Aromatherapy

The term and its attribution

An encyclopedia research starter states that René-Maurice Gattefossé (1881 to 1950), a French chemical engineer from a family perfumery business, introduced the term aromathérapie in his 1937 book Aromathérapie: Les Huiles Essentielles, Hormones Végétales. A university science-communication office also credits the term to Gattefossé in 1937. A Brazilian nursing-theory article instead says the term was first used in 1928, so the sources reviewed disagree on the date and none of them resolves it.

Several accounts tell of Gattefossé burning a hand in a laboratory and using lavender oil, but the research starter notes that the story is qualified and that many historians suggest the self-experimentation was deliberate. It gives no date for the incident. The episode is better treated as a founding anecdote than as documented evidence about lavender oil.

Valnet, Maury and the English-language field

A science-communication article states that Jean Valnet later studied essential oils for medicinal use by applying them to soldiers' wounds during World War II. A practitioner's retrospective gives 1964 as the publication year of Valnet's book Aromathérapie. The Brazilian article describes Marguerite Maury as born in Austria in 1895 and credits her with individualized blends of essential oils and with clinics in London, France and Switzerland.

The same practitioner describes Maury's approach as a mix of energetics and skin care and says that a UK school teaching through the 1960s and most of the 1970s drew on information from her. He describes his own 1977 book The Art of Aromatherapy as the first on the subject in English. These are one practitioner's recollections, and the sources reviewed give no firm dates for Maury's own work.

Earlier aromatic uses

The sources reviewed for this page do not document dated ancient or premodern medical uses of aromatic plant oils, so no date is given here. The nursing history article says only that aromatherapy is used within traditional medicine in Eastern countries, without naming traditions or periods. It reports that Florence Nightingale used lavender oil on the foreheads of wounded soldiers during the Crimean War to calm them, without a year, and that 1960s counterculture raised Western interest in such practices.

How bodies define the term

The US National Center for Complementary and Integrative Health (NCCIH) describes aromatherapy as the use of essential oils from plants as a complementary health approach, most often by inhaling them or applying a diluted form to the skin. Its page, last updated in January 2020, names oils such as Roman chamomile, geranium, lavender, tea tree, lemon, ginger, cedarwood and bergamot.

The UK Complementary and Natural Healthcare Council (CNHC) core curriculum, published in June 2025, defines aromatherapy as the use of essential oils and related aromatic extracts, such as absolutes, hydrolats and CO2 extracts (excluding CBD), in holistic treatments to improve physical and emotional well-being. It lists massage, inhalation, topical applications and compresses as example treatments.

How research trials use the word

Trial reviewers apply the label more loosely than professional definitions do. A Cochrane review of nausea after surgery counted inhaled isopropyl alcohol and peppermint products as aromatherapy, while a Cochrane review in dementia defined the intervention as fragrance from plants. A review of cancer symptoms separated massage with carrier oil only from massage with essential oils. Results therefore cannot be pooled across studies without checking what each trial actually used.

Methods of use

At overview level, the sources describe three delivery routes: inhalation, including by diffusers, vaporizers and steamers; massage with oils blended into a carrier medium; and other topical applications such as compresses and baths. NCCIH states that essential oils are most often inhaled or applied in diluted form. A UK hospital leaflet for pregnant patients lists massage, smelling oil from a tissue or cotton ball, and compresses. This page does not give blends, doses or dilution tables.

Consultation and professional practice

The CNHC curriculum expects practitioners to use consultation methods that include observation, smell, listening and touch, to recognize cautions, contraindications and adverse reactions, and to recognize conditions for which aromatherapy is inappropriate and the client should seek advice elsewhere. It also lists circumstances in which a practitioner may choose not to accept a client, and the assessment of oil quality and adulteration.

Nursing and maternity settings

A Brazilian review states that nurses have practiced aromatherapy historically and worldwide. In England, one NHS hospital trust leaflet describes midwife-led aromatherapy offered after 12 weeks of pregnancy to women whose pregnancy is going well, using massage, smelling or compresses, and it asks patients to disclose conditions including epilepsy before oils are used. This is one service description, not evidence of how widespread such care is.

Uses that have been proposed or studied

Claims reviewed in the sources include relief of anxiety, stress, pain, depression, sleep difficulty, nausea, agitation in dementia and discomfort in labor. NCCIH says aromatherapy is sometimes used for insomnia and is sometimes added to massage for knee pain from osteoarthritis and for pain, anxiety and other symptoms in people with cancer. Systematic reviews have tested many of these uses, as the next section reports.

Mechanisms and unsupported claims

Practitioner curricula teach that essential oils act through the olfactory and limbic systems on psychological functioning. The sources reviewed do not show this mechanism in human trials. A university science-communication article states that scientific evidence is thin for popular claims such as eucalyptus preventing colds or a blend relieving stress. Claims that oils detoxify the body, treat infection or treat cancer were not supported by any source reviewed here.

Dementia and postoperative nausea

A 2020 Cochrane review of aromatherapy for dementia included 13 randomized trials with 708 participants, all in institutional care where settings were reported. Fragrances were mainly lavender and lemon balm. Of five trials rated moderate or low confidence for agitation, four found no significant effect and one found benefit. The authors found no convincing evidence of benefit, citing small trials and poor reporting of results and harms.

A 2018 Cochrane review of postoperative nausea and vomiting included 16 studies with 1,036 participants. Compared with placebo, aromatherapy did not clearly reduce nausea severity (standardized mean difference -0.22, 95% confidence interval -0.63 to 0.18; low-certainty evidence) but was linked to less use of rescue antiemetics (risk ratio 0.60, 0.37 to 0.97; low certainty).

Labor pain and cancer symptoms

A 2011 Cochrane review of aromatherapy for pain in labor included two trials with 535 women and found no difference in pain intensity, assisted birth, cesarean section or use of pain medication, and it called for more studies before practice recommendations. A 2016 Cochrane review of massage with or without aromatherapy in cancer included 19 studies with 1,274 participants and rated all outcomes very low quality.

In that cancer review, aromatherapy massage compared with no massage showed small reductions in anxiety and medium-term pain that the authors judged unlikely to be clinically significant. Only two studies compared massage with and without essential oils, too few to judge what the oils add. Two studies reported adverse events, such as rash and malaise; the other 17 did not report them.

Anxiety and sleep

A 2023 systematic review and network meta-analysis of 44 randomized trials with 3,419 people found lower state anxiety scores with essential oils than with controls (weighted mean difference -6.63, 95% confidence interval -8.17 to -5.08). Heterogeneity was very high (I-squared 93.2 percent), 17 trials were at high risk of bias, and the authors rated evidence quality as low or very low. No trial reported adverse events, which the authors said left safety uncertain.

A 2024 meta-analysis of 10 mostly small Chinese trials with 692 older adults reported better sleep quality scores with aromatherapy (standardized mean difference -1.02). Seven of the 10 trials came from China, all were single-center, only two blinded the implementer, and heterogeneity fell sharply when one trial was excluded. The included trials reported no adverse event data.

Reading this evidence

Across these reviews, trials are small, outcomes are mostly self-rated, and the oils, doses and delivery routes vary. NCCIH notes that little rigorous research addresses insomnia, and it describes a study in people under stress in which lemon scent improved mood while neither lemon nor lavender affected stress indicators, immune markers or pain control. Pooled benefits for anxiety and sleep are therefore uncertain and do not carry over to other conditions.

Reported adverse effects

A systematic review of case reports and case series found 42 reports describing 71 patients with adverse effects of aromatherapy. Effects ranged from mild to severe and included one death, and dermatitis was the most common. Lavender, peppermint, tea tree oil and ylang-ylang were most often implicated. Case reports cannot show how often events occur.

Sunlight adds a risk for some oils: in a small photopatch study of citrus-type cosmetic oils, 3 of 8 oils caused reactions in volunteers at the highest concentration that was not phototoxic in a skin model.

Swallowing and children

A review of Ontario poison center calls from 1995 to 1997 recorded 251 children who had swallowed essential oils or oil-containing products, including eucalyptus oil (50) and camphorated oil (18). Cough and vomiting were most common, two children had seizures and recovered, and a literature search found 18 pediatric reports including one death. The data concern camphor products and eucalyptus oil.

Estrogenic activity: lavender, tea tree oil and the debate

In 2007 a US report described three prepubertal boys with gynecomastia that coincided with topical use of products containing lavender and tea tree oils and resolved after use stopped; cell-line tests showed estrogenic and antiandrogenic activity, and the authors concluded the oils probably caused it. A 2019 report added four children with similar exposure to lavender products and resolution after discontinuation.

Later work questions the link. A 2019 systematic review found four papers describing 11 cases, judged reporting insufficient and found no evidence for a tea tree oil link and little for lavender. A 2021 parent survey of 556 children found no gynecomastia and no clear difference in endocrine disorders (risk ratio 2.80, interval 0.35 to 22.16, very wide).

Epilepsy, pregnancy and pets

A 2019 systematic review of essential oils and seizures found eight human case reports linked to oils such as sage, thuja, hyssop, camphor, rosemary and eucalyptus, plus animal data on constituents such as thujone and 1,8-cineole. The authors advised caution, especially in children and people with epilepsy. Its funder and several authors are linked to an essential oil company.

An NHS hospital trust leaflet asks patients with epilepsy, heart problems or asthma to tell their midwife and states there is not enough evidence to confirm essential oils are safe in those situations. For pets, the sources reviewed hold one case report of liver damage in a cockatiel after tea tree oil on its skin.

Training and when to seek care

In the UK, the CNHC core curriculum sets a minimum standard for practitioners applying for CNHC registration: level 3 qualifications in England and Wales and 590 hours of total qualification time. No US standard is reported because US credentialing pages could not be opened, and no source reviewed describes statutory regulation.

Essential oils are not a substitute for assessment or prescribed treatment. A child who has swallowed essential oil, a seizure, breathing difficulty or a severe skin reaction call for urgent medical help or a poison center.