Kava for Anxiety: Extracts, Kavalactones and Liver Risk

What is kava?

Kava is a preparation of Piper methysticum, a Pacific shrub in the pepper family. Its roots and other underground parts are used to make a water-based drink and medicinal extracts. Kavalactones are a group of active plant compounds commonly used to describe extract strength. Six major kavalactones account for much of the chemical profile, alongside smaller amounts of other constituents. ^[1]

People take kava for relaxation, anxiety and sometimes sleep. Modern products include root powders, tablets, concentrated liquid extracts and ready-to-drink beverages. A serving's total root weight, its kavalactone amount and its other ingredients all matter when comparing those products.

History: from Pacific gatherings to medicinal extracts

Kava has a long social and ceremonial history across parts of the Pacific. FSANZ's historical account places its likely origin in northern Vanuatu and describes use for more than a thousand years. Traditional preparation involves grinding or pounding fresh or dried roots, mixing them with water and sharing the drink from a communal bowl. Practices and preferred varieties differ between communities. ^[1]

In Fiji, the drink is known as yaqona. Cambridge's Museum of Archaeology and Anthropology describes nineteenth-century chiefly and priestly use in religious rituals, alongside the development of more public and recreational consumption. Its collection includes wooden mixing bowls called tanoa and coconut-shell cups called bilo. Presenting roots and sharing yaqona remain important Fijian customs. ^[2]

Kava also entered European medical texts. The EMA's historical review describes the British Pharmaceutical Codex of 1911 listing solid and liquid kava extracts for urinary complaints, including an alcohol-based preparation. Later herbal references included oral urinary uses and external use for joint pain. These records document what people used, rather than demonstrate that kava treats an infection or relieves joint pain. ^[3]

Today's supplement research focuses largely on anxiety, while commercial drinks are often presented as alcohol alternatives. The ceremonial history helps explain kava's social role; the effects and risks of a modern concentrate or mixed drink still depend on what is actually in it. ^[4]

Why do people take kava?

When to ask a doctor before taking it

Discuss kava before use if you have liver disease or take medicines that affect alertness. Do not combine it with alcohol or sedatives such as benzodiazepines. Include sleep medicines, pain medicines and other herbal products in that discussion. Anxiety that interferes with daily life deserves an assessment and a review of established treatment options. ^[6]

Counting kavalactones

A label should state the actual extract amount and, when standardized, its kavalactone amount or percentage. Multiplying extract weight by that percentage gives the kavalactone amount: for example, 300 mg of an extract standardized to 30% supplies 90 mg of kavalactones. This is label arithmetic, not a suggested dose.

Root powder, extract weight and kavalactone weight are different measurements. The anxiety trials below used stated daily kavalactone amounts, while some older papers describe milligrams of a named extract. Neither the size of a drink nor the words “traditional strength” establish a safe daily intake. ^[6]

Which preparations differ?

Three details help explain what a product contains:

• Plant variety: “Noble” identifies traditionally preferred varieties used for beverages. Tu-dei, or “two-day,” varieties have been associated with longer-lasting effects, nausea, headache and lethargy. A noble designation is useful identity information, but it does not certify safety.
• Plant part: Roots and rhizomes—the underground stems—are common starting materials. Traditional beverage descriptions also include basal stems. Leaves and upper plant material have a different chemical composition; “kava” alone is an incomplete description of the ingredient.
• Extraction: Mixing ground plant material with water produces a different preparation from a concentrated water, ethanol or acetone extract. The extraction method affects which compounds and amounts enter the product. ^[1]

The protocol for the longer anxiety trial specified a noble Borogu water extract made into tablets by MediHerb. A later paper from the same research program notes that the trial batch was different from the manufacturer's subsequently marketed product. Even a named manufacturer or cultivar does not establish that a current bottle has the composition used in a study. ^[7], ^[8]

For a prepared drink, read the whole ingredient list. A kava-and-kratom beverage introduces another psychoactive plant and has different evidence from kava alone.

Drawbacks and side effects

These effects are described by NCCIH^[6]. Heavy, ongoing beverage consumption has also been associated with dry, flaky skin and yellow discoloration, sometimes called kava dermopathy. Yellow skin or eyes should not automatically be dismissed as a skin effect when liver injury is also possible. FSANZ advises against driving or operating heavy machinery after consuming kava beverage. ^[1]

Women who are pregnant or breastfeeding should avoid self-directed kava use. NIH identifies potentially harmful pyrone constituents—chemicals present in kava—as a concern in these stages, alongside incomplete safety information. ^[6]

Liver injury has more than one proposed explanation

Reports have included severe injury, liver transplantation and deaths. Proposed explanations include plant variety, non-root material, extraction, contaminants, other medicines and individual susceptibility. None has been established as the single explanation. Reports include water-prepared products as well as alcohol and acetone extracts. ^[6]

A 2009 case assessment compared five people who had used water-based extracts or kava-herb mixtures with nine who had used ethanol or acetone extracts. The clinical pattern was similar. Because some products were mixtures, the study cannot assign every case to kava alone or calculate the risk for a particular drink. It does challenge the idea that liver reports concern only organic-solvent extracts. ^[9]

The proposed toxin pipermethystine is another example of why the explanation remains unsettled. An analysis of retained German products and root extracts did not detect it at the study's measurement limit, although a leaf sample did contain it. That result does not support blaming pipermethystine for the injuries associated with those tested products. ^[10]

A 2025 report described liver injury in a man who had taken 10 g of kava a day for sleep. He was also taking several medicines; the paper did not identify a kavalactone amount or establish that the product was a water extract. A structured causality assessment rated kava a probable cause. Liver enzymes improved after stopping but were still above the listed reference ranges at three months. ^[11]

Medicine interactions: human data are more selective

Laboratory experiments suggest that kava compounds can affect enzymes that break down medicines. Human results are more selective. In a study of 12 healthy adults taking one root extract for 28 days, researchers found reduced activity in one of four tested enzyme pathways, with no significant change in the other three. ^[12]

Those measurements used test drugs to estimate enzyme activity. They do not establish how every prescription medicine will behave with every kava product. The separate concern about combining sedating substances remains: alcohol and sedative medicines may add to kava's effects on alertness. ^[6]

Kava and kratom in poison-center reports

A 2026 CDC report documented 203 kava-related reports to US poison centers in 2025, including 61 involving kratom. These were consultations about actual or suspected exposure, whether or not symptoms occurred. They were not 203 confirmed cases of kava poisoning. ^[4]

The increase in co-use occurred alongside more serious reported outcomes. The records could not distinguish product preparation or dose, and reporting was voluntary. They therefore identify a safety concern without proving that the combination caused a particular liver or nervous-system effect, or estimating the risk for everyone who drinks kava.

Dependence and stopping

The six-week anxiety trial's safety report found no difference from placebo in measured addiction or withdrawal symptoms. That is reassuring for that short study, but it is not evidence that dependence and withdrawal cannot occur with heavier or longer use. ^[13]

A later case described hallucinations, anxiety, insomnia and muscle jerking after a man stopped heavy use of a beverage containing both kava and kratom. The treating authors suspected a withdrawal syndrome involving kava, but the combined product prevents a clear separation of each ingredient's contribution. ^[14]

What do the human studies show?

Kava's clinical research is concentrated on anxiety, with smaller sleep and cognition studies. The exact extract matters: some older studies used the acetone-based WS 1490 preparation, while later anxiety trials used water-based extracts. Their results can be read together without treating those products as interchangeable.

Short courses and the longer generalized-anxiety trial

The 2009 Kava Anxiety Depression Spectrum Study recruited 60 adults with elevated anxiety for at least a month. In its three-week study, participants received kava and placebo at different times. It used a water-based extract supplying 250 mg of kavalactones daily and reported greater improvement in anxiety scores than placebo. This was a broader anxiety population than the later trial restricted to diagnosed generalized anxiety disorder. ^[15]

A separate six-week trial enrolled 75 adults with generalized anxiety disorder and no accompanying mood disorder. Its main analysis included 58 people after an initial week on placebo. Treatment began at 120 mg of kavalactones daily, increasing to 240 mg for people who did not respond. Anxiety scores improved more with kava, and 26% reached remission, meaning their symptoms fell below the trial's set cutoff, compared with 6% on placebo. Headaches were more frequent with kava. ^[16]

The later, larger trial analyzed 171 adults with diagnosed generalized anxiety disorder over 16 weeks. It used a fixed daily amount of 240 mg of kavalactones from a water-based extract. Kava did not significantly improve anxiety more than placebo. Remission occurred in 17.4% of the kava group and 23.8% of the placebo group, also without a significant difference. Memory complaints, tremor or shakiness, and liver-test abnormalities were more frequent with kava, although no participant met the trial's criteria for liver injury. ^[5]

The positive six-week study means the disagreement cannot be explained simply as “kava works for everyday stress but never for diagnosed anxiety.” These studies differed in participants, preparation, treatment length and whether the dose could increase. They did not directly test which of those differences caused the conflicting results.

Earlier studies included favorable and null results

Earlier studies of the named WS 1490 extract were sometimes favorable. A four-week trial of 141 adults^[17] found improvements in several anxiety measures with 150 mg/day extract, although differences were smaller than in earlier studies using 300 mg/day. Those are extract weights, not kavalactone weights. An individual-patient analysis of six WS 1490 trials^[18] also favored the extract in non-psychotic anxiety disorders. The later 16-week water-extract trial is an important counterweight; the difference cannot simply be attributed to diagnosed versus everyday anxiety.

Positive short trials are only part of the earlier evidence. A 2006 report combining three smaller trials in generalized anxiety disorder did not find an advantage for kava over placebo. Its abstract does not specify enough preparation detail to compare the extract directly with the later water-extract tablets. ^[19]

An internet-based trial randomized 391 adults with anxiety and insomnia to kava, valerian or double placebo for four weeks. The kava group did not improve anxiety significantly more than placebo. This was a different research setting from the clinic-based diagnosed-anxiety trials. ^[20]

What might explain the effects?

Kavalactones have several proposed actions on nerve-cell signaling, including effects involving ion channels and GABA-related signaling. GABA is a chemical messenger involved in limiting nerve activity. FSANZ's assessment describes the overall mechanism as incompletely established. Calling kava a natural equivalent of a benzodiazepine goes beyond that evidence. ^[1]

A brain-imaging follow-up from the 16-week trial illustrates the distinction. It found a change in GABA measurements in a small subset, with only 20 people scanned at eight weeks. The authors noted that this chemical change could occur without improvement in anxiety symptoms. It was a subset of the same trial, not an additional successful treatment study. ^[8]

Cognition and immediate calming

One small crossover trial tested a single dose containing 180 mg of kavalactones in 22 mildly or moderately anxious adults. Kava did not provide immediate anxiety relief in that experiment. It also did not impair the tested cognitive tasks; two attention-task comparisons favored kava over placebo. The benzodiazepine comparison, oxazepam, reduced alertness. ^[21]

These findings concern one dose and a set of laboratory tests. The memory complaints in the longer anxiety trial and the different quantities consumed in recreational drinks make “relaxation without any impairment” too broad a promise.

Sleep and next-day alertness

Kava is sometimes taken for sleep, but sedation alone does not tell us whether a person sleeps better or feels more alert the next day. A four-week trial in 61 people^[22] with sleep disturbance associated with anxiety found better self-rated sleep quality and feeling more restored after sleep with 200 mg/day of WS 1490 than with placebo. It used questionnaires, not overnight recordings of sleep stages.

By contrast, the four-week internet trial in adults with anxiety and insomnia concluded that neither kava nor valerian relieved those symptoms more than placebo. ^[20]

The positive questionnaire result does not establish deeper sleep stages, reliably faster sleep onset or freedom from next-day impairment. NIH considers the evidence for conditions beyond anxiety insufficient to draw firm conclusions. ^[6]

Sexual function was a secondary finding

The safety paper from the six-week anxiety trial also reported improved sexual-drive scores in women taking kava. This was a secondary result in the same anxiety study. It does not establish a separate treatment for sexual dysfunction. ^[13]

When do liver tests matter?

Liver tests may investigate a suspected adverse reaction or guide an individual medical plan. A normal test does not guarantee that continued use will be safe, and the research does not establish a universal home monitoring schedule. Tell the clinician about the exact kava product, amount, duration and other medicines or drinks involved.

The bottom line

Kava's history includes both Pacific social beverages and medicinal extracts. Some short anxiety trials and one self-rated sleep trial favored named extracts, while a larger sixteen-week anxiety trial did not. A short cognitive test found no immediate anxiety benefit, and the longer study recorded more memory complaints. A bar drink, concentrate or mixture cannot be assumed to produce the same results as a trial extract.

For a product decision, the useful details are the plant part, cultivar, extraction method, kavalactone amount and other ingredients. For a health decision, consider the actual anxiety or sleep problem, other medicines, sedation and liver history. The studies discussed here do not establish a universal course or safe all-day intake.

Help us improve this article

Spotted an error or found a source we should consider? Tell us which part of this article needs another look and include a link if you have one. Please leave out personal health details and lab results; we can’t answer individual medical questions.

Your name

Email address

What would you like to share?Select a topicCorrectionSource or research suggestionArticle requestAccessibility issueBusiness inquirySomething else

Your message

Continue to email →Email us directly