Essential Oils for Anxiety: What the Trials Actually Show

Essential oils do have measurable anxiety effects, but the evidence is smaller than the marketing suggests and it is unevenly distributed. Lavender carries almost all of it. The strongest lavender data comes from an oral standardised capsule, not from a diffuser. The other seven oils range from modest to untested.

📊 What the Evidence Actually Shows

A 2023 systematic review and network meta-analysis pooled 44 randomised trials, 50 study arms, 10 essential oils and 3,419 people with anxiety. Essential oils did reduce state anxiety scores, with a pooled weighted mean difference of about −6.6 points on the State-Trait Anxiety Inventory. Frontiers in Public Health, 2023

That is a real effect and a modest one. Two caveats matter more than the number:

• Blinding is nearly impossible. People can smell. Most trials cannot hide which group received the oil, which inflates reported benefit.
• Settings skew the results. A large share of these trials measured anxiety before surgery, during labour or after a C-section — situational anxiety that settles on its own. That is not the same as treating an anxiety disorder.

So the fair summary is: aromatherapy is a reasonable comfort measure with a small measurable effect, and it is not a treatment for generalised anxiety disorder. For context on where the line sits, see anxiety versus an anxiety disorder.

Why a smell could affect anxiety at all

Olfactory signals travel directly to the limbic system rather than routing through the thalamus, which is why a scent can shift arousal before you have named it. That anatomical shortcut is real and it explains the speed of the effect. It does not explain a lasting one.

The more likely mechanism in everyday use is associative: a familiar smell becomes a cue for a state you have practised. That is why pairing an oil with slow breathing works better than the oil alone, and why the scent you find pleasant outperforms the one with the better trial result. Expectation is not a flaw in these studies — it is part of the effect being measured.

🧮 The Eight Oils, Ranked by Evidence

🔥 Tier 1: Lavender

Lavender is the only oil on this list with a body of work big enough to review on its own. A systematic review and meta-analysis of lavender in any form found effects on anxiety across randomised and non-randomised studies, while noting the usual problems of small samples and weak blinding. Donelli, Phytomedicine, 2019

The most interesting trial is also the one most often misquoted. A multi-centre, double-blind study compared Silexan — a standardised oral lavender oil capsule — with lorazepam over six weeks in adults with generalised anxiety disorder. The lavender preparation performed comparably on the Hamilton Anxiety Rating Scale. Woelk, Phytomedicine, 2010

For inhalation, the results are consistent but smaller: reductions in state anxiety in procedural settings, with effects that appear within minutes and fade with the scent. That is a comfort effect worth having, not a treatment.

🌿 Tier 2: Bergamot, Chamomile and Rose

These three have genuine randomised trials, usually small and usually in a narrow context.

• Bergamot. A placebo-controlled crossover trial in 48 university students found improved sleepiness and refreshment on waking after using bergamot spray at bedtime and on waking, alongside changes in stress measures. Wakui, Complementary Therapies in Medicine, 2023 It was open-label, so expectation cannot be excluded.
• Chamomile. A 2024 randomised trial tested inhaled chamomile on anxiety and haemodynamic measures, reporting reductions in the intervention group. BMC Complementary Medicine and Therapies, 2024 Animal work on the neural pathway exists, but that is not human evidence of anxiolysis. Zhu, Advanced Science, 2025
• Rose. A 2024 trial compared Damask rose and chamomile aromatherapy and found both reduced anxiety compared with control. Journal of Perianesthesia Nursing, 2024

Aromatherapy also shows up in supportive cancer care, where a systematic review found reductions in anxiety and depression among patients receiving treatment. Li, Frontiers in Public Health, 2022 That is a population where comfort interventions matter and where the risk of harm is low.

What tier 2 means in practice

If lavender does nothing for you, these are reasonable alternatives with at least some trial support. If you are choosing between them on evidence, you are choosing within noise. Pick the scent you find pleasant, because the plausible mechanism — a learned relaxation response to a familiar smell — works better when you actually like it.

🧱 Tier 3: The Oils With Thin Evidence

Frankincense, ylang-ylang, clary sage and vetiver appear on every list of anxiety oils. They are traditional, they smell distinctive, and the human trial base for anxiety is either tiny or absent.

• Frankincense is studied mostly for inflammation and in animal models; anxiety trials in humans are scarce and small.
• Ylang-ylang has a handful of small studies, often with blood pressure or mood endpoints rather than clinical anxiety.
• Clary sage has been examined in labour and menopause contexts, where anxiety is one of several measured outcomes.
• Vetiver is largely supported by traditional use and very small studies, not replication.

That does not make them useless. It makes them unproven, which is a different claim. If a product page tells you vetiver “calms the nervous system”, it is describing a tradition, not a trial result.

💧 How to Use Them Properly

1. Inhalation is the lowest-risk route. A few drops on a tissue, a diffuser in a ventilated room, or a personal inhaler. Effects are rapid and short-lived, which suits situational anxiety, and pairing it with slow movement gives the cue something to attach to.
2. Dilute for skin contact. Two to three drops per 10 ml of carrier oil, so roughly 1–2%. Patch-test on the inner forearm and wait 24 hours before wider use.
3. Do not ingest. Unless it is a licensed oral preparation such as Silexan, essential oils are not food. Several are genuinely toxic when swallowed, particularly in children.
4. Use it as a cue, not a cure. Pair the scent with slow breathing — roughly four seconds in, six out — so the oil becomes a trigger for a response you already have. The breathing does the physiological work. Vagus nerve regulation explains why.

📦 Buying: What Separates a Good Bottle

Essential oils are an unregulated market in most countries, and the label rarely tells you what you need to know. Four checks do most of the work.

Two things to ignore: “therapeutic grade”, which is a marketing term with no legal definition, and any claim that a product is safe to ingest because it is pure. Purity and safety are different questions, and the second one depends on the oil.

⚠️ Safety, and the Risks People Skip

Essential oils are concentrated plant chemistry, not scented water. The common problems are predictable.

• Phototoxicity. Bergamot, lemon and other cold-pressed citrus oils can cause phototoxic skin reactions when applied before sun exposure. Documented since the 1980s, and still the most common avoidable injury. Archives of Dermatological Research, 1985
• Irritation and sensitisation. Undiluted application, repeated use of the same oil, and old oxidised bottles all increase reaction risk.
• Pregnancy and breastfeeding. Evidence for safety is thin, and some oils are traditionally avoided. Treat this as a category to discuss with a clinician rather than a wellness choice.
• Children. Topical and inhaled use in young children carries higher risk of accidental ingestion and respiratory irritation. Keep bottles out of reach and do not diffuse heavily in small rooms.
• Pets. Cats in particular lack certain liver enzymes and can be harmed by diffused oils and by skin contact. Ventilate, and let the animal leave the room.
• Asthma and sensitivity. Strong scents can trigger symptoms in people with reactive airways.

Interactions and medication

There is very little formal interaction data for aromatherapy, which is itself the point: absence of evidence is not reassurance. The theoretical concerns are enzyme induction from repeated exposure to certain terpenes and additive sedation when oils are combined with benzodiazepines, sleep aids or alcohol. If you take psychiatric medication, mention the oils you use rather than assuming a scent is inert.

🏥 When Scent Is Not the Answer

• Anxiety most days for six months or more, which meets the threshold for generalised anxiety disorder
• Panic attacks with physical symptoms that mimic cardiac events
• Avoidance behaviour that is shrinking your life — leaving the house less, declining work or social events
• Anxiety alongside low mood, hopelessness or thoughts of self-harm, which needs urgent assessment

The practical role for essential oils is adjunctive comfort: something to make a difficult moment more tolerable while the actual treatment does its work. That is a legitimate use. It is not a substitute.

What actually has stronger evidence

Cognitive behavioural therapy for anxiety has decades of trial support and effect sizes these oils do not approach. Exposure-based work, structured breathing practice and, where indicated, medication all outperform aromatherapy on every clinical endpoint that matters. None of that makes a diffuser pointless — it makes it a comfort measure alongside treatment, which is exactly how it should be positioned.