Human Clinical Studies on Essential Oils: What Exists, Limits and Conflicts
What systematic reviews of human trials of aromatherapy and essential oils found, where reviews disagree, and why most trials are small, short and hard to blind.
Content last updated . This page has not been medically reviewed.
Safety first. Natural products can cause side effects and interact with medicines. Read essential oil safety before you use anything on this page.
The short answer. Many human trials exist, but the best reviews we read describe them as small, varied and often at high risk of bias. Some reviews report benefit for anxiety, sleep or pain; others find the evidence unconvincing. Both can be true, because the reviews ask different questions. This page is information only, and it is not advice to use or swallow any oil.
What the reviews found
| Question | What the review found | Evidence label |
|---|---|---|
| Any condition | An overview of 10 systematic reviews (conditions included high blood pressure, depression, anxiety, pain and dementia): most reviews were poor quality, and for none of the conditions was the evidence convincing. [1] | Insufficient evidence |
| Anxiety, lavender | 65 RCTs (7,993 people) and 25 non-randomized studies. Inhaled lavender lowered anxiety scale scores in the pooled data; the authors said most trials had a high risk of bias. An oral lavender oil preparation and lavender massage were also reported to lower scores. [2] | Promising but limited |
| Sleep | A meta-analysis of 30 studies (2011 to 2019) reported an overall improvement in sleep quality. Subgroups differed widely, and the review included quasi-experimental studies. [3] | Promising but limited |
| Nausea after surgery | 16 studies, 1,036 people. Compared with placebo, aromatherapy was not shown to cut nausea severity (low-quality evidence); people may have needed fewer rescue anti-nausea drugs (low-quality). Peppermint did not show an effect on nausea severity at five minutes. [4] | Mixed, low certainty |
| Cancer symptoms, massage | 19 studies, 1,274 people, evidence graded very low. Some benefits for pain and anxiety were seen but judged unlikely to be clinically significant. [5] | Insufficient evidence |
| Cancer pain | A review of 12 studies (6 in meta-analysis) reported significant pain reduction with essential oils and called for better-designed trials. [6] | Promising but limited |
| Dementia, agitation | 13 trials, 708 people, in care homes. No convincing evidence of benefit; harms poorly reported. [7] | Insufficient evidence |
| Irritable bowel syndrome, peppermint oil (oral, tested in capsule trials) | 10 RCTs, 1,030 people: better than placebo for global symptoms and pain, but more adverse events, and the authors rated quality very low. [8] | Promising but limited |
| Tea tree oil (on skin or in mouthrinse) | 46 RCTs; some dental findings, more evidence needed for acne; side effects in 60% of studies, mostly minor except at concentrations of 25% or more. Overall quality poor to modest. [9] | Promising but limited |
| Eucalyptus oil, allergic rhinitis | A review with only 3 trials reported lower symptom scores and an improved quality-of-life score. [10] | Insufficient evidence (very few trials) |
Where reviews conflict
- "Not convincing" versus "improves". The 2012 overview judged the evidence unconvincing for every condition. [1] A 2021 meta-analysis concluded that aromatherapy improves sleep quality and reduces stress, pain, anxiety, depression and fatigue. [3] The reviews differ in date, in which studies they pooled, and in how they graded study quality. In our reading, the later review included quasi-experimental studies, which are less strict than RCTs. [3]
- "Significant" versus "not clinically significant". Two cancer reviews both noted weak study quality; one reported statistically significant pain reduction, while the Cochrane review judged similar findings unlikely to matter clinically. [6] [5]
- Lavender: NIH versus a meta-analysis. NIH says it is unclear whether lavender aromatherapy helps anxiety, stress, depression, some types of pain or sleep, while a meta-analysis reported lower anxiety scores. [11] [2]
A statistically significant result is not the same as a difference a person would notice.
Limits shared by most trials
- Small groups and short follow-up. In the dementia review, trials ranged from 18 to 186 people. [7]
- Hard to blind: people can smell what they are given. See How scent works.
- Self-reported outcomes, such as anxiety scores.
- Different oils, products, routes and amounts, so results do not add up neatly. The dementia review did not pool results because of heterogeneity. [7]
- Harms are poorly reported. Adverse events were poorly reported or not reported in most dementia trials, and 17 of 19 cancer-massage studies did not report adverse events. [7] [5]
- Independent funding and testing: NIH noted a lack of both in oral lavender studies. [11]
Frequently asked questions
Is there good human evidence that essential oils work?
Not clearly. The most cautious review we read judged the evidence not convincing for any condition, and others report some benefit with weak study quality. [1] [2]
Why do reviews reach different conclusions?
They differ in the questions asked, the trials included, and how quality was judged. Our reading of the abstracts shows differences in included study types and outcomes. [3] [1]
Which condition has the best evidence?
We could not name one with strong evidence. Peppermint oil capsules in IBS and inhaled lavender for anxiety are among the better studied uses in the reviews we read, but both reviews cited weak study quality or very low certainty. [8] [2]
Do the trials show essential oils are safe?
No. Safety reporting was poor in several reviews, and some oils cause skin reactions or serious poisoning if swallowed. [7] [12]
Are trials of capsules the same as aromatherapy?
No. Oral capsules and inhaled scent are different products with different risks. NIH describes the peppermint capsule research as studies of enteric-coated capsules. [13]
More details: evidence, limits and cautions
The details behind this page: cautions, limits of the evidence and where sources disagree. Open any note to read it.
How to read this page
- A randomized controlled trial (RCT) assigns people by chance to a treatment or a comparison. A systematic review gathers all such trials and judges them together. A meta-analysis combines their numbers.
- Aromatherapy is usually inhaled, or applied in a massage. Oral products such as enteric-coated peppermint oil capsules are pharmaceutical-type products tested in trials, and are described here as trials only.
- Evidence labels follow the authors' own wording where possible.
Safety and who should ask a clinician
Trial results do not make essential oils safe for everyone. Essential oils should not be swallowed; tea tree oil taken by mouth can cause confusion, unsteadiness, inability to walk and coma. [12] If someone has swallowed an oil, call Poison Help at 1-800-222-1222 (US), or 911 for collapse, a seizure or trouble breathing. NIH notes menthol from peppermint should not be inhaled by or applied to the face of an infant or small child. [13] Ask a pharmacist or clinician about medicines, pregnancy, children and surgery; see medicine interactions and essential oil safety.
See also Essential oil research and evidence and Research by oil.
References
- Lee MS, Choi J, Posadzki P, Ernst E. Aromatherapy for health care: an overview of systematic reviews. Maturitas. 2012;71(3):257-260 (abstract read). Read 2026-10-23.
- Donelli D, Antonelli M, Bellinazzi C, Gensini GF, Firenzuoli F. Effects of lavender on anxiety: a systematic review and meta-analysis. Phytomedicine. 2019;65:153099 (abstract read). Read 2026-10-23.
- Her J, Cho MK. Effect of aromatherapy on sleep quality of adults and elderly people: a systematic literature review and meta-analysis. Complement Ther Med. 2021;60:102739 (abstract read). Read 2026-10-23.
- Hines S, Steels E, Chang A, Gibbons K. Aromatherapy for treatment of postoperative nausea and vomiting. Cochrane Database Syst Rev. 2018;3:CD007598 (abstract read). Read 2026-10-23.
- Shin ES, Seo KH, Lee SH, et al. Massage with or without aromatherapy for symptom relief in people with cancer. Cochrane Database Syst Rev. 2016;2016(6):CD009873 (abstract read). Read 2026-10-23.
- Corasaniti MT, Bagetta G, Morrone LA, et al. Efficacy of essential oils in relieving cancer pain: a systematic review and meta-analysis. Int J Mol Sci. 2023;24(8):7085 (abstract read). Read 2026-10-23.
- Ball EL, Owen-Booth B, Gray A, et al. Aromatherapy for dementia. Cochrane Database Syst Rev. 2020;8:CD003150 (abstract read). Read 2026-10-23.
- Ingrosso MR, Ianiro G, Nee J, et al. Systematic review and meta-analysis: efficacy of peppermint oil in irritable bowel syndrome. Aliment Pharmacol Ther. 2022;56(6):932-941 (abstract read). Read 2026-10-23.
- Kairey L, Agnew T, Bowles EJ, et al. Efficacy and safety of Melaleuca alternifolia (tea tree) oil for human health: a systematic review of randomized controlled trials. Front Pharmacol. 2023;14:1116077 (abstract read). Read 2026-10-23.
- Hamayal M, Shahid W, Akhtar CH, et al. Effect of eucalyptus oil on symptoms of allergic rhinitis: a systematic review. Indian J Otolaryngol Head Neck Surg. 2025;77(2):967-974 (abstract read). Read 2026-10-23.
- National Center for Complementary and Integrative Health (NIH). Lavender: Usefulness and Safety. Page last updated February 2025. Read 2026-10-23.
- National Center for Complementary and Integrative Health (NIH). Tea Tree Oil: Usefulness and Safety. Page last updated April 2025. Read 2026-10-23.
- National Center for Complementary and Integrative Health (NIH). Peppermint Oil: Usefulness and Safety. Page last updated May 2025. Read 2026-10-23.