Ashwagandha Meta-Analysis Anxiety Evidence

---

Table of Contents

1. What This Article Covers
2. Why Meta-Analyses Matter More Than Single Studies
3. The 2022 Dose-Response Meta-Analysis
4. The 2024 Pooled Analysis: Stress, Anxiety, and Cortisol
5. The 2025 BJPsych Open Systematic Review
6. The 2025 Bayesian Confirmation
7. The 2026 PubMed-Indexed Meta-Analysis
8. What the Numbers Actually Mean: Interpreting SMD and Heterogeneity
9. Does Ashwagandha Lower Cortisol Too?
10. Dosing: What Doses Were Used Across Studies?
11. How Long Does It Take to Work?
12. KSM-66 and Extract Standardization: Does It Matter?
13. Safety, Side Effects, and Interactions
14. Can Ashwagandha Replace Prescription Anxiety Treatment?
15. Quality of Evidence: Limitations You Need to Know
16. Summary: What the Overall Ashwagandha Anxiety Evidence Shows
17. Frequently Asked Questions

---

What This Article Covers

If you have spent any time researching adaptogens, you have almost certainly come across bold claims about ashwagandha reducing anxiety, lowering stress hormones, and improving sleep. The problem is that most of those claims link back to individual small studies — not to the broader body of evidence.

This article is different. It focuses entirely on ashwagandha meta-analysis anxiety evidence: what the pooled data actually shows, what the numbers mean in plain language, what the limitations are, and what an honest, evidence-based consumer should take away from the research as of 2025–2026.

We pull from the top peer-reviewed sources currently indexed in PubMed and published in journals including BJPsych Open, so you can evaluate this evidence for yourself rather than taking marketing copy at face value.

We will cover:

• Every major ashwagandha anxiety systematic review and meta-analysis published between 2022 and 2026
• The statistical effect sizes, confidence intervals, and heterogeneity figures in plain English
• What doses and extract types were used across trials
• The honest limitations and quality concerns researchers themselves flag
• Practical takeaways about safety, timing, and realistic expectations

Let's start with the basics.

---

Why Meta-Analyses Matter More Than Single Studies

A single randomized controlled trial (RCT) testing ashwagandha might enroll 60 participants over eight weeks at one clinical site. A positive result from that trial is interesting — but it could reflect a fluke, a biased sample, a placebo response, or investigator enthusiasm. Single studies, even well-designed ones, get overturned all the time.

A systematic review takes a different approach. Researchers define a search strategy in advance, hunt down every relevant trial they can find using strict inclusion criteria, and assess each study's risk of bias. They synthesize the evidence qualitatively — meaning they describe what each study found and whether the designs are comparable.

A meta-analysis goes one step further. It pools the numerical results from multiple studies and calculates a single summary estimate with a confidence interval. This gives you more statistical power than any single trial, and it lets you see whether the effect is consistent or whether it varies wildly depending on the study.

When you see a result like SMD −1.55 (95% CI −2.37 to −0.74), that is not from one study. That is a weighted average across multiple trials, expressing how much ashwagandha moved anxiety scores compared to placebo in standard deviation units. We will explain exactly what that means throughout this article.

The bottom line: ashwagandha anxiety meta-analysis data is the strongest class of evidence available for evaluating whether this supplement works, and there is now a meaningful accumulation of it.

---

The 2022 Dose-Response Meta-Analysis

The 2022 systematic review and dose-response meta-analysis represents one of the early rigorous attempts to pool the ashwagandha anxiety evidence base in a statistically disciplined way.

What It Found

The pooled analysis found that ashwagandha significantly reduced anxiety compared to placebo, with a standardized mean difference (SMD) of −1.55 (95% CI: −2.37 to −0.74), and a p-value of 0.005.

In plain terms:

• The effect was in the right direction (lower anxiety scores with ashwagandha vs. placebo)
• The result was statistically significant (p = .005 means there is less than a 0.5% chance this result occurred by chance alone if there were no true effect)
• The effect size of −1.55 is considered large by conventional standards (Cohen's d > 0.8 is typically classified as large)

The Heterogeneity Problem

Here is where intellectual honesty matters. The I² statistic was 93.8%, which is very high.

I² measures the percentage of variation across studies that is due to true differences between studies (rather than random sampling error). An I² of 93.8% means the studies are telling substantially different stories. Some are showing large effects, others smaller ones, and combining them into a single number risks oversimplifying a messier reality.

Quality of Evidence

The researchers themselves rated the certainty of evidence as low. This is critical context that marketing materials never mention. Low certainty does not mean the effect is fake — it means we cannot yet be confident the true effect size is as large as the pooled estimate suggests, because of methodological limitations in the included studies (small sample sizes, short durations, varied outcome measures, and risk of bias).

Takeaway from 2022: Promising signal, statistically significant, but limited by high heterogeneity and low-certainty evidence overall. This is a foundation to build on, not a final answer.

---

---

The 2024 Pooled Analysis: Stress, Anxiety, and Cortisol

The 2024 ashwagandha anxiety pooled analysis included 9 studies with 558 participants and represented a meaningful expansion of the evidence base. Unlike some earlier reviews that focused narrowly on anxiety alone, this meta-analysis examined three interconnected outcomes simultaneously: perceived stress, anxiety scores on the Hamilton Anxiety Scale (HAM-A), and cortisol levels.

Anxiety Results

On the Hamilton Anxiety Scale, ashwagandha produced a mean difference of −2.19 (95% CI: −3.83 to −0.55) compared to placebo.

The HAM-A is a clinician-administered scale where lower scores indicate less anxiety. A drop of 2.19 points on a scale that typically ranges from 0 to 56 is statistically significant (given that the confidence interval does not include zero), though the clinical meaning of a ~2-point change is modest and context-dependent.

Stress Results

On the Perceived Stress Scale (PSS), the mean difference was −4.72 (95% CI: −8.45 to −0.99). The PSS is a widely used self-report tool measuring how unpredictable, uncontrollable, or overloaded respondents perceive their lives. A nearly 5-point drop is clinically more meaningful on this scale and falls within ranges that researchers typically consider a noticeable subjective improvement.

Cortisol Results

We cover this more fully in the cortisol section below, but the 2024 meta-analysis also found a significant reduction in cortisol: MD −2.58 (95% CI: −4.99 to −0.16).

Why This Meta-Analysis Matters

The 2024 ashwagandha systematic review anxiety findings are important because they:

1. Confirm the anxiety-reducing signal from 2022 in a larger pooled sample
2. Show that the anxiety effect is accompanied by measurable cortisol reduction — suggesting a biologically plausible mechanism
3. Use a validated clinical anxiety measure (HAM-A) rather than purely self-report tools

The evidence is converging: across multiple research groups using different methodologies, ashwagandha consistently moves anxiety and stress scores in the right direction versus placebo.

---

The 2025 BJPsych Open Systematic Review

Published in BJPsych Open — a peer-reviewed journal of the Royal College of Psychiatrists — the 2025 systematic review and meta-analysis represents one of the most rigorous and comprehensive pieces of ashwagandha systematic evidence anxiety research to date.

The full paper is available at the Cambridge University Press platform and is indexed in PubMed (PMID: 39348746).

Study Design

This ashwagandha anxiety systematic review conducted a comprehensive literature search, ultimately including ten studies for the anxiety outcome analysis. The researchers evaluated both a pre-treatment comparison and an 8-week outcome comparison, allowing them to examine how anxiety changes over time with supplementation.

Anxiety Results: Before Treatment

At pre-treatment assessment, ashwagandha groups showed an anxiety improvement of μ = −1.55 (95% CI: −2.45 to −0.65, p = 0.0007). This statistically significant finding at baseline is somewhat unusual and may reflect baseline differences between groups or the early onset of effect.

Anxiety Results: At 8 Weeks

By the 8-week mark, the effect had grown substantially: μ = −3.52 (95% CI: −6.00 to −1.04, p = 0.0053). This suggests that anxiety benefits accumulate over time rather than appearing immediately — an important practical point for people evaluating whether to continue supplementation.

The Pooled SMD Across Ten Studies

Across all ten studies included in the anxiety analysis, the pooled SMD was −1.62 (95% CI: −2.66 to −0.57), with an I² of 94%.

This effect size of −1.62 is large by any conventional benchmark. It is also strikingly consistent with the 2022 result (SMD −1.55) despite being drawn from a different set of included trials analyzed by a different research team. That convergence across independent meta-analyses is scientifically meaningful.

High Heterogeneity, Again

The I² of 94% echoes the 2022 finding. The authors acknowledge this as a significant limitation, noting that variation in study populations, doses, extract types, outcome measures, and trial durations all contribute to statistical heterogeneity. This is not a reason to dismiss the evidence — but it is a reason to interpret any single pooled number with appropriate caution.

---

The 2025 Bayesian Confirmation

One of the more methodologically sophisticated contributions in the recent ashwagandha anxiety overall evidence literature comes from the Bayesian analysis conducted within the 2025 paper focused on individuals with mental disorders.

Why Bayesian Analysis Matters

Traditional frequentist meta-analysis (which produces p-values and confidence intervals) is powerful but has known limitations. It treats each analysis as if there were no prior knowledge about the outcome — which is not quite true when you already have a growing body of ashwagandha RCTs.

Bayesian analysis allows researchers to incorporate prior evidence into the estimate and quantifies the result as a credible interval (CrI) rather than a confidence interval. It also handles heterogeneity across studies in a more flexible way.

The Bayesian Result

The Bayesian analysis confirmed a statistically significant anxiety-reducing effect: SMD −1.23 (95% Credible Interval: −2.19 to −0.12).

This is a somewhat more conservative estimate than the frequentist meta-analyses (which cluster around −1.55 to −1.62), which is expected — Bayesian analysis with weakly informative priors tends to "shrink" extreme estimates toward the center. But even this more conservative figure is:

• A large effect size (SMD > 1.0)
• Statistically significant (the CrI does not include zero)
• Confirmed by an independent analytical framework

The Bayesian confirmation matters because it provides convergent validity. The anxiety-reducing signal from ashwagandha is not an artifact of any single analytical approach.

The 600 mg/day Threshold

This 2025 analysis also found that the significant anxiety benefit was associated with dosing of at least 600 mg/day. Studies using lower doses did not consistently show the same effect. This is an important practical data point that we return to in the dosing section.

---

---

The 2026 PubMed-Indexed Meta-Analysis

The most recent large-scale ashwagandha meta analysis in the anxiety space was indexed in PubMed in 2026 and reported effect sizes that are notably larger than those from earlier meta-analyses.

The Findings

The 2026 meta-analysis reported:

• Anxiety SMD: −6.87 (95% CI: −8.77 to −4.97)
• Stress SMD: −5.88 (95% CI: −8.15 to −3.60)
• Depression SMD: −5.68 (95% CI: −8.43 to −2.94)

These are extremely large effect sizes — far larger than those reported in earlier meta-analyses that clustered around SMD −1.5 to −1.6.

Why the Discrepancy?

When a meta-analysis reports effect sizes dramatically larger than its predecessors, the most important question is: why?

Several factors could explain this:

1. Inclusion of different outcome measures: If the 2026 meta-analysis included studies using outcome tools that are more sensitive to change (or more prone to floor/ceiling effects), the SMDs will be larger.
2. Inclusion of higher-dose trials: If newer trials used higher doses and produced larger effects, the pooled estimate rises.
3. Publication bias: If only positive studies are published and included, pooled estimates are inflated upward.
4. Heterogeneous populations: Mixing clinical populations with severe anxiety and healthy volunteers in the same analysis can distort effect sizes.
5. Outcome measure variability: The 2026 meta-analysis covers stress, anxiety, and depression together, which may reflect different measurement tools with different sensitivity profiles.

At this writing, the 2026 figures should be treated as a signal worth monitoring but interpreted cautiously given how far they depart from earlier estimates. Replication and methodological scrutiny will be needed before these numbers anchor any clinical recommendations.

What remains clear across all meta-analyses: the direction is consistent — ashwagandha reduces anxiety compared to placebo — and the effect is statistically significant. The true magnitude remains uncertain.

---

What the Numbers Actually Mean: Interpreting SMD and Heterogeneity

Standardized Mean Difference (SMD)

Because different studies use different anxiety scales (HAM-A, PSS, GAD-7, DASS-21, etc.), researchers cannot simply average raw score changes. Instead, they divide each study's mean difference by its standard deviation, producing a unitless number called the SMD (also known as Cohen's d or Hedges' g in most meta-analyses).

Conventional benchmarks:

• SMD 0.2 = small effect
• SMD 0.5 = medium effect
• SMD 0.8 = large effect
• SMD > 1.0 = very large effect

The ashwagandha anxiety SMDs ranging from −1.23 (Bayesian estimate) to −1.62 (BJPsych Open) are therefore in the "very large" category by conventional standards. Even the more conservative Bayesian estimate clears the "large" threshold.

However, very large SMDs in supplement research are a yellow flag — not necessarily a red flag, but a prompt for scrutiny. Drug trials for well-established anxiety medications like SSRIs typically show SMDs in the 0.3–0.5 range. An SMD of 1.5+ from a botanical supplement warrants asking whether every included study is high quality.

Confidence Intervals

A 95% confidence interval (CI) tells you the range within which the true population effect most likely falls. A CI of −2.37 to −0.74 means you can be 95% confident the true SMD is somewhere in that range. When a CI does not cross zero, the result is statistically significant.

Wider CIs indicate more uncertainty. Compare:

• SMD −1.55 (95% CI: −2.37 to −0.74) → moderately wide, reasonable certainty
• SMD −6.87 (95% CI: −8.77 to −4.97) → wide but consistently large, suggesting meaningful effect but substantial variability

I² and Heterogeneity

I² quantifies the percentage of variation between studies that is due to true differences rather than chance.

• I² < 25%: low heterogeneity (studies broadly agree)
• I² 25–75%: moderate heterogeneity
• I² > 75%: high heterogeneity (studies vary substantially)

The consistent I² of 93–94% across multiple ashwagandha meta-analyses means the individual trials are telling quite different stories about effect size. This is not unusual for supplement research — it typically reflects genuine variation in dose, population, extract quality, duration, and outcome tool. But it does mean the single pooled number should be treated as a rough central estimate, not a precise truth.

---

Does Ashwagandha Lower Cortisol Too?

One of the most asked reader questions in this space is whether ashwagandha reduces cortisol alongside anxiety — and whether the cortisol reduction might actually explain the anxiety benefit.

What the Meta-Analytic Evidence Shows

The 2024 meta-analysis found a significant cortisol reduction: MD −2.58 nmol/L (95% CI: −4.99 to −0.16). This is a statistically significant finding, though the lower bound of the confidence interval (−0.16) is close to zero, suggesting the effect may be small in absolute terms.

The 2025 BJPsych Open systematic review also examined cortisol outcomes and found evidence supporting cortisol reduction with ashwagandha supplementation.

Interestingly, a separate 2025 systematic review and meta-analysis on perceived stress found no significant effect on perceived stress in its pooled analysis — but did find a significant cortisol reduction. This dissociation is intriguing: it suggests ashwagandha might affect biological stress markers (cortisol) more reliably than subjective stress perception, or it may reflect differences in the populations and measurement tools used across studies.

The Mechanism Question

Ashwagandha (Withania somnifera) contains withanolides — steroidal lactones — that may modulate the hypothalamic-pituitary-adrenal (HPA) axis, the body's central stress regulation system. Cortisol is the primary output hormone of the HPA axis when you are stressed. If ashwagandha dampens HPA reactivity, lower cortisol would be expected to accompany reduced anxiety — which is broadly consistent with what the meta-analyses show.

However, the mechanism remains incompletely understood. Preclinical evidence from animal models supports HPA-axis modulation, GABA-A receptor activity, and anti-inflammatory pathways, but these have not been definitively established as the primary drivers of the clinical anxiety effect in humans.

Practical Implication

If you are evaluating ashwagandha for stress or anxiety, the cortisol data suggests the benefit may go beyond pure placebo or self-report bias. The existence of an objective biomarker (cortisol) moving in the expected direction strengthens confidence in the overall signal.

---

Dosing: What Doses Were Used Across Studies?

The included trials across the major ashwagandha anxiety systematic review papers used a range of doses, but the emerging picture is relatively consistent.

The 600 mg/day Threshold

The 2025 meta-analysis focused on individuals with mental disorders specifically found that significant anxiety benefits were associated with at least 600 mg/day of ashwagandha extract. Studies using lower doses did not show consistent effects.

This is an important finding because:

• Many consumer products contain 300–500 mg per capsule or serving
• Labeling does not always specify whether the stated dose refers to whole root powder or a standardized extract (which has a very different potency profile)
• Standardized extracts (like KSM-66 at 5% withanolides) at 600 mg are not equivalent to 600 mg of generic ashwagandha root powder

Dose Ranges Across Trials

Across the meta-analyzed studies, doses generally ranged from 300 mg/day to 1,000 mg/day of ashwagandha extract. The most commonly studied doses were 300 mg twice daily (600 mg total) and 600 mg once daily.

Some trials split doses (morning and evening), while others used a single daily dose. The evidence does not yet clearly distinguish whether splitting is superior.

Extract vs. Root Powder

Multiple included studies used standardized extracts rather than raw root powder. This matters because:

• Standardized extracts have a defined withanolide content (typically 5–10%)
• Root powder withanolide content varies with growing conditions, harvest timing, and processing
• The clinical evidence is stronger for standardized extracts than for non-standardized forms

---

How Long Does It Take to Work?

Based on the trial designs included in the meta-analyses, here is what the data suggests about timing:

4–8 Weeks for Measurable Effects

Most trials ran for 8–12 weeks. The 2025 BJPsych Open meta-analysis specifically compared pre-treatment to 8-week outcomes and found the effect grew over time:

• Pre-treatment: μ = −1.55 (modest early effect)
• At 8 weeks: μ = −3.52 (substantially larger effect)

This time-dependent pattern suggests that ashwagandha is not an immediate anxiolytic in the way that a benzodiazepine is. The effect appears to accumulate with consistent use.

What This Means Practically

People who try ashwagandha for one week and notice nothing are not necessarily non-responders. The evidence base supports continued use for at least 4–8 weeks before evaluating efficacy.

There is also some preliminary evidence from shorter trials (4 weeks) showing early effects on cortisol and perceived stress, which may precede improvements on formal anxiety scales.

---

KSM-66 and Extract Standardization: Does It Matter?

KSM-66 is a proprietary ashwagandha root extract developed by Ixoreal Biomed, standardized to ≥5% withanolides. It is one of the most clinically studied ashwagandha extracts available, and several of the trials included in the major meta-analyses used KSM-66 specifically.

KSM-66 Anxiety Evidence Synthesis

The KSM-66 anxiety evidence synthesis draws from multiple double-blind, placebo-controlled RCTs. In the included trials, KSM-66 at 300–600 mg/day consistently showed statistically significant anxiety reductions versus placebo.

The KSM-66 meta-analysis contribution to the broader ashwagandha evidence base is substantial — because KSM-66 trials typically have better-defined extract composition than studies using uncharacterized root powder, they introduce less noise into the pooled analysis.

KSM-66 vs. Sensoril vs. Root Powder

A separate proprietary extract, Sensoril (from root and leaf), has also been studied, though with fewer large RCTs. Root powder studies are included in meta-analyses but generally show more variable results.

The KSM-66 meta-analysis subgroup findings, where reported, tend to show effect sizes in the range of the overall pooled estimates — suggesting that standardization to withanolide content, rather than any proprietary "magic," is likely what drives consistency.

The Practical Implication

When evaluating an ashwagandha product, look for:

1. A standardized extract (KSM-66, Sensoril, or equivalent)
2. A stated withanolide content (typically 5% minimum)
3. A dose of at least 600 mg/day of the standardized extract
4. Third-party testing for purity and heavy metals

---

---

Safety, Side Effects, and Interactions

The anxiety benefits from the ashwagandha anxiety overall evidence are meaningful — but any honest evaluation must include the safety picture.

What the Clinical Trials Show

Across the trials included in the meta-analyses, ashwagandha was generally well tolerated at doses of 300–1,000 mg/day for up to 12 weeks. The most commonly reported side effects in RCTs include:

• Gastrointestinal upset (nausea, loose stools, stomach discomfort) — typically mild and dose-dependent
• Drowsiness or sedation — reported in some participants, particularly at higher doses
• Headache — occasional, usually early in supplementation

Serious adverse events were rare in clinical trial populations.

Liver Safety Concerns

Case reports in the medical literature have described rare instances of drug-induced liver injury (DILI) associated with ashwagandha supplementation. While causality is not definitively established in all cases, the signal exists and should not be dismissed.

People with pre-existing liver conditions should avoid ashwagandha or use it only under medical supervision. Periodic liver function monitoring may be appropriate for long-term users, though no formal guideline currently mandates this.

Thyroid Interactions

Ashwagandha has been shown in some studies to increase thyroid hormone levels (T3 and T4). This is concerning for:

• People with hyperthyroidism or Graves' disease
• People on thyroid hormone replacement (levothyroxine) — the interaction could lead to over-medication
• People scheduled for thyroid testing — supplementation may skew results

Interactions With CNS-Active Medications

Given ashwagandha's potential GABA-A receptor activity and sedative effects, there is theoretical concern about additive sedation with:

• Benzodiazepines (diazepam, lorazepam, etc.)
• Barbiturates
• Sleep medications (zolpidem, eszopiclone)
• Alcohol
• Some anticonvulsants

These interactions have not been rigorously studied in humans, so clinical caution is warranted.

Contraindications

• Pregnancy: Ashwagandha has traditional use as an abortifacient and should be avoided during pregnancy
• Autoimmune conditions: Ashwagandha may stimulate immune activity, which is potentially problematic in rheumatoid arthritis, lupus, or multiple sclerosis
• Active thyroid disease: As above
• Scheduled surgery: Discontinue at least 2 weeks prior due to potential CNS-depressant interactions with anesthesia

Long-Term Safety

Most RCTs run for 8–12 weeks. The long-term safety profile beyond 3 months is not well characterized in clinical trial data. Observational data and traditional use suggest reasonable tolerability, but the meta-analytic evidence base does not extend to long-term use.

---

Can Ashwagandha Replace Prescription Anxiety Treatment?

This is one of the most important questions in the ashwagandha anxiety evidence overall discussion, and it deserves a direct and honest answer.

The Short Answer: No

The ashwagandha systematic review anxiety literature does not support using ashwagandha as a replacement for prescription anxiety medications in people with diagnosable anxiety disorders. Here is why:

1. The study populations are mostly subclinical The majority of RCTs included in the meta-analyses enrolled healthy adults with elevated perceived stress or mild-to-moderate anxiety symptoms — not people meeting DSM-5 criteria for Generalized Anxiety Disorder, Panic Disorder, Social Anxiety Disorder, or PTSD. The evidence that ashwagandha helps people with diagnosed anxiety disorders is much more limited.

2. Head-to-head comparisons do not exist There are no adequately powered trials directly comparing ashwagandha to SSRIs, SNRIs, or benzodiazepines for anxiety treatment. Without such comparisons, it is impossible to say whether ashwagandha's effect size is comparable to established treatments.

3. The effect size context SSRIs like sertraline or escitalopram for anxiety disorders typically show SMDs in the 0.3–0.6 range in meta-analyses of large, rigorously conducted trials. Ashwagandha shows larger SMDs in meta-analyses, but those meta-analyses have high heterogeneity, variable quality studies, and mostly subclinical populations — meaning the comparison is not apples to apples.

4. Discontinuation carries risk If someone discontinues established anxiety medication to try ashwagandha and experiences a relapse or deterioration, the consequences can be severe. This is not a theoretical concern — anxiety disorders are serious conditions with functional and quality-of-life implications that require evidence-based management.

Where Ashwagandha May Fit

For people with subclinical stress and anxiety who do not meet criteria for a clinical disorder, who are not on medications, and who want to support their stress response through lifestyle and supplementation, the evidence from ashwagandha anxiety meta-analysis research is genuinely supportive.

It may also be considered as a complementary approach alongside standard care for people whose clinicians are comfortable with it — but this should be disclosed to and coordinated with a healthcare provider, particularly given the interaction concerns above.

---

Quality of Evidence: Limitations You Need to Know

Any fair-minded review of the ashwagandha meta-analysis anxiety evidence must acknowledge its limitations clearly. Here are the main ones:

1. High Heterogeneity Across All Meta-Analyses

As noted repeatedly, I² values of 93–94% appear across multiple independent meta-analyses. This tells us the included trials are measuring something real, but the magnitude of the effect varies enormously depending on who is being studied, what dose they took, how long they took it, and how anxiety was measured.

2. Small Individual Trial Sizes

Many of the included RCTs enrolled fewer than 100 participants. Small trials are more susceptible to random variation and tend to show larger effect sizes than larger, more adequately powered trials. This "small study effect" is a well-known driver of inflated SMDs in meta-analyses.

3. Short Trial Duration

Most trials ran 8–12 weeks. The long-term efficacy and safety of ashwagandha for anxiety is not established from the clinical trial literature.

4. Publication Bias Risk

Positive studies are more likely to be published than negative ones (publication bias). If a handful of unpublished negative trials exist, the true pooled effect would be smaller than what the published meta-analyses show. Funnel plot asymmetry tests in the reviewed meta-analyses suggest some publication bias may be present, though results are inconclusive.

5. Outcome Measure Variability

The studies use a range of anxiety measures: HAM-A (clinician-rated), GAD-7 (self-report), PSS (perceived stress, not strictly anxiety), DASS-21 (depression/anxiety/stress composite), and others. Pooling these different instruments contributes to heterogeneity and makes interpretation of the pooled SMD harder.

6. Extract and Dose Variability

Not all ashwagandha is the same. The meta-analyses pool studies using KSM-66, Sensoril, uncharacterized root powder, and root-plus-leaf combinations. This extract variability inflates heterogeneity and makes it difficult to assign the clinical evidence to any single product form.

7. The 2022 Certainty of Evidence Rating

The 2022 meta-analysis rated the certainty of evidence as low using formal GRADE methodology. The BJPsych Open 2025 paper also acknowledges methodological limitations. These are not arbitrary hedges — they reflect real gaps in the evidence base that subsequent research needs to address.

---

Summary: What the Overall Ashwagandha Anxiety Evidence Shows

Here is the honest, complete picture from the ashwagandha anxiety overall evidence as of 2025–2026:

What We Can Say With Reasonable Confidence

✅ Ashwagandha reduces anxiety scores compared to placebo in randomized controlled trials. This finding is replicated across multiple independent meta-analyses, using different included studies, different analytical methods, and even different statistical frameworks (frequentist and Bayesian).

✅ The effect is statistically significant across all major meta-analyses, with p-values well below 0.05 and confidence/credible intervals that do not include zero.

✅ The effect size is large by conventional benchmarks, with pooled SMDs ranging from −1.23 (Bayesian) to −1.62 (BJPsych Open frequentist) across the most methodologically rigorous analyses.

✅ Cortisol is also reduced, providing an objective biomarker to complement the self-reported anxiety improvements.

✅ The effect appears dose-dependent, with 600 mg/day of standardized extract emerging as a clinically relevant threshold.

✅ Effects accumulate over 4–8 weeks, meaning ashwagandha is a supplement requiring consistent, sustained use — not an acute intervention.

✅ KSM-66 and other standardized extracts have the strongest individual trial evidence.

What We Cannot Yet Say

❌ The certainty of evidence remains low to moderate due to high heterogeneity, small trial sizes, and risk of bias.

❌ Long-term efficacy and safety are not established beyond 3 months from RCT data.

❌ Ashwagandha's efficacy in diagnosed anxiety disorders (versus subclinical stress and anxiety) has not been established.

❌ Head-to-head comparisons with established anxiety medications do not exist.

❌ The dramatically larger effect sizes in the 2026 meta-analysis need independent replication before they can anchor clinical guidance.

The Bottom Line

The ashwagandha meta-analysis anxiety evidence is genuine, replicable, and growing. This is not a supplement where researchers keep finding nothing. The convergence of results across independent research teams, using different analytical approaches, is meaningful.

At the same time, "large SMD in a meta-analysis of predominantly subclinical stress trials with high heterogeneity and low certainty of evidence" is not the same as "proven highly effective treatment for anxiety disorders." The honest position is: the evidence supports ashwagandha as a useful supplement for stress and anxiety in generally healthy adults, particularly at ≥600 mg/day of standardized extract, taken consistently for at least 8 weeks — while recognizing that the true effect size remains uncertain and clinical-level anxiety disorders require evidence-based medical treatment.

---

---

Frequently Asked Questions

Does ashwagandha actually reduce anxiety?

Based on the ashwagandha anxiety meta-analysis evidence, yes — the pooled data from multiple independent systematic reviews shows statistically significant and large reductions in anxiety scores compared to placebo. The 2025 BJPsych Open meta-analysis (ten studies, SMD −1.62), the 2024 pooled analysis (nine studies, HAM-A MD −2.19), and the 2022 dose-response meta-analysis (SMD −1.55) all point in the same direction. A 2025 Bayesian analysis confirmed the effect at SMD −1.23. The consistency across studies and analytical frameworks is notable.

How strong is the evidence from meta-analyses?

Strong enough to take seriously, but not without caveats. The meta-analyses are methodologically sound, but the certainty of evidence has been rated as low in formal GRADE assessments due to high heterogeneity (I² ~94%), small individual trial sizes, and risk of bias. The signal is real; the precise magnitude is uncertain.

What dose was used in the studies?

Most trials used 300–1,000 mg/day of standardized ashwagandha extract, with 600 mg/day being the most common effective dose. The 2025 meta-analysis found that at least 600 mg/day was necessary for consistent anxiety benefit. Doses should refer to standardized extract (e.g., KSM-66 at 5% withanolides), not raw root powder.

How long does it take to work?

Clinical trials show effects building over 4–8 weeks of consistent use. The 2025 BJPsych Open meta-analysis found the anxiety effect at 8 weeks (μ = −3.52) was substantially larger than at earlier assessment (μ = −1.55). Ashwagandha is not a rapid-acting anxiolytic.

Is it safe for long-term use?

Short-term use (up to 12 weeks) is generally well tolerated in clinical trial populations. Long-term safety beyond 3 months is not well established from RCT data. Rare cases of liver injury have been reported. People with liver conditions, thyroid disorders, autoimmune diseases, or who are taking CNS-active medications should consult a healthcare provider before use. Avoid during pregnancy.

What side effects or interactions should be watched for?

Common mild side effects include gastrointestinal upset, drowsiness, and headache. More serious concerns include rare liver injury, thyroid hormone elevation (relevant for people on thyroid medication), and potential additive sedation with benzodiazepines, sleep medications, and alcohol. Disclose use to your healthcare provider, especially if you take any prescription medications.

Does it lower cortisol as well as anxiety?

Yes — the 2024 meta-analysis found a significant cortisol reduction (MD −2.58 nmol/L, 95% CI −4.99 to −0.16). A separate 2025 systematic review found significant cortisol reduction even in a meta-analysis that did not find significant effects on perceived stress. This cortisol data supports a biological mechanism (HPA axis modulation) beyond placebo.

Are the benefits consistent across different ashwagandha extracts?

The strongest evidence is for standardized extracts, particularly KSM-66 (≥5% withanolides). Studies using uncharacterized root powder show more variable results. The KSM-66 anxiety evidence synthesis within the broader meta-analytic literature supports its use specifically, though the overall meta-analyses pool different extract types together.

Is the evidence high quality or limited by bias and heterogeneity?

Both. The evidence base has genuine signal — consistent direction, statistical significance across multiple independent analyses, and Bayesian confirmation. But it is also limited by high I² (93–94%), small trials, publication bias risk, varied outcome measures, and a formal certainty-of-evidence rating of low from the 2022 meta-analysis. Honest interpretation requires holding both truths simultaneously.

Can it replace prescription anxiety treatment?

No. The evidence does not support using ashwagandha as a replacement for prescription treatment in people with diagnosed anxiety disorders. There are no head-to-head comparisons with SSRIs or other evidence-based treatments, the study populations are mostly subclinical, and the long-term evidence is limited. Ashwagandha may have a role as a complementary or adjunctive tool in healthy adults with subclinical stress and anxiety — but this should be discussed with a healthcare provider, particularly for anyone currently on psychiatric medication.

---

This article is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional before starting any supplement, particularly if you have a diagnosed medical condition, take prescription medications, or are pregnant or breastfeeding.

---

Sources Referenced:

• BJPsych Open (2025). Effects of ashwagandha supplements on cortisol, stress, and anxiety levels in adults: a systematic review and meta-analysis. Cambridge University Press. https://www.cambridge.org/core/journals/bjpsych-open/article/effects-of-ashwagandha-supplements-on-cortisol-stress-and-anxiety-levels-in-adults-a-systematic-review-and-metaanalysis/6F2D7847C1F64707F2034A45FD6CF0C0
• PubMed PMID: 36017529 — 2022 dose-response meta-analysis
• PubMed PMID: 39348746 — 2025 systematic review/meta-analysis in individuals with mental disorders