Kava for Anxiety: What the Evidence Shows

Kava is the plant that beats placebo.

Of every supplement marketed for anxiety, kava has the cleanest efficacy signal in the peer-reviewed record — and the ugliest safety history.

Here’s the honest bottom line. Multiple meta-analyses show kava extract reduces anxiety more than placebo, with effect sizes most other “calm” supplements never reach. But the largest and longest trial ever run was flatly negative, and a wave of liver-injury reports pulled kava off shelves across Europe, the UK, Canada, and Australia in the early 2000s. It’s the most convincing supplement anxiolytic and the one with the most reason for caution — both at once.

Does kava actually work for anxiety?

This is the strongest part of the case, and it’s stronger than the herb’s reputation suggests.

The definitive synthesis is the Cochrane review by Pittler and Ernst (2003), which pooled randomized, double-blind, placebo-controlled trials of kava mono-preparations. Across 11 trials and 645 participants, a meta-analysis of six studies using the Hamilton Anxiety scale found a significant reduction favoring kava — a weighted mean difference of 5.0 points (95% CI 1.1 to 8.8; p = 0.01). The reviewers’ own earlier meta-analysis in the Journal of Clinical Psychopharmacology (2000) put the effect even higher: a weighted mean difference of 9.69 points on the Hamilton scale across three trials. Both concluded kava is superior to placebo as a short-term symptomatic treatment for anxiety.

For context, that’s a real, measurable signal on a validated clinician-rated scale — not the self-reported wobble that props up most supplement claims. That’s more than can be said for most herbs sold for anxiety.

What does the single-trial evidence show?

After the safety scare, researchers went back and tested the aqueous (water-based) extract the WHO had flagged as potentially safer.

In the Kava Anxiety Depression Spectrum Study (KADSS), Sarris et al. (2009), 60 adults with elevated generalized anxiety took an aqueous kava extract delivering 250 mg of kavalactones per day in a three-week placebo-controlled crossover. Kava cut Hamilton Anxiety scores by 9.9 points versus 0.8 on placebo in the first controlled phase, and the pooled cross-phase effect was highly significant. It also worked when anxiety came with depression.

Sarris then ran a proper generalized anxiety disorder (GAD) trial in 2013: 75 participants, aqueous kava titrated to 120–240 mg kavalactones/day over six weeks. Kava beat placebo with a moderate effect size (Cohen’s d = 0.62; p = 0.046) — and in the moderate-to-severe GAD subgroup the effect was larger (d = 0.82). 26% of the kava group remitted (Hamilton Anxiety ≤ 7) versus 6% on placebo. Liver function tests showed no between-group difference.

Why isn’t kava a first-line anxiety treatment?

Here’s the information-gain most “kava works” posts leave out: the biggest, longest, most rigorous trial failed.

Sarris and colleagues ran a phase III, multi-site, 16-week RCT (2020) in 171 people with diagnosed GAD — the same team, a similar aqueous extract, 120 mg kavalactones twice daily. The result: no significant difference from placebo. The point estimate actually favored placebo by 1.37 points on the Hamilton scale (p = 0.25). Only 17.4% of the kava group remitted. The authors’ own read is worth quoting: kava may help “more ‘situational’ anxiety as a short-term anxiolytic,” but “this particular extract was not effective for diagnosed generalised anxiety disorder.”

So the honest synthesis is narrower than the meta-analyses alone imply: kava has a genuine short-term signal for milder or situational anxiety, and a negative verdict for the clinical disorder over four months. Efficacy that evaporates in your largest trial is exactly the pattern you should distrust — the same skeptical read that applies to valerian for sleep and anxiety.

Is kava safe? The liver story

This is why kava isn’t sold like a vitamin. The first cases of kava-linked hepatotoxicity were described in 1998, and per a 2016 review by Pantano et al., more than 100 cases of suspected severe liver injury following kava exposure have since been reported worldwide. Products were withdrawn across Europe, the USA, and Australia, with FDA and CDC warnings issued from 2002 onward.

The critical, under-reported nuance is which kava. Pacific Islanders have consumed aqueous kava for centuries without a hepatotoxicity epidemic. Western supplements, by contrast, were typically made with ethanol or acetone extraction — solvents that concentrate kavalactones far higher than water does. Per the same review, an acetone–water extract runs about 70% kavalactones and an ethanol extract about 30%, versus roughly 30 times lower in the traditional aqueous drink. Liver injury clustered especially in women, appearing on average 111 days after first exposure.

Was it the solvent, the plant part (aerial leaves and stems contain the alkaloid pipermethystine, largely absent from the roots used traditionally), a poor-quality cultivar, or contamination? The literature still hasn’t settled it — some reports of injury involve aqueous preparations too, which points back at raw-material quality and the near-total absence of standardization rather than the extraction solvent alone. That unresolved mechanism is the caveat: nobody can yet promise you a clean extract.

And the safety story isn’t fully reassuring even in the modern aqueous trials. The negative 2020 phase III RCT reported liver function test abnormalities significantly more often in the kava group — though no participant met criteria for herb-induced liver injury — plus more tremor and poorer memory. “Well tolerated” is not the same as “inert.”

Who might it help, and how to use it?

If you’re considering kava for occasional, situational anxiety over a short window, the case is defensible — but only with a noble-cultivar, root-only, aqueous (water-based) product, never an ethanol or acetone extract, and never with alcohol, acetaminophen, or other liver-stressing drugs. Skip it entirely if you have any liver condition, take hepatotoxic medications, drink heavily, or are pregnant. Get liver enzymes checked if you use it beyond a few weeks, and stop at the first sign of fatigue, dark urine, or yellowing skin.

What kava won’t do is fix an over-aroused nervous system that resets itself every night through anticipatory dread. A short-term GABA-nudging plant can blunt a symptom; it doesn’t retrain the response. The same ceiling limits ashwagandha and most supplements sold for stress and anxiety — the lever is regulation, not chemistry.

The takeaway

Kava is the rare anxiety supplement with a real, replicated, clinician-scale signal against placebo — and also the one whose largest, longest trial found nothing and whose liver-injury history triggered outright bans. Read it as a possible short-term tool for mild or situational anxiety, sourced carefully as an aqueous root extract, not as a treatment for diagnosed anxiety disorder and not as a risk-free one. For the system this actually sits inside — how to wind an over-aroused nervous system down — see our anxiety regulation and sleep restoration work.