STAR LABA result for a mixture cannot identify which ingredient caused it.
The 2025 formulation study
A randomised crossover study enrolled 125 people with insomnia and compared an oral CBD formulation containing eight terpenes with placebo. The measured combined share of slow-wave and REM sleep rose by about 1.3 percentage points on average. Sleep stages were estimated by a wrist-worn device, and the authors called for confirmation with polysomnography. The paper appeared in the January 2025 issue after online publication in 2024. [1]
Why this does not prove “myrcene for sleep”
The active formulation contained CBD plus multiple terpenes, including myrcene and linalool. It did not isolate either molecule’s effect. It also did not test live hash rosin. Calling the study a demonstration that one terpene treats insomnia would exceed its design.
Lavender is also a mixture
A 2021 trial involving 35 postmenopausal women compared lavender essential-oil inhalation with a control. It did not find a statistically significant between-group improvement in the primary sleep-quality outcome. Improvements within a group are not equivalent to superiority over a control. Lavender findings cannot simply be relabelled as evidence for isolated linalool. [2]
The useful conclusion
The studies justify further investigation, not a universal “best sleep terpene” list. When reading a claim, ask whether the experiment measured sleep onset, duration, sleep stages or subjective quality. Those outcomes are not interchangeable. Persistent insomnia deserves clinical assessment; this educational series is not a recommendation to self-treat with concentrates or essential oils.
Sources & further reading
- J Clin Sleep Med (2025; online 2024) — Effects of a cannabidiol/terpene formulation on sleep in individuals with insomnia
- Lucena et al. (2021) — Lavender essential oil on postmenopausal women with insomnia: Double-blind randomized trial
Research findings and public-agency definitions are attributed above. Reading checklists are editorial guidance, not laboratory findings.