Does Aromatherapy Work? Evidence for Sleep, Stress and Nausea

A drop of lavender on a pillow is the most common piece of aromatherapy in the United States and the one with the most trials behind it, roughly forty of them, most small, most short, most measuring how well people say they slept. That research says something modest and real. The bottle in the shop next to it, labelled for immunity and hormone balance, says something with no research at all. The difference between those two bottles is the whole answer to whether aromatherapy works, and it is what our practitioners learn to explain in their first week.

This article stays on evidence. The safety arithmetic of percentages and drops lives in our essential oil dilution chart, and the training and scope questions in our guide to aromatherapy certification.

What do the studies show, use by use?

Small effects on symptoms people rate themselves, mostly from inhalation, mostly lasting minutes to weeks. The table sorts the common claims by what has been published. Where a use has no credible trial, it says so.

Why is smell so hard to blind in a trial?

Because the participant knows whether the room smells of lavender. A placebo pill can be made identical to a drug; a placebo smell cannot be made identical to a real one, and a participant who recognises lavender arrives with everything they believe about lavender. That is why the best-designed trials compare lavender with a different pleasant smell, such as sweet almond oil or a neutral fragrance, and why the gap between the two groups is usually smaller than the gap between lavender and nothing.

Our aromatherapy manual describes three levels at which oils are said to act: physiological, through active molecules; emotional, through olfaction stimulating the limbic brain, the seat of emotion and memory; and energetic, through a vibrational frequency acting on subtle fields. We teach all three as the tradition, and we tell students plainly which has evidence. The emotional level, smell reaching memory and mood, is where the trials sit. The physiological level has laboratory support for some molecules and little human evidence at aromatherapy doses. The energetic level has no measurable basis and is taught as a belief some clients hold, not as a mechanism.

Inhalation, topical or ingestion: which route has evidence?

Inhalation, by a wide margin. The sleep, anxiety and nausea trials nearly all used a diffuser, a cotton pad or an inhaler stick. Topical use has evidence mostly as part of massage, which cannot be separated from the touch. Ingestion has one serious exception, a standardised oral lavender preparation studied in Europe for anxiety, and that is a pharmaceutical product, not a bottle of essential oil from a shop.

• Inhalation. Where the trials are. Fast onset, minutes; short duration; lowest risk. Our manual's route for emotions.
• Topical, diluted. Evidence mainly via massage. Skin risks: irritation, sensitisation, phototoxicity. Always at the dilution chart's percentages.
• Ingestion. Outside our scope entirely. Our manual reserves internal use for medically trained practitioners; swallowed oils have caused poisonings, especially in children.

What does the FDA say about claims on oils?

That the claim, not the bottle, decides what the product is. Under U.S. law an essential oil sold for fragrance or to cleanse or beautify is a cosmetic; the same oil sold to relieve anxiety, fight infection or balance hormones is a drug, and an unapproved one. The FDA applied that reasoning in 2014 when it sent warning letters to two of the largest multi-level marketing sellers over distributors' claims that oils could address viral infections and other conditions.

For a practitioner, the consequence is practical. A label that reads "lavender, for a calming evening ritual" is a cosmetic statement. A label that reads "lavender, treats insomnia" turns a $12 bottle into an unapproved drug and its seller into someone the FDA can write to. We teach students to write labels and websites in the first register only, and the Federal Trade Commission's rule that health claims need competent scientific evidence applies to what a practitioner says out loud as much as to what is printed.

Where does marketing outrun the data?

At every claim that names a disease or a body system. The pattern is easy to spot once you know it: the evidence exists for feelings people can rate on a scale, and it vanishes for anything a laboratory would have to measure.

1. 1

   "Antiviral" and "antibacterial"

   Laboratory dishes, not human bodies. An oil that kills bacteria on a plate at 100 percent concentration says nothing about a 2 percent blend on skin or a few molecules in the air.

2. 2

   "Supports immunity"

   Unmeasurable and therefore unfalsifiable, which is why it is popular. No trial has shown an essential oil changes infection rates in people.

3. 3

   "Balances hormones"

   A handful of case reports link certain oils to hormone-related effects in children, in the direction of harm, not balance. No trial supports the benefit claim.

4. 4

   "Detoxifies"

   The liver and kidneys detoxify. Nothing inhaled or rubbed on changes that, and a client who hears otherwise may delay real care.

5. 5

   "Therapeutic grade"

   A trademark, not a standard. No U.S. body grades essential oils; the phrase is a marketing term owned by the seller who uses it.

How do our practitioners describe effects honestly?

In three tiers, in a fixed order: what the tradition says, what trials found, and what the client is likely to notice, with the mechanism named as smell. It sounds like this for lavender: "Traditionally used for sleep and restlessness. Trials show many people report sleeping somewhat better over a few weeks with lavender in the room; the effect is small and stops when the oil does. What you'll probably notice is that the smell becomes a cue that it's time to wind down." Fifty words, and every one defensible.

Our manual's own description of true lavender lists sleep and stress, insomnia, anxiety and mental agitation among its traditional uses, and peppermint's entry lists nausea, heaviness and digestive spasm. We keep those entries as the tradition. We add the evidence tier ourselves, oil by oil, across the 50 to 60 oils in the working repertoire, so that a graduate can say for each one whether the trials exist, and if so how big they were and what they measured.

Why train in something with modest evidence?

Because modest, honest and safe is a reasonable thing to offer, and because most of the harm in this field comes from people who never learned the limits. A client who leaves a session with a 2 percent lavender blend, a cue for winding down and an accurate sentence about what it will and will not do has been well served. A client who leaves with a bottle of oregano oil to take by mouth for an infection has been endangered.

That is why our live online aromatherapy program puts chemistry and safety before blending: 6 live online teaching days (42 teaching hours) for $2,000, with the evidence tier attached to every oil in the repertoire. It is also why we find a comparison with our review of the Reiki research useful for applicants: the honest posture is the same across modalities, whatever the size of the effect.

What does an honest first consultation sound like?

Quiet, and mostly questions. Our practitioners open with what the client wants from the session, what they are already using and what they have been told. Then intake: age, pregnancy, skin, medications, and any diagnosed condition that sends the client back to their provider before oils are discussed at all. Only then is an oil chosen, usually one, with a stated dilution and a route, inhalation first.

The close is a sentence the client can repeat to a sceptical partner: this is a smell that many people find calming, with a small amount of research behind it, used safely at a low percentage. If a practitioner cannot say that sentence without adding to it, the training was not finished. Ours ends there on purpose.