Generalized anxiety disorder (GAD) is treatable. In this review, the clearest financially independent human outcome evidence supports cognitive behavioral therapy (CBT) for persistent worry, with population and endpoint limits. Clinical guidelines also recommend medicines and applied relaxation. Confidence: moderate for the reviewed CBT benefit; low for supplements as GAD treatment. Government guidance and sponsor-linked trials have different evidential roles (Stanley 2009 CBT trial; NICE CG113).
- Persistent, difficult-to-control worry and functional impairment deserve assessment (NIMH: generalized anxiety disorder).
- CBT improved worry in a public-funded trial, but not every anxiety endpoint (Stanley 2009 CBT trial).
- Mindfulness is a possible adjunct; a small trial missed its primary superiority endpoint (Hoge 2013 mindfulness trial).
- Chamomile remains preliminary; kava carries liver-injury risks (Mao 2016 chamomile trial; NCCIH: kava safety).
Table of contents
- Evidence summary
- What it is
- How it works
- The evidence-based treatments
- Supplement and lifestyle evidence
- What works and what does not
- Risks and side effects
- Important interactions
- Who should avoid
- Prescribing information and how to use treatment
- Animal and in-vitro evidence
- Funding and source roles
- Frequently asked questions
- Sources and funding notes
Evidence summary
Clinical recommendations, eligible trial outcomes and conflicted context are separated below. A positive secondary endpoint cannot replace a negative primary endpoint.
| Claim | Evidence reviewed | Source | Funding / conflict | Interpretation / limits |
|---|---|---|---|---|
| CBT reduces worry | 134 older adults; worry-defined response at 3 months: 40.0% versus 21.9% with enhanced usual care. | Stanley 2009 CBT trial | NIMH/VA; no conflicts reported. | GAD-severity outcome was not significantly different; limited generalizability. |
| CBT, relaxation or medicine | Stepped adult treatment guidance. | NICE CG113 | Public body with indirect fee income. | Recommendation, not a new independent trial or universal treatment ranking. |
| Mindfulness | 93 randomized; 89 attended treatment. Primary HAM-A between-group change not significant; some secondary outcomes favored mindfulness. | Hoge 2013 mindfulness trial | Public/foundation support; author commercial ties. | Small, short study; mixed endpoints. |
| Chamomile | 93 initial responders randomized for continuation; relapse hazard ratio 0.52, 95% CI 0.20–1.33. | Mao 2016 chamomile trial | NIH grants; extract-producer supply role. | Primary relapse endpoint not significant; not independent proof. |
| Other complementary approaches | Benefits for situational anxiety do not necessarily treat diagnosed GAD. | NCCIH: anxiety and complementary approaches | NIH education; underlying trial finances not all cleared. | No general supplement replacement established by this review. |
What GAD is
GAD involves excessive worry across ordinary life domains that is hard to control. Adult diagnosis usually requires worry on most days for at least six months, associated symptoms and meaningful distress or impairment. Restlessness, fatigue, poor concentration, irritability, muscle tension and disrupted sleep are common. A screening score supports assessment; it is not a diagnosis (NIMH: generalized anxiety disorder).
How it works
There is no single proven cause. Biology, family vulnerability, experience and environment interact. CBT works with patterns such as threatening interpretations, avoidance and repetitive worry; it does not require a claim that anxiety is simply a chemical imbalance. Treatment outcomes do not prove every proposed brain or stress-hormone mechanism (NIMH: generalized anxiety disorder; Stanley 2009 CBT trial).
The evidence-based treatments
Assessment should include other mental-health conditions, substance use and medical explanations. NICE offers stepped care, progressing from education and structured self-help to CBT, applied relaxation or medication when symptoms persist or impair functioning. An SSRI is a recommended drug option; alternatives depend on response and tolerability. Benzodiazepines are reserved for short crises in NICE guidance. These are clinical recommendations; the drug trials behind them were not individually cleared for independence here (NICE CG113).
Supplement and lifestyle evidence
Sleep regularity, reducing excessive caffeine and appropriate activity can support care. Mindfulness may help some symptoms, but its reviewed trial had indirect author conflicts and mixed results. Chamomile continuation research selected people who already responded; the nonsignificant primary result and supplier role prevent an independent treatment verdict. Tea and standardized trial extract are not interchangeable (NIMH: generalized anxiety disorder; Hoge 2013 mindfulness trial; Mao 2016 chamomile trial).
What works and what does not
CBT has credible evidence for worry reduction in the reviewed older-adult setting. Results should not be presented as a guaranteed cure or extended unchanged to children. This review does not establish routine magnesium, probiotics, omega-3 or herbal combinations as substitutes for GAD care. That is a bounded evidence gap, not proof that every supplement is ineffective. Complementary symptom relief in other populations answers a different question (Stanley 2009 CBT trial; NCCIH: anxiety and complementary approaches).
Risks and side effects
Medicines can cause adverse effects and withdrawal symptoms. Seek urgent help for suicidal intent, inability to stay safe or a serious reaction. Antidepressants require monitoring, particularly when starting or changing treatment. Kava products have been linked to rare severe liver injury, including fatal cases (NCCIH: kava safety; NIMH: mental health medication safety).
Important interactions
Ask a pharmacist to review medicines, alcohol and supplements together. Chamomile has reported interactions with warfarin and some liver-metabolized medicines; sedative interactions are suspected. Kava should not be combined with alcohol or other sedating substances. St. John’s wort can interact with antidepressants and cause serotonin syndrome. Do not abruptly stop prescribed psychiatric medicines (NCCIH: chamomile safety; NCCIH: kava safety; NIMH: mental health medication safety).
Who should avoid
Pregnancy, breastfeeding, liver disease, bipolar symptoms, substance-use problems and multiple medicines require individualized assessment. Chamomile can trigger severe allergy, especially with related-plant allergy; pregnancy safety is uncertain. People at risk of self-harm need prompt clinical support. Adult guideline recommendations should not be applied automatically to children (NCCIH: chamomile safety; NICE CG113).
Prescribing information and how to use treatment
No personalized drug or supplement dose is provided. The clinician should agree goals, review symptoms and function, monitor harms, and plan any discontinuation. A studied extract or class schedule is descriptive, not a self-treatment instruction. Before paying for therapy or an app, check practitioner competence, the treatment model and access to follow-up (NIMH: mental health medication safety; Mao 2016 chamomile trial).
Animal and in-vitro evidence
Animal and cell findings are excluded from the human treatment verdict. Receptor activity, stress-hormone changes and biological plausibility cannot establish remission, daily functioning or long-term safety in people.
Funding and source roles
Who paid for the evidence?
Follow named sources to the funding and relationships disclosed in this article. Numbered article disclosures preserve notes where a source is not identified. A public or university name alone does not establish independence.
View 7 more funding disclosures
This graphic reorganizes the article's disclosures; it is not a new financial audit or independence classification. Highlighted notes show different funding relationships where available. Review funding is separate from underlying trial funding. Disclosure is not proof of falsehood, and no declared conflict is not proof of complete independence.
Therapists, app providers, drug companies, supplement sellers and training publishers can profit from anxiety treatment. These are commercial interests, not allegations about this article’s sources. Study funding and author income are assessed separately. Tier describes financial independence; the credibility grade also considers methods and accountability. Unknown is not independent. Most sources here are US-based, with UK guidance; broader applicability remains uncertain.
| Source | Funding / backers | Country / jurisdiction | Independence | Credibility / incentives / gaps |
|---|---|---|---|---|
| NIMH: generalized anxiety disorder | US federal NIH/HHS funding; budget disclosure. | United States; federal agency | Tier 1 — institutional education | A — Public accountability and clinical accuracy. Political priorities and synthesis choices remain; not trial-level clearance. |
| NICE CG113 | Mainly DHSC grants plus NHS England and fee/research income; 2025–26 accounts. | United Kingdom; public body, London | Tier 2 — indirect institutional ties | B — Transparent recommendations and accountability. Cost-effectiveness remit; industry-facing fee activity; full underlying financial audit not completed. |
| Stanley 2009 CBT trial | NIMH R01-MH53932 and Houston VA HSR&D HFP90-020; authors declared no conflicts. | United States; Houston trial | Tier 1 — checked disclosures | A — Registered randomized comparison and accountable public grants. Older, mostly well-educated sample; therapist/participant blinding impossible. |
| Hoge 2013 mindfulness trial | NIH NCCAM K23AT4432, Highland Street Foundation and Harvard clinical support. Pollack disclosed pharma consulting/equity and interview royalties. | United States; Boston trial | Tier 2 — author indirect ties | B — Active control and blinded raters. Foundation provenance not fully traced; primary anxiety endpoint not superior; small trial. |
| Mao 2016 chamomile trial | NIH grants; Swedish Herbal Institute AB produced/provided extract in the same trial; purchase/gift terms untraced. | US trial; Swedish extract producer | Tier 4 — maker-associated extract supply and preparation support; transaction terms unresolved | D for independent efficacy; methods assessed separately — Randomization and published primary endpoint. Responder-enriched sample; commercial supply role; not cleared as independent. |
| Keefe 2016: same chamomile trial, open-label phase | NIH NCCAM/NCCIH and NCI support; Swedish Herbal Institute produced/provided extract and preparation expertise. | US study; Swedish product supplier | Tier 4 — maker-associated extract supply and preparation support; transaction terms unresolved | D for independent efficacy; methods assessed separately — Transparent companion report identifies supplier. Open-label phase of the same trial, not independent replication; purchase/gift terms untraced. |
| NCCIH: anxiety and complementary approaches | US federal NIH/HHS funding; budget disclosure. | United States; NIH, Bethesda | Tier 1 — institutional education | B — Public safety remit and linked evidence. Complementary-health mission; cited trial finances not individually cleared. |
| NCCIH: chamomile safety | US federal NIH/HHS public funding. | United States; NIH, Bethesda | Tier 1 — safety education | A — Publicly accountable risk information. Reported and theoretical interactions have different certainty. |
| NCCIH: kava safety | US federal NIH/HHS public funding. | United States; NIH, Bethesda | Tier 1 — safety education | A — Publicly accountable harm warnings. Rare harms and product variation complicate risk estimates. |
| NIMH: mental health medication safety | US federal NIH/HHS public funding. | United States; federal agency | Tier 1 — safety education | A — Publicly accountable medication precautions. General information; current product-specific instructions still require checking. |
Frequently asked questions
Can ordinary stress become GAD?
Stress can contribute, but persistent symptoms require assessment rather than self-labeling (NIMH: generalized anxiety disorder).
Can a supplement replace CBT or medicine?
The reviewed evidence does not justify replacement. Discuss any supplement alongside existing care (NCCIH: anxiety and complementary approaches; Mao 2016 chamomile trial).
Does public funding guarantee unbiased results?
No. Outcome choice, missing data, researcher allegiance and author ties still matter. The small mindfulness trial illustrates why its primary and secondary outcomes must be distinguished (Hoge 2013 mindfulness trial).
Sources and funding notes
Original trial methods, results and financial disclosures were checked. Guidance describes professional recommendations; it does not certify every underlying trial as independent. The scorecard gives the source-specific funding and limitations. This is a focused evidence review, not an exhaustive systematic review.
- NIMH: generalized anxiety disorder
- NICE CG113
- Stanley 2009 CBT trial
- Hoge 2013 mindfulness trial
- Mao 2016 chamomile trial
- Keefe 2016: same chamomile trial, open-label phase
- NCCIH: anxiety and complementary approaches
- NCCIH: chamomile safety
- NCCIH: kava safety
- NIMH: mental health medication safety
- NIMH budget disclosure
- NICE 2025–26 accounts
- NCCIH public-budget disclosure
Last reviewed: October 4, 2026. Educational information; diagnosis, treatment selection and monitoring require a qualified clinician.
Have a question — or want us to cover something?
Ask about anything on this page, or request the next deep dive: an ingredient, a supplement, or a health concern. We use published research, evidence syntheses, and regulatory guidance, with clear source links.
One daily research roundup
Get the topics, key findings and links from our new articles in one email. At most one digest a day, only when there is something new.