Aromatherapy for menopausal symptoms: a large effect on weak evidence

Design

Systematic review and meta-analysis, 17 studies in 21 articles; randomised and quasi-experimental

Population

1,347 perimenopausal and postmenopausal women

Primary outcome

Menopausal symptom scores across vasomotor, physical, psychological and sleep domains

Effect

Overall symptoms SMD -1.00 (95% CI -1.43 to -0.58); vasomotor SMD -0.79 (-1.11 to -0.47); no effect on urogenital symptoms; GRADE low to very low

Seventeen studies reported in 21 articles, covering 1,347 women, were pooled from a search of seven databases to January 2026, under a registered protocol. Randomised and quasi-experimental designs were both eligible, assessed with the Cochrane risk of bias tool and graded with GRADE.

The effect estimates are large across the board. Overall menopausal symptoms improved with a standardised mean difference of -1.00 (95% CI -1.43 to -0.58), vasomotor symptoms -0.79 (-1.11 to -0.47), physical symptoms -0.78 (-1.10 to -0.46), psychological symptoms -0.59 (-0.96 to -0.23), anxiety -0.63 (-1.10 to -0.16) and depression -0.57 (-0.85 to -0.28). Sleep quality improved by a mean difference of -3.95 (-5.94 to -1.97) and sexual function by 7.05 (3.64 to 10.45). Urogenital symptoms did not change.

And the certainty is low. GRADE was rated 'very low' for sexual function and sleep quality and 'low' for everything else; the authors state plainly that firm clinical recommendations cannot be made. That combination - big effects, weak evidence - is the signature of small unblinded trials of a pleasant intervention measured by self-report. What it supports in clinic is permission rather than prescription: if a woman is using aromatherapy and finds it helps, there is no reason to talk her out of it, and no basis for offering it instead of treatment that works.

• Do not offer aromatherapy as an alternative to hormone therapy or to a non-hormonal drug that has evidence
• Record what a woman is already using, including oils, so symptom changes can be interpreted
• Note that urogenital symptoms did not improve - vaginal dryness still needs local treatment
• Blinding is close to impossible with a scented intervention; treat self-reported outcomes accordingly
• Useful framing for a woman who wants to try something before committing to hormone therapy

Why it matters

Patients raise this in clinic and deserve an answer more precise than approval or dismissal.

Don't overread it

GRADE certainty was low or very low throughout, and the authors say firm clinical recommendations cannot yet be made.

The statistics, in plain English

A standardised mean difference of -1.00 is conventionally a large effect, and that should provoke suspicion rather than enthusiasm in a set of small trials where participants know what they received. GRADE ratings of 'low' and 'very low' mean the reviewers judged the true effect likely to be substantially different from the estimate. The single negative result - no effect on urogenital symptoms - is arguably the most informative one in the analysis, because it shows the measurements were capable of returning a null, which makes the uniformity of the positive findings elsewhere harder to explain by bias alone but still does not establish them.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.