Kava and anxiety: what the clinical trials actually show

2023-07-25 · 8 min read

By the kava.fyi editorial team · Last updated: 25 July 2023

Kava (Piper methysticum) is one of the most studied plants in the context of anxiety. Contrary to what sales copy suggests, the results do not add up to one simple story: alongside a Cochrane review with a positive conclusion and several smaller studies with a clear effect, there is also the largest and longest study in the field, published in 2020, which did not confirm the effect. Below we gather only the papers that can be opened and checked, with the study design, the number of participants and the outcome.

This article is for information only and does not replace medical advice. Anxiety disorders are an illness that a specialist diagnoses and manages.

How to read studies on anxiety

Most of the papers discussed here use the same measure: the Hamilton Anxiety Rating Scale (HAMA). This is a questionnaire completed by a clinician, in which a higher score means stronger anxiety, and a score of 7 or below is treated in these studies as remission. It is also worth keeping three concepts in mind:

• Placebo and double blinding: neither the participant nor the researcher knows who received kava and who received an inactive tablet. Without this it is hard to separate the plant's effect from expectations.
• Effect size (Cohen's d): tells you how large the difference between groups is. By convention, 0.2 is a small effect, 0.5 a medium one and 0.8 a large one.
• Statistical significance (p): a p value below 0.05 means the difference is unlikely to be down to chance. It does not, however, say whether the difference matters in everyday life.

The specific preparation also matters. Some of the older European studies used standardised pharmaceutical extracts, such as WS 1490, while the Australian studies of recent years used water extracts of the root, closer to the traditional drink. Read more about the active constituents in the article on kavalactones.

The 2003 Cochrane review

The most frequently cited source is the Cochrane systematic review by Pittler and Ernst (2003). The authors included only randomised, double-blind, placebo-controlled trials in which kava was the sole ingredient of the preparation. Eleven trials with a total of 645 participants met the criteria. A meta-analysis of six of them that reported a combined HAMA score showed significantly greater anxiety reduction with kava than with placebo: a weighted mean difference of 5.0 points (95% confidence interval 1.1 to 8.8; p = 0.01; n = 345). Adverse events were described as mild, transient and rare.

The authors' conclusion was cautious: kava "appears to be an effective symptomatic option" for anxiety, and the data suggest relative safety with short-term use, from 1 to 24 weeks. At the same time, the review called for further research into long-term safety. It is worth remembering that the review covers the literature up to August 2002, that is, before all the water-extract trials.

The 25-week trial: Volz and Kieser 1997

The longest of the classic studies was carried out in Germany. Volz and Kieser enrolled 101 outpatients with non-psychotic anxiety (per DSM-III-R: agoraphobia, specific phobia, generalised anxiety disorder and adjustment disorder with anxiety) in a multicentre, randomised, placebo-controlled trial lasting 25 weeks. The preparation was the standardised extract WS 1490. On the main measure, HAMA, kava's advantage over placebo became significant from week 8. The extract also performed better on secondary measures, including the clinical global impression and the somatic and psychic anxiety subscales. Adverse events were rare and evenly distributed between the two groups.

This study has two weaknesses from today's perspective: a mixed diagnostic group and a preparation that is no longer available on the market in this form. It does, however, show something important: the effect, where it appeared, showed up after weeks, not after a single dose.

The Australian trials with the water extract

After kava medicinal products were withdrawn from the European and North American markets in 2002 (more in the article on kava and the liver), Jerome Sarris's team in Australia began a series of trials with a water extract of the root. Three of them form the core of current knowledge.

KADSS, 2009: a short trial, a very large effect

The Kava Anxiety Depression Spectrum Study was a three-week placebo-controlled crossover trial in which every participant went through both phases. Sixty adults with heightened generalised anxiety lasting at least a month were recruited; the dose used in the trial was 250 mg of kavalactones daily. In the first phase, HAMA fell by 9.9 points with kava versus 0.8 with placebo; in the second, by 10.3 points versus a rise of 3.3 with placebo. The combined effect was very large (d = 2.24; p < 0.0001). The authors found no clinical hepatotoxicity. Such a large effect size in a small, short trial is, however, a signal to wait for replication, and later studies from the same group showed considerably more modest effects.

2013: the first trial in diagnosed GAD

In 2013 the same team published the first placebo-controlled trial in people with diagnosed generalised anxiety disorder (GAD) without co-occurring mood disorders. Seventy-five people were enrolled, and after a one-week placebo run-in phase 58 were analysed. The trial lasted 6 weeks, with a dose of 120 or 240 mg of kavalactones daily depending on response. Kava reduced anxiety significantly more than placebo, but the effect was moderate (d = 0.62; p = 0.046); it was larger in people with moderate or severe GAD (d = 0.82). Remission was achieved by 26% of the kava group versus 6% of the placebo group. Kava was well tolerated; only headaches were reported more often, and liver test results did not differ between groups.

2020: the largest trial, a negative result

The pivotal paper appeared in 2020 in the Australian and New Zealand Journal of Psychiatry. This was a phase III, multicentre, randomised, double-blind, placebo-controlled trial lasting 16 weeks. It involved 171 pharmacologically untreated people with diagnosed GAD, who received a water extract of the root standardised to 120 mg of kavalactones twice daily, or placebo.

The difference in anxiety reduction between the groups was not significant, numerically 1.37 points in favour of placebo (p = 0.25). Remission was achieved by 17.4% of people in the kava group and 23.8% in the placebo group (p = 0.46). Worsening memory was reported more often in the kava group (36 versus 23 reports; p = 0.044), as was tremor (36 versus 23; p = 0.024), and abnormal liver test results were significantly more common, although no one met the criteria for drug-induced liver injury. The authors' conclusion: this particular extract was not effective in diagnosed GAD, and the earlier data support kava instead for situational anxiety, as a short-term remedy in non-clinical populations.

This trial matters because it was the largest, the longest, and designed precisely to settle the uncertainty. Its result cannot be brushed aside.

What newer reviews and guidelines say

• Ooi et al., 2018: a systematic review of 12 papers on GAD. A meta-analysis of three placebo-controlled trials (n = 130) pointed to an advantage for kava (standardised mean difference of 0.59 to 0.99), but without statistical significance. Conclusion: promising data, but insufficient to confirm an effect beyond placebo. The authors were awaiting the results of a large trial, which appeared two years later.
• Barić et al., 2018: a meta-analysis of complementary methods in GAD. A direct comparison of kava with placebo (4 kava arms, n = 139; 5 placebo arms, n = 359) gave an end-of-treatment HAMA difference of 3.24 points in favour of kava, at the borderline of significance (p = 0.059). The authors described the effect as moderate and safety reporting across the trials as poor.
• WFSBP and CANMAT guidelines, 2022: an international team of 31 experts assessed plant-based preparations in psychiatry based solely on meta-analyses and at least two randomised trials. Kava received a rating of "not recommended" for GAD, while ashwagandha and lavender were rated "conditionally recommended". The guidelines' first author is Sarris, who had previously run positive trials on kava, which lends this assessment extra weight.

It is worth setting this against the 2011 review by Sarris et al., which, before the 2020 trial, summed up that a significant result occurred in four of the six trials analysed (mean d = 1.1). The state of knowledge changed over the course of a decade, and that is the most important lesson from this list.

Safety and caveats

The 2011 review recommended traditional, water-soluble extracts from the rhizome of appropriate cultivars, avoiding combination with alcohol, caution with other psychotropic medications, avoiding large portions before driving, and regular liver checks in people using kava on an ongoing basis. The 2016 FAO/WHO report on the safety of the kava drink adds that kavalactones inhibit certain cytochrome P450 enzymes, which can affect the metabolism of medications taken at the same time, and that high consumption is linked to a flaking rash, weight loss, nausea and a reversible rise in the liver enzyme GGT. The report recommends using only noble cultivars and peeled root; we discuss the differences between cultivars in our guide to kava cultivars.

Summary

An honest summary is this: in short trials, usually up to 6 weeks, kava has repeatedly reduced anxiety symptoms more than placebo, and the 2003 Cochrane review judged it a symptomatic option. The largest and longest trial, from 2020, found no advantage over placebo in people with diagnosed GAD, and current guidelines do not recommend kava for this diagnosis. Kava is not a medicine and does not replace therapy; if anxiety is interfering with everyday functioning, the first step is to talk to a specialist. We write about how kava works at the receptor level in the article kava and GABA.

Sources

1. Pittler MH, Ernst E (2003): Kava extract for treating anxiety. Cochrane Database of Systematic Reviews https://pubmed.ncbi.nlm.nih.gov/12535473/
2. Volz HP, Kieser M (1997): Kava-kava extract WS 1490 versus placebo in anxiety disorders: a randomized placebo-controlled 25-week outpatient trial. Pharmacopsychiatry https://pubmed.ncbi.nlm.nih.gov/9065962/
3. Sarris J et al. (2009): The Kava Anxiety Depression Spectrum Study (KADSS): a randomized, placebo-controlled crossover trial using an aqueous extract of Piper methysticum. Psychopharmacology https://pubmed.ncbi.nlm.nih.gov/19430766/
4. Sarris J et al. (2013): Kava in the treatment of generalized anxiety disorder: a double-blind, randomized, placebo-controlled study. J Clin Psychopharmacol https://pubmed.ncbi.nlm.nih.gov/23635869/
5. Sarris J et al. (2020): Kava for generalised anxiety disorder: A 16-week double-blind, randomised, placebo-controlled study. Aust N Z J Psychiatry https://pubmed.ncbi.nlm.nih.gov/31813230/
6. Ooi SL, Henderson P, Pak SC (2018): Kava for Generalized Anxiety Disorder: A Review of Current Evidence. J Altern Complement Med https://pubmed.ncbi.nlm.nih.gov/29641222/
7. Barić H et al. (2018): Complementary and Alternative Medicine Treatments for Generalized Anxiety Disorder: Systematic Review and Meta-analysis of Randomized Controlled Trials. Adv Ther https://pubmed.ncbi.nlm.nih.gov/29508154/
8. Sarris J et al. (2022): Clinician guidelines for the treatment of psychiatric disorders with nutraceuticals and phytoceuticals: WFSBP and CANMAT Taskforce. World J Biol Psychiatry https://pubmed.ncbi.nlm.nih.gov/35311615/
9. Sarris J, LaPorte E, Schweitzer I (2011): Kava: a comprehensive review of efficacy, safety, and psychopharmacology. Aust N Z J Psychiatry https://pubmed.ncbi.nlm.nih.gov/21073405/
10. FAO/WHO (2016): Kava: a review of the safety of traditional and recreational beverage consumption. Technical Report, Rome https://openknowledge.fao.org/server/api/core/bitstreams/f97fbe5f-244e-4301-be2a-03ac39ef630b/content