and mindfulness programmes	^{20 18 28 32 33 8 7 23 17 12 11 10 27 9 22 6 25 2 13 14 16 24 1 3 21 15 4}
Stress Mood	Biological association	Moderate certainty	Studied for: Regular exercise and physical activity Studied for: Mindfulness-based stress reduction (MBSR) and mindfulness programmes	^{20 18 28 32 33 8 7 23 17 12 11 10 27 9 22 6 25 2 13 14 16 24 1 3 21 15 4}

E. Health conditions

Connected to	Kind of connection	Evidence status	Context	Studies
Panic attacks and panic disorder		Health agency guidance	Large caffeine amounts brought on panic attacks in about half of people with panic disorder in blinded studies. Preparation or compound: Caffeine	¹⁰

I. Biomarkers

Connected to	Kind of connection	Evidence status	Context	Studies
Anxiety symptom scores (clinician-rated and self-rated)	Biological association	Moderate certainty	Measured in trials: lower scores with several lifestyle and mind-body approaches; smaller gaps against active treatments. A lab change is not the same as a health benefit.	^{8 17}
Panic attacks after a caffeine challenge	Biological association	Low certainty	Measured in trials: more panic attacks after caffeine in people with panic disorder. A lab change is not the same as a health benefit.	¹⁰
Hamilton Anxiety Rating Scale (HAM-A) total score	Biological association	Low certainty	Measured in trials: lower with oral lavender oil and chamomile extract than with placebo in short trials. A lab change is not the same as a health benefit.	^{9 2}

L. Active compounds

Connected to	Kind of connection	Evidence status	Context	Studies
Cognitive behavioural therapy (CBT)	Biological association	Human	First-line talking therapy for anxiety disorders. A 2018 analysis of 41 placebo-controlled trials (2,843 people) found moderate effects of CBT on the main symptoms across anxiety-related disorders, with large effects for generalized anxiety disorder.	^{5 32 30 33}
Acceptance and commitment therapy (ACT)	Biological association	Agency summary	NIMH describes it as another option for GAD with a growing body of supporting research.	³²
SSRI and SNRI antidepressants	Biological association	Human	Commonly used for anxiety disorders and may take several weeks to work (NIMH). A 2019 network analysis of 89 trials found that several medicines across drug classes were more effective than placebo with relatively good tolerability, while some were poorly tolerated.	^{32 30 19}
Benzodiazepines and buspirone	Biological association	Agency summary	NIMH says benzodiazepines can reduce anxiety rapidly but cause tolerance and dependence in some people, so they are prescribed only briefly; buspirone is not a sedative and has less potential to be addictive. NICE and WHO also limit benzodiazepines.	^{32 30 33}
Exercise, sleep, caffeine and alcohol habits	Biological association	Human	Studied as support alongside care.	^{20 18 11 10}
Mindfulness, yoga and relaxation	Biological association	Human	Studied alone or alongside care.	^{8 17 12}
Lavender oil, chamomile, kava, ashwagandha	Biological association	Human	Studied as capsule or tablet extracts.	^{6 13 16 1}
Omega-3, L-theanine, magnesium	Biological association	Human	Studied as supplements.	^{21 15 4}

Each connection is stored with the ingredient, the connected record, the kind of connection, the evidence status, the supporting studies, the preparation or compound involved, and a review date (last reviewed October 31, 2026).

Evidence at a glance: what the research found for 16 outcomes

2Moderate certainty

8Low certainty

1Very low certainty

2Mixed findings

3Not enough evidence

Each outcome below is one thing researchers measured. Certainty is how far the evidence can be trusted, not how well it works. The full studies are on the next tab.

Regular exercise and physical activity 5 studies Moderate certainty Mindfulness-based stress reduction (MBSR) and mindfulness programmes 3 studies · 276 people in trials Moderate certainty Yoga (Kundalini yoga in the main trial) 2 studies · 226 people in trials Mixed findings Relaxation training (progressive relaxation, autogenic training, applied relaxation) 3 studies Low certainty Better sleep (digital cognitive behavioural therapy for insomnia) 3 studies Low certainty Cutting back on caffeine 3 studies Low certainty Cutting back on alcohol 4 studies Not enough evidence Lavender oil capsules (the Silexan preparation) and lavender aromatherapy 5 studies · 221 people in trials Low certainty Chamomile extract (supplement) 3 studies · 236 people in trials Low certainty Kava (Piper methysticum) 4 studies · 171 people in trials Mixed findings Ashwagandha (Withania somnifera) root extract 2 studies Low certainty Omega-3 fatty acids (fish oil type supplements) 1 study Low certainty L-theanine (an amino acid from tea) 1 study · 46 people in trials Not enough evidence Magnesium supplements 1 study Very low certainty Acupuncture 1 study Low certainty Valerian and passionflower 1 study Not enough evidence

The evidence

The research centre for Anxiety Disorders: Symptoms, Evaluation and Care. Every record here was checked against PubMed or the publisher's page; we never list a study we have not verified.

Evidence by outcome

Bars show how many studies we found for each outcome, not how well it works. People are added up across trials only (reviews are not added, to avoid double counting).

How we rate evidence

High certainty

Several well-run trials or reviews agree; further research is unlikely to change the picture much.

Moderate certainty

Good trials point the same way, but some doubt remains.

Low certainty

Small, short or inconsistent trials. The finding could change with better research.

Very low certainty

Very little or very weak human data. Treat as a possibility only.

Mixed findings

Trials or reviews disagree with each other.

Not enough evidence

Not enough human research to say either way.

Evidence against

Good research found no benefit, or found harm.

We take the reviewers' own risk-of-bias and GRADE statements where they exist, and otherwise judge from study size, design, consistency, directness and whether a product matches what is sold. Many publications do not make evidence high. Method note: We searched PubMed, read the abstract of every review and trial cited, and read the NIMH, NHS, NICE, WHO and NCCIH pages as downloaded text. Full texts were not read. Certainty ratings are the compiler's own reading unless a review gives its own grade (stated in the text). Funding is not stated in most abstracts, and this is noted where relevant.

Study library: every study we checked 33

Design Research type Year from Search

Study	Design	People	Product, dose, length	Result
Akhgarjand 2022 Does Ashwagandha supplementation have a beneficial effect on the management of anxiety and stress? A systematic review and meta-analysis of randomized controlled trials Phytother Res · PMID 36017529 · DOI	Systematic review meta analysis	1002 12 randomised trials, 1,002 participants aged 25 to 48; searched to December 2021	Ashwagandha (Withania somnifera) extract Not given in the abstract · vs Placebo	Anxiety and stress scores were lower with ashwagandha than with placebo, with very high heterogeneity. The reviewers rated the certainty of evidence low for both outcomes. Anxiety SMD -1.55; stress SMD -1.75 (Anxiety -2.37 to -0.74 (I-squared 93.8 percent); stress -2.29 to -1.22 (I-squared 83.1 percent)) Methods, limits and funding Outcomes: Anxiety and stress scores Limitations: Very high heterogeneity; low certainty by the reviewers' own rating; the abstract does not say whether participants had anxiety disorders. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Amsterdam 2009 A randomized, double-blind, placebo-controlled trial of oral Matricaria recutita (chamomile) extract therapy for generalized anxiety disorder J Clin Psychopharmacol · PMID 19593179 · DOI	Controlled trial	57 57 outpatients with mild to moderate generalized anxiety disorder randomised (61 enrolled)	Chamomile (Matricaria recutita) extract 8 weeks · vs Placebo	Greater fall in total HAM-A score on chamomile than on placebo; secondary outcomes moved in the same direction. Adverse events were not significantly different. p = 0.047 for the group difference (Not given in the abstract) Methods, limits and funding Outcomes: Hamilton Anxiety Rating total score; secondary scales Limitations: Small; not powered for secondary outcomes; borderline p value; the authors call for replication. Funding and conflicts: Record lists NIH extramural research support; no conflict statement read. Read: abstract
Bokan 2023 Herb-Induced Liver Injury by Ayurvedic Ashwagandha as Assessed for Causality by the Updated RUCAM: An Emerging Cause Pharmaceuticals (Basel) · PMID 37631044 · DOI	Case reports and review	All previously reported cases of ashwagandha-related liver injury, plus two new cases (a 36-year-old man and a 30-year-old woman)	Ashwagandha (Withania somnifera root) capsules Onset after about 45 days to 6 months in the two new cases	Liver enzymes and bilirubin rose in both new cases, causality was rated probable, and both improved after stopping the supplement. One earlier reported case ended in liver transplantation. The authors conclude ashwagandha supplements are not free of safety concerns. RUCAM score 7 in both new cases (probable) (Not given: a case report) Methods, limits and funding Outcomes: Liver injury and causality assessment (RUCAM) Limitations: Case reports cannot show how common liver injury is, and product contents were not tested in the abstract. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Boyle 2017 The Effects of Magnesium Supplementation on Subjective Anxiety and Stress-A Systematic Review Nutrients · PMID 28445426 · DOI	Systematic review	18 studies of magnesium alone or with up to five other ingredients in people vulnerable to anxiety (mildly anxious, premenstrual syndrome, postpartum, hypertension); searched May 2016	Magnesium supplements, alone or in combination Not given in the abstract · vs Various	Four of eight studies in anxious samples, four of seven in premenstrual syndrome and one of two in hypertension reported benefit; no effect in postpartum anxiety. No validated stress measure was used. No pooled effect size (Not given: a narrative review without a pooled estimate) Methods, limits and funding Outcomes: Subjective anxiety and stress Limitations: The reviewers call the quality of existing evidence poor; combination products. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Carpenter 2018 Cognitive behavioral therapy for anxiety and related disorders: A meta-analysis of randomized placebo-controlled trials Depress Anxiety · PMID 29451967 · DOI	Systematic review meta analysis	2843 41 randomised placebo-controlled studies (2,843 patients) with acute stress disorder, generalized anxiety disorder, obsessive compulsive disorder, panic disorder, post-traumatic stress disorder or social anxiety disorder	Cognitive behavioural therapy (CBT) Not given in the abstract · vs Psychological or pill placebo	Moderate placebo-controlled effects of CBT on target symptoms, small to moderate effects on other anxiety symptoms, depression and quality of life. Large effects for obsessive compulsive disorder, GAD and acute stress disorder; small to moderate for PTSD, social anxiety and panic disorder. In PTSD studies dropout was higher with CBT. Target symptoms Hedges's g 0.56; response odds ratio 2.97 (Not given in the abstract) Methods, limits and funding Outcomes: Target disorder symptoms, other anxiety symptoms, depression, quality of life Limitations: Effects stronger in completer than intention-to-treat samples; results vary by disorder; the authors say more effective treatments are needed for PTSD, social anxiety and panic. Funding and conflicts: Record lists NIH extramural and non-US government support; no conflict statement read. Read: abstract
Donelli 2019 Effects of lavender on anxiety: A systematic review and meta-analysis Phytomedicine · PMID 31655395 · DOI	Systematic review meta analysis	7993 65 randomised trials (7,993 participants) and 25 non-randomised studies (1,200) of lavender in any form for anxiety or anxiety-inducing settings; searched to November 2018	Lavender: oral Silexan, inhalation, and massage with lavender oil Oral product 6 weeks or more · vs Any type of control	Lavender inhalation, oral Silexan and lavender oil massage were each linked to lower anxiety. The reviewers rate the average quality of studies low and most trials at high risk of bias. Inhalation Hedges's g -0.73; oral HAM-A mean difference -2.90; massage g -0.66 (Inhalation -1.00 to -0.46; oral -4.86 to -0.95; massage -0.97 to -0.35) Methods, limits and funding Outcomes: Anxiety scales Limitations: Low average study quality, high risk of bias, and varied designs especially for non-oral use. Funding and conflicts: Review not funded (as stated in the abstract). Registration: CRD42019130126 Read: abstract
Hofmann 2010 The effect of mindfulness-based therapy on anxiety and depression: A meta-analytic review J Consult Clin Psychol · PMID 20350028 · DOI	Systematic review meta analysis	1140 39 studies, 1,140 people receiving mindfulness-based therapy for cancer, generalized anxiety disorder, depression and other psychiatric or medical conditions	Mindfulness-based therapy Not given in the abstract; effects unrelated to number of sessions · vs Before-and-after change (pre- to post-treatment)	Moderate improvement in anxiety and mood from before to after treatment, larger in people with anxiety and mood disorders, and maintained at follow-up. Anxiety g 0.63 overall and 0.97 in people with anxiety and mood disorders; mood g 0.59 and 0.95 (Not given in the abstract) Methods, limits and funding Outcomes: Anxiety and mood symptoms Limitations: Effect sizes are before-and-after changes, not comparisons with a control group; mixed conditions; older searches. Funding and conflicts: Record lists NIH extramural research support; no conflict statement read. Read: abstract
Hoge 2023 Mindfulness-Based Stress Reduction vs Escitalopram for the Treatment of Adults With Anxiety Disorders: A Randomized Clinical Trial JAMA Psychiatry · PMID 36350591 · DOI	Controlled trial	276 276 adults with a diagnosed anxiety disorder at 3 US medical centres (208 completed); mean age 33, 75 percent women	Mindfulness-based stress reduction (MBSR), weekly group course 8 weeks, follow-up to 24 weeks · vs Escitalopram (an antidepressant), flexibly dosed as reported	MBSR was noninferior to escitalopram on the main anxiety measure. At least one study-related adverse event occurred in 15.4 percent on MBSR and 78.6 percent on escitalopram. Score fell by 1.35 on MBSR and 1.43 on escitalopram; difference -0.07 (-0.38 to 0.23 (noninferiority margin -0.495)) Methods, limits and funding Outcomes: Clinical Global Impression of Severity (anxiety), rated by a blinded interviewer Limitations: No placebo or waiting-list arm, so it cannot show how much either beat no treatment; about one in four did not complete it. Funding and conflicts: Record lists NIH extramural research support; no conflict statement read. Registration: NCT03522844 Read: abstract
Kasper 2010 Silexan, an orally administered Lavandula oil preparation, is effective in the treatment of 'subsyndromal' anxiety disorder: a randomized, double-blind, placebo controlled trial Int Clin Psychopharmacol · PMID 20512042 · DOI	Controlled trial	221 221 adults with anxiety disorder not otherwise specified (DSM-IV 300.00 or ICD-10 F41.9), from 27 general and psychiatric practices; HAMA total score 18 or more	Silexan, a standardised oral Lavandula oil preparation, 80 mg a day 10 weeks · vs Placebo	Anxiety scores fell by 16.0 points on silexan and 9.5 on placebo; responders 76.9 versus 49.1 percent; remitters 60.6 versus 42.6 percent. Sleep quality also improved more. HAMA fall 16.0 vs 9.5 points (Not given; p < 0.01 one-sided, intention to treat) Methods, limits and funding Outcomes: Hamilton Anxiety Scale and Pittsburgh Sleep Quality Index total score change Limitations: One-sided p value; a single branded preparation; funding not stated in the abstract. NCCIH notes a lack of independent funding and testing. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Klevebrant 2022 Effects of caffeine on anxiety and panic attacks in patients with panic disorder: A systematic review and meta-analysis Gen Hosp Psychiatry · PMID 34871964 · DOI	Systematic review meta analysis	237 10 blinded placebo-controlled caffeine challenge studies in adults with panic disorder (237 patients in the 9 panic-attack studies), with healthy adult comparison groups	Caffeine, one-off challenge doses of 400 to 750 mg as reported Single acute challenge · vs Placebo; healthy controls	51.1 percent of patients had a panic attack after caffeine and none after placebo. Patients (53.9 percent) were far more vulnerable than healthy controls (1.7 percent). Caffeine also raised anxiety, more in patients than in healthy adults. Log risk ratio 3.47 (patients vs controls); Hedges's g 1.02 for subjective anxiety (Log RR 95% CI 2.06 to 4.87; g 95% CI 0.09 to 1.96) Methods, limits and funding Outcomes: Panic attacks and subjective anxiety Limitations: Few studies; only 2 reported anxiety data for placebo; narrow dose range, so smaller amounts and dose-response could not be assessed. Funding and conflicts: Record lists non-US government support; no conflict statement read. Registered: PROSPERO CRD42019120220. Registration: CRD42019120220 Read: abstract
Lee 2023 Digital cognitive behavioral therapy for insomnia on depression and anxiety: a systematic review and meta-analysis NPJ Digit Med · PMID 36966184 · DOI	Systematic review meta analysis	22 randomised trials in adults with insomnia who also reported depression or anxiety measures (anxiety in 18 trials)	Digital cognitive behavioural therapy for insomnia (dCBT-I) Not given in the abstract · vs Control groups in each trial	dCBT-I gave a small to moderate drop in depressive symptoms, a small drop in anxiety symptoms and a large improvement in sleep. Effects were stronger with high adherence and also seen with fully automated programmes. Anxiety SMD -0.29; depression -0.42; sleep -0.76 (Anxiety -0.40 to -0.19; depression -0.56 to -0.28; sleep -0.95 to -0.57) Methods, limits and funding Outcomes: Depressive symptoms, anxiety symptoms, sleep measures Limitations: People had insomnia, not an anxiety disorder; abstract gives no follow-up length. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Manzoni 2008 Relaxation training for anxiety: a ten-years systematic review with meta-analysis BMC Psychiatry · PMID 18518981 · DOI	Systematic review meta analysis	27 studies (1997 to 2007): randomised trials, observational studies and studies without a control group, in people with anxiety problems and disorders and in volunteers	Relaxation training: Jacobson's progressive relaxation, autogenic training, applied relaxation and meditation Not given in the abstract; longer treatments did better · vs Within-group change and between-group comparisons	Relaxation training lowered anxiety by a medium to large amount. Efficacy was higher for meditation, among volunteers and for longer treatments. Cohen's d 0.57 within groups; 0.51 between groups (Within 0.52 to 0.68; between 0.46 to 0.634) Methods, limits and funding Outcomes: Anxiety on questionnaires Limitations: Mixed designs including uncontrolled studies; searches ended in 2007; bigger effects in volunteers than patients. Funding and conflicts: Record lists non-US government support; no conflict statement read. Read: abstract
Mao 2016 Long-term chamomile (Matricaria chamomilla L.) treatment for generalized anxiety disorder: A randomized clinical trial Phytomedicine · PMID 27912875 · DOI	Controlled trial	179 179 outpatients with moderate to severe GAD enrolled; 93 responders were randomised to continue or switch to placebo (46 chamomile, 47 placebo)	Chamomile (Matricaria chamomilla L.) pharmaceutical-grade extract, open-label treatment followed by a double-blind period 12 weeks open-label then 26 weeks randomised · vs Placebo substitution	Relapse 15.2 percent on chamomile versus 25.5 percent on placebo, not statistically significant. Symptoms stayed significantly lower on chamomile, and adverse events were similar. The authors say the sample was small. Relapse hazard ratio 0.52; symptoms p = 0.0032 (Hazard ratio 95% CI 0.20 to 1.33 (p = 0.16)) Methods, limits and funding Outcomes: Time to relapse (primary); relapse proportion, adverse events and vital signs Limitations: Primary outcome not significant; small sample; first phase had no placebo group; low placebo relapse rate. Funding and conflicts: No funding or conflict statement in the record read. Registration: NCT01072344 Read: abstract
Ooi 2018 Kava for Generalized Anxiety Disorder: A Review of Current Evidence J Altern Complement Med · PMID 29641222 · DOI	Systematic review	130 12 articles on kava for generalized anxiety disorder; meta-analysis of 3 placebo-controlled trials (130 people)	Kava (Piper methysticum) preparations Short-term trials, 4 to 8 weeks as the abstract says for safety · vs Placebo or reference treatment	Two placebo-controlled trials and a reference-controlled trial supported kava; one placebo-controlled trial found it no better than placebo. A pooled analysis of three trials favoured kava without reaching statistical significance. The reviewers call the evidence insufficient to confirm an effect beyond placebo. Standardised mean differences 0.59 to 0.99, not significant (Not given in the abstract) Methods, limits and funding Outcomes: GAD symptoms Limitations: Few small trials; preparations differ; English-language studies only. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Sarris 2019 L-theanine in the adjunctive treatment of generalized anxiety disorder: A double-blind, randomised, placebo-controlled trial J Psychiatr Res · PMID 30580081 · DOI	Controlled trial	46 46 adults with DSM-5 generalized anxiety disorder on stable antidepressant treatment	L-theanine added to current antidepressant, 450 to 900 mg a day as reported 8-week double-blind period within a 10-week study · vs Matching placebo added to current antidepressant	L-theanine did not beat placebo for anxiety or insomnia severity. Sleep satisfaction was better on L-theanine, and in people with little insomnia the sleep score separated in its favour. No cognitive effects were seen. HAMA p = 0.73; ISI p = 0.35; sleep satisfaction item p = 0.015 (Not given in the abstract) Methods, limits and funding Outcomes: Anxiety (HAMA), insomnia severity, sleep satisfaction, cognition Limitations: Very small and preliminary; the authors say further studies of sleep are warranted. Funding and conflicts: Record lists non-US government support; no conflict statement read. Read: abstract
Sarris 2020 Kava for generalised anxiety disorder: A 16-week double-blind, randomised, placebo-controlled study Aust N Z J Psychiatry · PMID 31813230 · DOI	Controlled trial	171 171 non-medicated adults with diagnosed generalised anxiety disorder in Australia	Aqueous extract of dried kava root in tablets, standardised for kavalactones, twice a day, as reported 16 weeks · vs Placebo	No significant difference from placebo (a 1.37-point relative reduction favouring placebo). Memory problems and tremor were more frequent with kava, and liver function test abnormalities were significantly more frequent, though no participant met criteria for herb-induced liver injury. Difference favouring placebo 1.37 points; remission 17.4% vs 23.8% (p = 0.25 for the main difference; p = 0.46 for remission) Methods, limits and funding Outcomes: Anxiety reduction on the Hamilton Anxiety Rating Scale; safety; genetic moderators Limitations: One extract only; non-medicated participants; liver test changes more frequent on kava. Funding and conflicts: Record lists non-US government support; no conflict statement read. Read: abstract
Simon 2021 Efficacy of Yoga vs Cognitive Behavioral Therapy vs Stress Education for the Treatment of Generalized Anxiety Disorder: A Randomized Clinical Trial JAMA Psychiatry · PMID 32805013 · DOI	Controlled trial	226 226 adults with generalized anxiety disorder from 2 academic centres (155 completed); mean age 33.4, 69.9 percent women	Kundalini yoga in groups of 4 to 6 Twelve 120-minute sessions plus 20 minutes of daily homework over 12 weeks · vs CBT for GAD, or stress education	Response was 54.2 percent with yoga, 70.8 percent with CBT and 33.0 percent with stress education. Yoga was not shown to be as effective as CBT. Yoga vs stress education OR 2.46; CBT vs stress education OR 5.00; yoga vs CBT difference 16.6 percent (Yoga vs stress education 1.12 to 5.42; CBT vs stress education 2.12 to 11.82) Methods, limits and funding Outcomes: Acute GAD response (much or very much improved) at 12 weeks, rated by trained independent raters Limitations: Single-blind; about 31 percent did not complete; stress education is a weak comparison; one yoga style. Funding and conflicts: Record lists NIH extramural research support (NCCIH supported per its anxiety page); no conflict statement read. Registration: NCT01912287 Read: abstract
Singh 2023 Effectiveness of physical activity interventions for improving depression, anxiety and distress: an overview of systematic reviews Br J Sports Med · PMID 36796860 · DOI	Umbrella review	128119 97 systematic reviews with meta-analyses (1,039 trials, 128,119 adults), including healthy adults, people with mental health disorders and people with chronic disease; searched to 1 January 2022	Physical activity interventions Not given; benefit was smaller in longer interventions · vs Usual care	Physical activity had medium effects on depression, anxiety and distress compared with usual care. Higher-intensity activity was linked to greater improvement; effectiveness diminished with longer interventions. Anxiety median effect size -0.42; depression -0.43; distress -0.60 (Anxiety IQR -0.66 to -0.26; distress 95% CI -0.78 to -0.42) Methods, limits and funding Outcomes: Depression, anxiety, psychological distress Limitations: 77 of the 97 reviews scored critically low on a quality tool; anxiety results mix healthy people and people with disorders. Funding and conflicts: No funding or conflict statement in the record read. Registered: PROSPERO CRD42021292710. Registration: CRD42021292710 Read: abstract
Slee 2019 Pharmacological treatments for generalised anxiety disorder: a systematic review and network meta-analysis Lancet · PMID 30712879 · DOI	Network meta analysis	25441 89 randomised trials (25,441 adult outpatients with generalised anxiety disorder) of 22 active drugs against placebo or each other; studies published 1994 to 2017	Medicines for anxiety (several drug classes; individual drugs are not named on this page) Not given in the abstract · vs Placebo or active drugs	Several medicines across drug classes were more effective than placebo with relatively good acceptability. Some others were effective but poorly tolerated compared with placebo. The authors say the failure of initial drug therapy might not be a reason to abandon drug treatment. Not repeated here: effect sizes differ by drug (Not repeated here) Methods, limits and funding Outcomes: Change in Hamilton Anxiety Scale score; study discontinuations for any cause Limitations: Some findings limited by small sample sizes; a network analysis rests on indirect comparisons. Funding and conflicts: Authors state no funding was received. Registration: CRD42018087106 Read: abstract
Stubbs 2017 An examination of the anxiolytic effects of exercise for people with anxiety and stress-related disorders: A meta-analysis Psychiatry Res · PMID 28088704 · DOI	Systematic review meta analysis	262 6 randomised trials, 262 adults (132 exercise, 130 control), with an anxiety and/or stress-related disorder; searched to December 2015	Exercise (types not listed in the abstract) Not given in the abstract · vs Usual treatment or control conditions	Exercise lowered anxiety symptoms more than control conditions, a moderate effect. Standardised mean difference -0.582, p = 0.02 (The abstract prints -1.0 to -0.76, which does not contain the estimate, so no interval is repeated) Methods, limits and funding Outcomes: Anxiety symptoms Limitations: Only six small trials; exercise types and diagnoses are not given in the abstract. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Su 2018 Association of Use of Omega-3 Polyunsaturated Fatty Acids With Changes in Severity of Anxiety Symptoms: A Systematic Review and Meta-analysis JAMA Netw Open · PMID 30646157 · DOI	Systematic review meta analysis	2240 19 clinical trials (1,203 people on omega-3 and 1,037 controls), mean age about 44 and 41, 55 percent women; searched to 4 March 2018	Omega-3 polyunsaturated fatty acids, mean 1,605.7 mg a day as reported Not given in the abstract · vs Control arms (placebo-controlled and not)	Anxiety symptoms were lower with omega-3. The benefit was larger in groups with specific clinical diagnoses and significant only at the higher amounts (at least 2,000 mg a day). Hedges's g 0.374 (0.081 to 0.666 (p = 0.01)) Methods, limits and funding Outcomes: Change in severity of anxiety symptoms Limitations: Mixed populations; anxiety often not the main complaint; some trials not placebo-controlled; the authors call for better-designed studies. Funding and conflicts: Record lists non-US government support; no conflict statement read. Read: abstract
Woelk 2010 A multi-center, double-blind, randomised study of the Lavender oil preparation Silexan in comparison to Lorazepam for generalized anxiety disorder Phytomedicine · PMID 19962288 · DOI	Controlled trial	Adults with generalized anxiety disorder in a multi-centre trial (number not given in the abstract)	Silexan, an oral lavender oil capsule preparation 6 weeks · vs Lorazepam (a benzodiazepine)	Anxiety scores fell to a similar extent in both groups: 11.3 points (45 percent) on silexan and 11.6 points (46 percent) on lorazepam. HAM-A fall 11.3 vs 11.6 points (Not given in the abstract) Methods, limits and funding Outcomes: Change in Hamilton Anxiety Rating Scale total score Limitations: No placebo arm, so it cannot show how much either beat no treatment; number of participants and sponsor not given in the abstract. Funding and conflicts: No funding or conflict statement in the record read; the record lists non-US government support. Read: abstract
s-nccih-anxiety NCCIH: Anxiety and Complementary Health Approaches (Last Updated September 2024) · Source	Health agency n/a			Methods, limits and funding Read: full-text
s-nccih-kava NCCIH: Kava (Last Updated April 2025) · Source	Health agency n/a			Methods, limits and funding Read: full-text
s-nccih-lavender NCCIH: Lavender (Last Updated February 2025) · Source	Health agency n/a			Methods, limits and funding Read: full-text
s-nccih-stj NCCIH: St. John's Wort and Depression: In Depth (Last Updated December 2017) · Source	Health agency n/a			Methods, limits and funding Read: full-text
s-nhs-anxiety NHS: Anxiety, fear and panic (page last reviewed 13 July 2026) · Source	Health agency n/a			Methods, limits and funding Read: full-text
s-nhs-gad NHS: Generalised anxiety disorder (GAD) (page last reviewed 22 October 2024) · Source	Health agency n/a			Methods, limits and funding Read: full-text
s-nhs-suicidal NHS: Help for suicidal thoughts (page last reviewed 26 April 2024) · Source	Health agency n/a			Methods, limits and funding Read: full-text
s-nice-cg113 NICE guideline CG113: Generalised anxiety disorder and panic disorder in adults: management (published 2011, last updated 15 June 2020) · Source	Guideline n/a			Methods, limits and funding Read: full-text
s-nimh-anxiety NIMH: Anxiety Disorders (last reviewed December 2024) · Source	Health agency n/a			Methods, limits and funding Read: full-text
s-nimh-gad NIMH: Generalized Anxiety Disorder: What You Need to Know (revised 2025) · Source	Health agency n/a			Methods, limits and funding Read: full-text
s-who-anxiety WHO: Anxiety disorders fact sheet (15 September 2026) · Source	Health agency n/a			Methods, limits and funding Read: full-text

What we do not know yet

No placebo or waiting-list arm in the large mindfulness versus escitalopram trial, so it cannot say how much either beat no treatment.
Herb and supplement trials are small and short, usually test one named product, and are not enough to compare products.
No trial was read of cutting down everyday caffeine or alcohol as a treatment for anxiety disorders.
Safety in pregnancy and breastfeeding for several supplements, and interactions of ashwagandha, omega-3, magnesium and L-theanine with anxiety medicines, were not covered by the sources read.
The LiverTox kava entry could not be opened, so kava liver safety rests on NCCIH and the 16-week trial.

How new research is added

New studies are found by a saved PubMed search and placed in a review queue before the page changes.

Lab markers

What changed in blood or saliva tests during trials. A changed number is not the same as feeling or being healthier, and we give no "normal ranges".

Other 3

●

Anxiety symptom scores (clinician-rated and self-rated)

Lower scores with several lifestyle and mind-body approaches; smaller gaps against active treatments in trials

Scores on scales such as the Clinical Global Impression of Severity, completed by a trained rater or by the person.

Moderate certainty

Results and what it means

Why it was measured

To see whether an approach lowers how anxious people feel and how much anxiety interferes with life.

Statistical vs. clinical meaning

Average changes in symptom scores. They do not show cure and do not predict any one person's result.

Study	Result	People	Product
⁸	Clinician-rated anxiety fell by 1.35 points on MBSR and 1.43 on escitalopram over 8 weeks, a difference of -0.07 (95% CI -0.38 to 0.23).	276 adults with an anxiety disorder (208 analysed)	8-week MBSR course vs escitalopram
¹⁷	Response at 12 weeks was 54.2 percent with Kundalini yoga, 70.8 percent with CBT and 33.0 percent with stress education.	226 adults with generalized anxiety disorder	Kundalini yoga, CBT or stress education in groups

^{8 17}

●

Panic attacks after a caffeine challenge

More panic attacks after caffeine in people with panic disorder in trials

Whether a person had a panic attack after being given caffeine or a matching placebo in a blinded test.

Low certainty

Results and what it means

Why it was measured

To see how sensitive people with panic disorder are to caffeine.

Statistical vs. clinical meaning

A short-term lab-style response to large amounts, not a measure of everyday habits.

Study	Result	People	Product
¹⁰	51.1 percent of 237 patients had a panic attack after caffeine and none after placebo; healthy adults were far less affected (1.7 percent versus 53.9 percent of patients).	Adults with panic disorder and healthy comparison adults in 10 studies	One-off caffeine challenge of 400 to 750 mg versus placebo

¹⁰

●

Hamilton Anxiety Rating Scale (HAM-A) total score

Lower with oral lavender oil and chamomile extract than with placebo in short trials in trials

A clinician-rated score of the severity of anxiety symptoms, where lower is better.

Low certainty

Results and what it means

Why it was measured

To compare how much anxiety falls with a treatment or with placebo.

Statistical vs. clinical meaning

Average changes in small, short trials of specific products.

Study	Result	People	Product
⁹	The score fell by 16.0 points (59.3 percent) on lavender oil and 9.5 points (35.4 percent) on placebo over 10 weeks.	221 adults with anxiety disorder not otherwise specified	Standardised oral lavender oil capsule vs placebo
²	The score fell more on chamomile extract than on placebo over 8 weeks (p = 0.047).	57 adults with mild to moderate generalized anxiety disorder	Chamomile extract vs placebo

^{9 2}

Safety

What studies and official sources say about side effects, medicines and who should check first. Each point shows what kind of evidence it comes from.

Main points to know 5

If you are thinking about harming yourself, get help now. If you are thinking about ending your life or harming yourself, or you feel you cannot keep yourself safe, contact your local emergency services or a crisis line in your country right away, or go to the nearest emergency department, and tell someone you trust. You do not have to cope alone. The NHS says to tell someone, and to call the emergency number or go to emergency care if you have seriously harmed yourself or feel you may be about to. NIMH and WHO both note that anxiety disorders raise the risk of suicidal thoughts and behaviours, and that effective help exists. In the United States, NIMH lists the 988 Suicide & Crisis Lifeline; elsewhere, look up your country's service. Health agency guidance ^{29 32 33}
Natural options are not a replacement for care. NIMH says healthy habits can reduce anxiety symptoms when paired with standard care, but this alone cannot replace treatment. NCCIH says natural does not always mean safe, and that a product being labelled standardised does not guarantee quality or consistency. If anxiety interferes with school, work or relationships, NIMH says it is time to seek professional help. Health agency guidance ^{32 23}
Do not stop or change a prescribed medicine on your own. NIMH, NICE and WHO all describe medicines for anxiety that can be effective, and NICE notes some are linked to withdrawal syndromes. Because of this, a prescriber should guide any change. Do not stop, lower or swap a prescribed medicine because a herb or supplement seems to be working, and do not add one without telling your prescriber or pharmacist. Guideline and health agency guidance ^{32 30 33}
Kava has been linked to serious liver injury. NCCIH says various kava products have been linked to rare cases of liver injury, some serious or even fatal, and that the first cases involved alcohol- or acetone-extracted supplements but other cases involved water-based kava drinks. In the 16-week trial in generalized anxiety disorder, liver test abnormalities were significantly more frequent on kava. Anyone with liver disease, or who drinks alcohol or takes other medicines, should talk to a doctor before considering it. Health agency guidance and randomised trial ^{24 16}
St John's wort interacts with many medicines. NCCIH says St John's wort can weaken many prescription medicines, and that combining it with certain antidepressants can cause a potentially life-threatening rise in serotonin. NCCIH also says it is a stimulant that may worsen feelings of anxiety in some people. It is not a studied anxiety treatment on this page, and is listed here only as a warning. Health agency guidance ²⁶

Side effects 7

Effect	How often	Severity and recovery	Evidence
Liver injury with kava NCCIH says suggested causes include undesirable plant varieties or parts, use with alcohol, contamination, and genetic differences between people.	Described as rare by NCCIH. In a 16-week trial of 171 adults with generalized anxiety disorder, liver function test abnormalities were significantly more frequent on kava, but no participant met criteria for herb-induced liver injury.	From abnormal blood tests to serious or fatal liver injury (NCCIH) Not stated in the sources read	Health agency summary and randomised trial ^{24 16}
Other kava side effects Long-term use of high amounts may cause kava dermopathy, a dry, scaly skin condition, according to NCCIH.	Not given. NCCIH lists digestive upset, headache, dizziness, fatigue, muscle aches and trembling; the 16-week trial found memory problems and tremor or shakiness more often on kava.	Mostly mild Not stated	Health agency summary and randomised trial ^{24 16}
Liver injury with ashwagandha Rated probable by a standard causality scale in both new cases. How common this is cannot be told from case reports.	Case reports. A 2023 review gathered all reported cases and added two new ones; one reported case ended in liver transplant.	Hepatocellular and mixed liver injury with itching and dark urine in the two new cases Both new cases improved after the supplement was stopped	Case reports and review ³
Digestive upset, headache or burping with oral lavender oil; allergic skin reactions with skin products A few cases of breast tissue swelling in children using skin products containing lavender were reported; NCCIH says it is unclear whether lavender was the cause.	Not given. NCCIH lists diarrhoea, headache, nausea or burping for oral lavender oil and headache or coughing for aromatherapy.	Mild Not stated	Health agency summary ²⁵
Headache, dizziness and digestive or muscle problems with chamomile NCCIH calls chamomile relatively safe.	Not given. In the two trials read, adverse events were similar to placebo.	Mild Not stated	Health agency summary and randomised trials ^{23 2 13}
Side effects of established medicines, which a clinician can discuss NICE says SSRIs and SNRIs are linked to an increased risk of suicidal thinking and self-harm in a minority of people under 30, so people in that age group are asked to be monitored closely in the first month. Benzodiazepines can lead to tolerance and dependence, which is why NIMH and NICE say they are used only briefly, and WHO says they are generally not recommended.	NIMH lists headaches, nausea and difficulty sleeping with SSRI and SNRI antidepressants; NICE adds a risk of increased anxiety, agitation and sleep problems early on, withdrawal syndromes with some, and bleeding risk with SSRIs in some people.	Usually not severe, according to NIMH Not stated	Guideline and health agency guidance ^{32 30}
Panic attacks and raised anxiety after large amounts of caffeine Smaller amounts were not studied in the review.	51.1 percent of 237 people with panic disorder had a panic attack after a large caffeine challenge, and none after placebo.	Short-term panic attacks Short-lived	Systematic review of blinded challenge studies ¹⁰

Medicines and supplements 5

With	Why it matters	Importance	Evidence	What to do
Prescription medicines (St John's wort) herb	NCCIH says St John's wort can weaken many prescription medicines and lead to a dangerous rise in serotonin when combined with certain antidepressants.	NCCIH lists antidepressants, birth control pills, cyclosporine, digoxin, oxycodone, some HIV and cancer drugs and warfarin as examples. The NCCIH page read is dated December 2017.	Health agency guidance Moderate certainty ²⁶	Do not start St John's wort without asking a doctor or pharmacist, especially if you take any prescribed medicine.
Sedatives, alcohol and benzodiazepines (kava) herb	Additive sedating effects.	NCCIH says kava should not be used together with other substances that have sedative effects, such as benzodiazepines or alcohol, and that people who take any medicine should talk with their health care provider first.	Health agency guidance Moderate certainty ²⁴	Avoid combining kava with alcohol or sedating medicines unless a doctor says otherwise.
Sedative drugs or herbs (lavender) herb	Theoretical additive sedation.	NCCIH says there are theoretical reasons to suspect lavender might interact with some sedative drugs or herbs, which matters before surgery.	Health agency guidance Low certainty ²⁵	Tell your doctor or pharmacist about lavender products you use, particularly before surgery.
Antidepressants and other anxiety medicines (any supplement) prescription	Not established for most supplements.	In the trials read, L-theanine was added to a stable antidepressant without an interaction being reported in the abstract. Interactions of ashwagandha, omega-3 and magnesium supplements with anxiety medicines were not covered by the sources read.	Absence of evidence Not enough evidence ¹⁵	Absence of a study is not proof of safety. Tell your prescriber and pharmacist about every supplement you use.
Alcohol and recreational drugs other	The NHS and WHO say alcohol and drugs can worsen anxiety and contribute to poor mental health.	NIMH notes anxiety disorders commonly occur with alcohol and substance misuse, which can make treatment more complex.	Health agency guidance Moderate certainty ^{27 33 32}	Tell your clinician about alcohol and drug use so that the whole picture is treated.

Who should check first 6

Situation	Advice	Evidence
Pregnancy and breastfeeding	NIMH notes some women experience generalized anxiety disorder during pregnancy and the postpartum period. The NHS says if you are pregnant or gave birth within the last 12 months, your midwife, doctor or health visitor can help you get support. NCCIH says kava may carry special risks in pregnancy and breastfeeding because of harmful pyrone constituents, and that little is known about lavender in pregnancy or breastfeeding. Safety of ashwagandha, chamomile, omega-3, L-theanine and magnesium supplements in pregnancy was not covered by the sources read. A 2023 overview found physical activity helped symptoms in pregnant and postpartum women, among other groups. Speak to a midwife, doctor or pharmacist before using any herb or supplement in pregnancy or while breastfeeding.	Health agency guidance ^{32 28 24 25 18}
Children and teenagers	The trials read for this page were in adults, and the NHS and NICE pages read cover adults. NIMH notes children and teens can have generalized anxiety disorder, often about school, friendships and the future. Herbs and supplements studied in adults should not be assumed safe or effective for children. NICE says SSRIs and SNRIs carry a warning for suicidal thinking in people under 30. Ask a paediatrician or child mental health professional.	Health agency guidance ^{32 30 28}
Liver disease, heavy alcohol use or regular liver-affecting medicines	Kava and ashwagandha have been linked to liver injury in reports. NCCIH links kava liver injury to use with alcohol. People with a liver condition, who drink heavily or who take medicines that can affect the liver should ask a doctor before using either.	Health agency guidance and case reports ^{24 3}
Bipolar disorder, psychosis or schizophrenia (St John's wort)	NCCIH cites case reports of St John's wort worsening psychotic symptoms in people with bipolar disorder or schizophrenia. Anyone with these conditions should talk to their clinician before using any herb.	Health agency guidance ²⁶
Taking prescribed anxiety or antidepressant medicines	See "Do not stop or change a prescribed medicine on your own" above. Add-on supplements need the prescriber's knowledge.	Guideline and health agency guidance ^{30 32 26}
Panic attacks, chest pain or racing heartbeat that are new or severe	NIMH says a provider may do a physical exam to make sure an unrelated physical problem is not causing symptoms. The NHS lists chest pains, a racing heart and breathlessness among symptoms of anxiety, and says a panic attack usually lasts 5 to 20 minutes and is not dangerous. Because the same symptoms can come from physical problems, chest pain that is severe, new or unlike earlier episodes, or that comes with fainting or shortness of breath that does not settle, should be treated as an emergency: contact local emergency services.	Health agency guidance and general safety advice ^{32 27}

Overdose, product quality and regulators

Body	Action	Date
National Institute of Mental Health (NIMH, United States) (United States)	Describes generalized anxiety disorder, says treatment typically involves psychotherapy, medication or both, calls CBT the gold standard psychotherapy, lists SSRI and SNRI antidepressants, benzodiazepines (brief use because of tolerance and dependence) and buspirone as medicines, says healthy habits and stress management can help alongside standard care but not replace it, and lists the 988 Suicide & Crisis Lifeline. source	2025
National Institute for Health and Care Excellence (NICE, England) (United Kingdom)	Guideline CG113 recommends stepped care with the least intrusive, most effective option first, high-intensity CBT or applied relaxation or a drug treatment at step 3, an SSRI as the first drug choice if drug treatment is chosen, benzodiazepines only as a short-term measure during crises, and warnings about suicidal thinking in people under 30 and about withdrawal syndromes. source	2020-06-15
NHS (England) (United Kingdom)	Says a GP will usually advise talking therapies, usually CBT, before medicines, which are usually an SSRI; advises exercise, enough sleep, cutting caffeine and not using alcohol, cigarettes, gambling or drugs to relieve anxiety; and says adults can refer themselves to NHS talking therapies. source	2024-10-22
World Health Organization (WHO) (Global)	Says there are effective treatments, that CBT-based psychological interventions have the most evidence, that SSRIs can be useful, that benzodiazepines are generally not recommended because of dependence and limited long-term effectiveness, and lists self-care steps. source	2026-09-15
National Center for Complementary and Integrative Health (NCCIH, United States) (United States)	Summarises complementary approaches for anxiety, says psychological and physical approaches generally have good safety records, that dietary supplements may have side effects and interactions, and that kava has been linked to serious liver injury. source	2024-09

When to get medical help

Thoughts of harming yourself or ending your life, or feeling you cannot keep yourself safe: contact local emergency services or a crisis line in your country right away, or go to the nearest emergency department. Tell someone you trust.
Chest pain that is severe, new or unlike earlier episodes, fainting, or breathlessness that does not settle: treat as an emergency and contact local emergency services, even if you think it is anxiety.
Anxiety that is affecting school, work, relationships or sleep, or that is not improving with the treatment you have: see a GP or doctor, or a mental health professional (NIMH and NHS).
Anxiety together with heavy drinking, drug use, depression or low mood: tell a clinician so that all of it can be treated together (NIMH).
Itching, dark urine, nausea, stomach pain or yellowing of the skin or eyes while taking kava or ashwagandha: stop it and contact a doctor promptly. These can be signs of liver injury; itching, nausea and dark-coloured urine were reported in the ashwagandha case reports.
Agitation, fast heartbeat, fever or diarrhoea after combining St John's wort with an antidepressant: seek urgent medical help (NCCIH describes these as signs of high serotonin).

Stop using the product and contact a clinician or emergency service. In the US, Poison Control is 1-800-222-1222.

General information only, not medical advice. Ask a clinician or pharmacist before use, especially with medicines, in pregnancy or with a health condition. See the Natural Remedy Safety guide.

Frequently asked questions

What is the difference between normal worry and an anxiety disorder?

NIMH says feeling anxious is a normal part of life. With an anxiety disorder the anxiety does not go away, is felt in many situations and can get worse over time. If it starts to interfere with school, work or relationships, it is time to seek help.

^{31 32}

What types of anxiety disorders are there?

NIMH lists generalized anxiety disorder, panic disorder, social anxiety disorder and phobia-related disorders. WHO also lists agoraphobia, separation anxiety disorder, specific phobias and selective mutism. People can have more than one.

^{31 33}

What are the symptoms of generalized anxiety disorder?

NIMH lists excessive worry about everyday things, trouble controlling worry, feeling on edge, restlessness, trouble concentrating, sleep problems or tiredness, aches and tension, trembling and sweating. The NHS adds palpitations, dizziness and low mood. A list cannot diagnose you.

^{32 28}

How are anxiety disorders evaluated?

A provider asks when symptoms began, how long they have lasted, how often they occur and how they affect your life, and may do a physical exam to rule out an unrelated physical cause. For generalized anxiety disorder NIMH says worry must be hard to control on most days for at least 6 months, with at least three of six listed symptoms.

^{32 28 30}

What does established care look like?

Talking therapy, medicines, or both. CBT is described by NIMH as the gold standard talking therapy. SSRI and SNRI antidepressants are among the medicines named. The NHS says a GP will usually advise talking therapies before medicines. A clinician can explain choices and side effects.

^{32 28 30 33}

Do exercise, mindfulness or yoga help anxiety?

Reviews found a medium drop in anxiety with exercise, and one large trial found an 8-week mindfulness course as good as an antidepressant in adults with anxiety disorders, though it had no placebo group. Kundalini yoga helped but not as much as CBT. They can support care, not replace it.

^{20 18 8 17}

Do lavender, chamomile or ashwagandha help?

A few small trials of lavender oil capsules and chamomile extract found lower anxiety than placebo, and a review of ashwagandha trials found lower anxiety with low certainty. NCCIH says conclusions are not firm, and products differ. Talk to a doctor or pharmacist first, especially if you take medicines or are pregnant.

^{9 13 1 23 25}

Is kava safe for anxiety?

Not automatically. The largest trial in generalized anxiety disorder found no benefit over placebo, and NCCIH says kava has been linked to rare but serious liver injury, and should not be combined with alcohol or sedatives. Ask a doctor before considering it.

^{16 24}

Does caffeine make anxiety worse?

In people with panic disorder, large amounts of caffeine brought on panic attacks in about half of those tested and almost never in healthy adults. The NHS and NIMH advise cutting back on caffeine. Smaller everyday amounts were not studied in that review.

^{10 28 32}

Can I stop my medicine once I feel better?

Do not stop or change a prescribed medicine on your own. Some are linked to withdrawal effects when stopped, so talk with the prescriber about any change. A herb or supplement should never replace it.

^{30 32}

Is it safe to use these options in pregnancy?

Ask a midwife, doctor or pharmacist before using any herb or supplement. NCCIH says kava may carry special risks in pregnancy and breastfeeding, and that little is known about lavender. The NHS says your midwife, doctor or health visitor can help with anxiety during pregnancy or after birth.

^{24 25 28}

When should I get urgent help?

If you are thinking about harming yourself or ending your life, contact your local emergency services or a crisis line in your country now. Treat chest pain that is severe, new or unlike earlier episodes as an emergency. Herbs and supplements are not a replacement for care.

^{29 32 27}

Explore by health goal

Stress Anxiety

Editorial transparency

Written by: AlternativeMed editorial team. Research compiled with AI assistance, then checked against PubMed records.
Medical review: None. This page has not been reviewed by a physician or pharmacist.
First published: 2026-10-31
Last substantive update: 2026-10-31
How we research: We searched PubMed, read the abstract of every review and trial cited, and read the NIMH, NHS, NICE, WHO and NCCIH pages as downloaded text. Full texts were not read. Certainty ratings are the compiler's own reading unless a review gives its own grade (stated in the text). Funding is not stated in most abstracts, and this is noted where relevant.
How we rate evidence: See "How we rate evidence" in the Evidence tab.
Independence: No product is promoted on this page. No affiliate links are shown.

Sources 33

Each source was opened and checked. Where we read only the abstract, it says so. The small numbers in the text point here.

Akhgarjand C, Asoudeh F, Bagheri A et al.. Does Ashwagandha supplementation have a beneficial effect on the management of anxiety and stress? A systematic review and meta-analysis of randomized controlled trials. Phytother Res. 2022. PMID 36017529. DOI 10.1002/ptr.7598. (abstract)
Amsterdam JD, Li Y, Soeller I et al.. A randomized, double-blind, placebo-controlled trial of oral Matricaria recutita (chamomile) extract therapy for generalized anxiety disorder. J Clin Psychopharmacol. 2009. PMID 19593179. DOI 10.1097/JCP.0b013e3181ac935c. (abstract)
Bokan G, Glamočanin T, Mavija Z et al.. Herb-Induced Liver Injury by Ayurvedic Ashwagandha as Assessed for Causality by the Updated RUCAM: An Emerging Cause. Pharmaceuticals (Basel). 2023. PMID 37631044. DOI 10.3390/ph16081129. (abstract)
Boyle NB, Lawton C, Dye L. The Effects of Magnesium Supplementation on Subjective Anxiety and Stress-A Systematic Review. Nutrients. 2017. PMID 28445426. DOI 10.3390/nu9050429. (abstract)
Carpenter JK, Andrews LA, Witcraft SM et al.. Cognitive behavioral therapy for anxiety and related disorders: A meta-analysis of randomized placebo-controlled trials. Depress Anxiety. 2018. PMID 29451967. DOI 10.1002/da.22728. (abstract)
Donelli D, Antonelli M, Bellinazzi C et al.. Effects of lavender on anxiety: A systematic review and meta-analysis. Phytomedicine. 2019. PMID 31655395. DOI 10.1016/j.phymed.2019.153099. (abstract)
Hofmann SG, Sawyer AT, Witt AA et al.. The effect of mindfulness-based therapy on anxiety and depression: A meta-analytic review. J Consult Clin Psychol. 2010. PMID 20350028. DOI 10.1037/a0018555. (abstract)
Hoge EA, Bui E, Mete M et al.. Mindfulness-Based Stress Reduction vs Escitalopram for the Treatment of Adults With Anxiety Disorders: A Randomized Clinical Trial. JAMA Psychiatry. 2023. PMID 36350591. DOI 10.1001/jamapsychiatry.2022.3679. (abstract)
Kasper S, Gastpar M, Müller WE et al.. Silexan, an orally administered Lavandula oil preparation, is effective in the treatment of 'subsyndromal' anxiety disorder: a randomized, double-blind, placebo controlled trial. Int Clin Psychopharmacol. 2010. PMID 20512042. DOI 10.1097/YIC.0b013e32833b3242. (abstract)
Klevebrant L, Frick A. Effects of caffeine on anxiety and panic attacks in patients with panic disorder: A systematic review and meta-analysis. Gen Hosp Psychiatry. 2022. PMID 34871964. DOI 10.1016/j.genhosppsych.2021.11.005. (abstract)
Lee S, Oh JW, Park KM et al.. Digital cognitive behavioral therapy for insomnia on depression and anxiety: a systematic review and meta-analysis. NPJ Digit Med. 2023. PMID 36966184. DOI 10.1038/s41746-023-00800-3. (abstract)
Manzoni GM, Pagnini F, Castelnuovo G et al.. Relaxation training for anxiety: a ten-years systematic review with meta-analysis. BMC Psychiatry. 2008. PMID 18518981. DOI 10.1186/1471-244X-8-41. (abstract)
Mao JJ, Xie SX, Keefe JR et al.. Long-term chamomile (Matricaria chamomilla L.) treatment for generalized anxiety disorder: A randomized clinical trial. Phytomedicine. 2016. PMID 27912875. DOI 10.1016/j.phymed.2016.10.012. (abstract)
Ooi SL, Henderson P, Pak SC. Kava for Generalized Anxiety Disorder: A Review of Current Evidence. J Altern Complement Med. 2018. PMID 29641222. DOI 10.1089/acm.2018.0001. (abstract)
Sarris J, Byrne GJ, Cribb L et al.. L-theanine in the adjunctive treatment of generalized anxiety disorder: A double-blind, randomised, placebo-controlled trial. J Psychiatr Res. 2019. PMID 30580081. DOI 10.1016/j.jpsychires.2018.12.014. (abstract)
Sarris J, Byrne GJ, Bousman CA et al.. Kava for generalised anxiety disorder: A 16-week double-blind, randomised, placebo-controlled study. Aust N Z J Psychiatry. 2020. PMID 31813230. DOI 10.1177/0004867419891246. (abstract)
Simon NM, Hofmann SG, Rosenfield D et al.. Efficacy of Yoga vs Cognitive Behavioral Therapy vs Stress Education for the Treatment of Generalized Anxiety Disorder: A Randomized Clinical Trial. JAMA Psychiatry. 2021. PMID 32805013. DOI 10.1001/jamapsychiatry.2020.2496. (abstract)
Singh B, Olds T, Curtis R et al.. Effectiveness of physical activity interventions for improving depression, anxiety and distress: an overview of systematic reviews. Br J Sports Med. 2023. PMID 36796860. DOI 10.1136/bjsports-2022-106195. (abstract)
Slee A, Nazareth I, Bondaronek P et al.. Pharmacological treatments for generalised anxiety disorder: a systematic review and network meta-analysis. Lancet. 2019. PMID 30712879. DOI 10.1016/S0140-6736(18)31793-8. (abstract)
Stubbs B, Vancampfort D, Rosenbaum S et al.. An examination of the anxiolytic effects of exercise for people with anxiety and stress-related disorders: A meta-analysis. Psychiatry Res. 2017. PMID 28088704. DOI 10.1016/j.psychres.2016.12.020. (abstract)
Su KP, Tseng PT, Lin PY et al.. Association of Use of Omega-3 Polyunsaturated Fatty Acids With Changes in Severity of Anxiety Symptoms: A Systematic Review and Meta-analysis. JAMA Netw Open. 2018. PMID 30646157. DOI 10.1001/jamanetworkopen.2018.2327. (abstract)
Woelk H, Schläfke S. A multi-center, double-blind, randomised study of the Lavender oil preparation Silexan in comparison to Lorazepam for generalized anxiety disorder. Phytomedicine. 2010. PMID 19962288. DOI 10.1016/j.phymed.2009.10.006. (abstract)
. NCCIH: Anxiety and Complementary Health Approaches (Last Updated September 2024). . . (full-text)
. NCCIH: Kava (Last Updated April 2025). . . (full-text)
. NCCIH: Lavender (Last Updated February 2025). . . (full-text)
. NCCIH: St. John's Wort and Depression: In Depth (Last Updated December 2017). . . (full-text)
. NHS: Anxiety, fear and panic (page last reviewed 13 July 2026). . . (full-text)
. NHS: Generalised anxiety disorder (GAD) (page last reviewed 22 October 2024). . . (full-text)
. NHS: Help for suicidal thoughts (page last reviewed 26 April 2024). . . (full-text)
. NICE guideline CG113: Generalised anxiety disorder and panic disorder in adults: management (published 2011, last updated 15 June 2020). . . (full-text)
. NIMH: Anxiety Disorders (last reviewed December 2024). . . (full-text)
. NIMH: Generalized Anxiety Disorder: What You Need to Know (revised 2025). . . (full-text)
. WHO: Anxiety disorders fact sheet (15 September 2026). . . (full-text)

Education, not medical advice. This page is for general information. It does not diagnose or treat any condition and is not a substitute for care from a qualified health professional. Natural does not automatically mean safe. Ask a clinician or pharmacist before changing your care, especially if you take medication, are pregnant or breastfeeding, or have a health condition.