Key Takeaways
What Helps With Anxiety: What the Trials Agree On
What helps with anxiety, according to the largest bodies of trial evidence, is cognitive behavioral therapy (CBT), antidepressant medication from the SSRI and SNRI classes, or the two together. A 2015 meta-analysis of 234 randomized trials covering 37,333 patients with panic disorder, generalized anxiety disorder (GAD) or social anxiety disorder found that drug, therapy and combined treatments all produced larger improvements than a waitlist[1].
The National Institute of Mental Health (NIMH) describes treatment for GAD as psychotherapy, medication or both, chosen around a person's needs, preferences and medical situation, and notes that finding the right fit can take trial and error[8]. U.S. prevalence figures are on the anxiety statistics page; this article covers what reduces it.
Mindfulness training and exercise also have randomized evidence behind them. Supplements and support groups have thinner evidence, and some fast-acting medications carry dependence risks.
This page compares those options against each other. Step-by-step techniques have their own guides: reducing anxiety in the moment, lowering it over weeks and coping tools.
Anxiety Treatments Compared: Average Improvement in Trials
| Treatment | Pre-post effect size (d) | Study arms |
|---|---|---|
| SNRIs | 2.25 | 23 |
| Benzodiazepines | 2.15 | 42 |
| CBT plus medication | 2.12 | 16 |
| SSRIs | 2.09 | 62 |
| Tricyclic antidepressants | 1.83 | 15 |
| Mindfulness therapies | 1.56 | 4 |
| Relaxation | 1.36 | 17 |
| Individual CBT or exposure | 1.30 | 93 |
| Pill placebo | 1.29 | 111 |
| Group CBT | 1.22 | 18 |
| Psychodynamic therapy | 1.17 | 5 |
| Psychological placebo | 0.83 | 16 |
| Waitlist | 0.20 | 50 |
Average before-and-after improvement (Cohen's d) in randomized trials of panic disorder, generalized anxiety disorder and social anxiety disorder, by treatment arm, from a 2015 meta-analysis of 234 trials[1].
1Pre-post effect sizes measure change within each group, so they include placebo response and natural improvement; the pill placebo arms improved by d = 1.29[1].
2Rows come from different trials and populations, so small gaps between rows are not a reliable ranking.
Sources
- •Bandelow et al., International Clinical Psychopharmacology, 2015
Effect size
An effect size expresses how much symptoms changed in standard-deviation units, so studies that used different anxiety scales can be pooled. Cohen's d, Hedges' g and the standardized mean difference (SMD) are common versions. Reviews cited here describe effects of about 0.4 to 0.6 as medium or moderate[2][5].
Related Terms
Source: Carpenter et al., 2018; Singh et al., 2023
Therapy: CBT and Exposure Have the Most Trials
CBT teaches people to test anxious predictions and to stop avoiding feared situations; exposure therapy is the part that involves facing those situations step by step. In a 2018 meta-analysis of 41 placebo-controlled trials with 2,843 patients, CBT outperformed psychological or pill placebo on the target disorder with a moderate effect (Hedges' g = 0.56)[2].
Patients who received CBT were more likely to respond than those given placebo, with an odds ratio of 2.97[2]. Effects were large for GAD and small to moderate for social anxiety disorder and panic disorder, and individual CBT outperformed group CBT for social anxiety[2]. CBT also had smaller benefits beyond the target disorder, including depression (g = 0.31) and quality of life (g = 0.30)[2].
Other talk therapies have less data. In the 2015 meta-analysis, relaxation averaged a before-and-after effect of d = 1.36 across 17 study arms and psychodynamic therapy d = 1.17 across 5 arms[1]. Mindfulness therapies scored d = 1.56, but from only 4 study arms, which is too few to rank them with confidence[1].
For specific phobias, NIMH describes exposure therapy as a CBT method that is particularly effective[9]. Details on session structure and trial results are in the guides to therapy for anxiety and exposure therapy.
Therapy gains tend to hold. A 2018 meta-analysis of 93 studies found that improvements after psychotherapy were maintained for up to 24 months, and CBT patients kept improving after treatment ended[7].
Medication: Which Anxiety Drugs Outperformed Placebo
The largest network meta-analysis of GAD drugs by its authors' account, published in The Lancet in 2019, pooled 89 trials with 25,441 patients and 22 drugs[3]. Duloxetine, pregabalin, venlafaxine and escitalopram were more effective than placebo and relatively well tolerated[3].
Quetiapine had the largest effect on the Hamilton Anxiety Scale but was poorly tolerated, and paroxetine and benzodiazepines were effective but also poorly tolerated compared with placebo[3]. Mirtazapine, sertraline, fluoxetine, buspirone and agomelatine also looked effective, though those results rested on small samples[3].
The authors concluded that several drug classes work for GAD and that a first medication failing is not by itself a reason to drop medication as a strategy[3]. Across panic, generalized and social anxiety, the 2015 meta-analysis found similar average before-and-after effects for SNRIs (d = 2.25) and SSRIs (d = 2.09), with tricyclic antidepressants lower at d = 1.83[1].
Timing differs by drug. NIMH notes that SSRIs and SNRIs may take several weeks to start working and that buspirone needs 3 to 4 weeks to be fully effective[8].
Benzodiazepines act quickly, but some people build tolerance and become dependent, so NIMH says they may be prescribed only for brief periods[8]. In 2020 the FDA required an updated boxed warning on all benzodiazepines covering abuse, addiction, physical dependence and withdrawal reactions[10]. The FDA estimated 92 million benzodiazepine prescriptions were dispensed in the U.S. in 2019, led by alprazolam (38%), clonazepam (24%) and lorazepam (20%)[10].
Drug-by-drug side effects and dosing are covered in the anxiety medication and SSRI for anxiety guides.
Is Therapy or Medication Better for Anxiety?
Measured by before-and-after change, medications averaged a larger effect than psychotherapies in the 2015 meta-analysis (d = 2.02 versus 1.22)[1]. The same analysis found pill placebo arms improved by d = 1.29, about as much as psychotherapy, so the gap cannot be read as a head-to-head result[1].
Direct comparisons are rarer. The Child/Adolescent Anxiety Multimodal Study (CAMS) randomized 488 children aged 7 to 17 to CBT, sertraline, both, or placebo for 12 weeks[6]. Combination treatment produced the highest response rate, and all three active arms beat placebo[6].
After treatment stops, both approaches appear to hold up. In the 2018 follow-up meta-analysis, patients who had stopped medication stayed stable and did not differ significantly from those who had finished psychotherapy[7].
CAMS Trial: Children Much or Very Much Improved at 12 Weeks
| Treatment arm | Much or very much improved |
|---|---|
| CBT plus sertraline | 80.7% |
| CBT alone | 59.7% |
| Sertraline alone | 54.9% |
| Placebo pill | 23.7% |
Share of 488 children aged 7 to 17 with separation anxiety disorder, GAD or social phobia rated much or very much improved on the Clinical Global Impression-Improvement scale after 12 weeks[6].
1CBT was delivered in 14 sessions; sertraline was dosed up to 200 mg per day[6].
Sources
- •Walkup et al., New England Journal of Medicine, 2008
Children rated much or very much improved after 12 weeks of CBT plus sertraline in the CAMS trial
Combination treatment outperformed CBT alone (59.7%), sertraline alone (54.9%) and placebo (23.7%).
"80.7% for combination therapy"
Mindfulness and Exercise: What Helps Without a Prescription
The first trial to compare a standard mindfulness course with a first-line anxiety drug, according to its authors, was TAME, which randomized 276 U.S. adults with diagnosed anxiety disorders to an 8-week mindfulness-based stress reduction (MBSR) course or escitalopram[4]. Clinician-rated severity fell by 1.35 points with MBSR and 1.43 points with escitalopram, a difference small enough to meet the trial's noninferiority test[4].
Side effects split sharply. At least one study-related adverse event occurred in 78.6% of the escitalopram group and 15.4% of the MBSR group[4]. The most common MBSR side effect was increased anxiety, in 11% of participants[4].
Among people who started treatment, 10 (8%) dropped out of the escitalopram group because of adverse events and none dropped out of the MBSR group for that reason[4]. More on meditation formats is in meditation for anxiety.
Exercise has broad but less precise support. An overview of 97 reviews spanning 1,039 trials and 128,119 participants found physical activity had a medium effect on anxiety compared with usual care (median effect size −0.42)[5]. Most of those reviews were rated critically low in quality, which limits confidence in the exact size of the effect[5].
NIMH reports that reducing caffeine and getting enough sleep can reduce anxiety symptoms when paired with standard care, and that exercise, mindfulness and meditation can enhance the effects of psychotherapy[8]. NIMH also states that a healthy lifestyle alone cannot replace treatment for an anxiety disorder[8].
Supplements, Support Groups and Weaker Evidence
Federal summaries of supplement evidence are more tentative. The National Center for Complementary and Integrative Health (NCCIH) says kava may help anxiety after several weeks but does not appear helpful for GAD symptoms[12].
Kava products have been linked to rare cases of liver injury, some serious or fatal, and NCCIH advises against combining kava with sedatives such as benzodiazepines or alcohol[12]. Trial results for other products are summarized in supplements for anxiety and ashwagandha for anxiety.
Support groups are described by NIMH as helpful for some people with anxiety disorders, with the caution that advice from members does not replace a provider's treatment plan[8]. Psychological placebo control conditions improved by d = 0.83 in the 2015 meta-analysis, well below active therapies[1]. Evidence on acupuncture is covered separately.
What Helps by Type of Anxiety Disorder
| Anxiety type | Therapy with evidence | Medication with evidence |
|---|---|---|
| Generalized anxiety disorder | CBT (large effect vs placebo) | Duloxetine, pregabalin, venlafaxine, escitalopram; buspirone |
| Panic disorder | CBT (small to moderate vs placebo) | SSRIs, SNRIs |
| Social anxiety disorder | Individual CBT over group CBT | SSRIs, SNRIs |
| Specific phobia | Exposure therapy | Anti-anxiety drugs or beta-blockers for specific situations |
Treatments with supporting evidence by disorder, as described in the sources cited in this article[1][2][3][8][9].
1Panic and social anxiety medication rows reflect the SSRI and SNRI trials pooled across those disorders in the 2015 meta-analysis[1].
Sources
- •Carpenter et al., 2018
- •Slee et al., 2019
- •Bandelow et al., 2015
- •NIMH, 2025
How Many People With Anxiety Get Help That Works
Effective treatment exists, but most people with an anxiety disorder do not receive it. The World Mental Health Surveys interviewed 51,547 people in 21 countries and found 9.8% had an anxiety disorder in the past year[11].
Of those, only 41.3% perceived a need for care, 27.6% received any treatment and 9.8% received treatment that met a minimum standard of adequacy[11]. The analysis was published in 2018[11] and draws on surveys conducted years earlier, so current rates may differ.
For U.S. figures by condition, see the pages on generalized anxiety disorder, panic disorder, social anxiety disorder and specific phobias. Cost barriers are covered in affordable therapy.
Frequently Asked Questions
Sources & References
- Bandelow B, Reitt M, Röver C, Michaelis S, Görlich Y, Wedekind D. Efficacy of treatments for anxiety disorders: a meta-analysis. International Clinical Psychopharmacology. Published 2015. Accessed October 2026. https://doi.org/10.1097/YIC.0000000000000078 [Link]
- Carpenter JK, Andrews LA, Witcraft SM, Powers MB, Smits JAJ, Hofmann SG. Cognitive behavioral therapy for anxiety and related disorders: a meta-analysis of randomized placebo-controlled trials. Depression and Anxiety. Published 2018. Accessed October 2026. https://doi.org/10.1002/da.22728 [Link]
- Slee A, Nazareth I, Bondaronek P, Liu Y, Cheng Z, Freemantle N. Pharmacological treatments for generalised anxiety disorder: a systematic review and network meta-analysis. The Lancet. Published 2019. Accessed October 2026. https://doi.org/10.1016/S0140-6736(18)31793-8 [Link]
- Hoge EA, Bui E, Mete M, Dutton MA, Baker AW, Simon NM. Mindfulness-based stress reduction vs escitalopram for the treatment of adults with anxiety disorders: a randomized clinical trial. JAMA Psychiatry. Published 2023. Accessed October 2026. https://jamanetwork.com/journals/jamapsychiatry/fullarticle/2798510 [Link]
- Singh B, Olds T, Curtis R, et al. Effectiveness of physical activity interventions for improving depression, anxiety and distress: an overview of systematic reviews. British Journal of Sports Medicine. Published 2023. Accessed October 2026. https://bjsm.bmj.com/content/57/18/1203 [Link]
- Walkup JT, Albano AM, Piacentini J, et al. Cognitive behavioral therapy, sertraline, or a combination in childhood anxiety. New England Journal of Medicine. Published 2008. Accessed October 2026. https://www.nejm.org/doi/full/10.1056/NEJMoa0804633 [Link]
- Bandelow B, Sagebiel A, Belz M, Görlich Y, Michaelis S, Wedekind D. Enduring effects of psychological treatments for anxiety disorders: meta-analysis of follow-up studies. The British Journal of Psychiatry. Published 2018. Accessed October 2026. https://doi.org/10.1192/bjp.2018.49 [Link]
- National Institute of Mental Health. Generalized Anxiety Disorder: What You Need to Know. National Institute of Mental Health, National Institutes of Health. Published 2025. Accessed October 2026. https://www.nimh.nih.gov/health/publications/generalized-anxiety-disorder-gad [Link]
- National Institute of Mental Health. Phobias and Phobia-Related Disorders. National Institute of Mental Health, National Institutes of Health. Published 2025. Accessed October 2026. https://www.nimh.nih.gov/health/publications/phobias-and-phobia-related-disorders [Link]
- U.S. Food and Drug Administration. FDA requiring Boxed Warning updated to improve safe use of benzodiazepine drug class. Published 2020. Accessed October 2026. https://www.fda.gov/drugs/drug-safety-and-availability/fda-requiring-boxed-warning-updated-improve-safe-use-benzodiazepine-drug-class [Link]
- Alonso J, Liu Z, Evans-Lacko S, et al. Treatment gap for anxiety disorders is global: results of the World Mental Health Surveys in 21 countries. Depression and Anxiety. Published 2018. Accessed October 2026. https://doi.org/10.1002/da.22711 [Link]
- National Center for Complementary and Integrative Health. Kava: Usefulness and Safety. National Center for Complementary and Integrative Health, National Institutes of Health. Published 2025. Accessed October 2026. https://www.nccih.nih.gov/health/kava [Link]