Essential Oils for Depression: Yes, Really
A Masterclass in clinical trial design
The Frontier Psychiatrists is a Health-themed newsletter. I did not think I would be writing a piece explaining an amazing study about essential oils. One of the nice things about science? I was surprised. To be clear, the study I'm about to describe to you does not describe a breakthrough treatment for depression. It describes an effective and tolerable treatment for mild depression that is a little bit better than placebo and compares well to Zoloft— in a study not powered to show that difference. What the authors do, remarkably well, is design a clinical trial that can do something remarkable: prove, to the satisfaction of this grumpy author, that treatment based on an essential oil of lavender is an active agent in the treatment of major depressive disorder. And, to my surprise, they demonstrated it in one of the hardest cohorts to demonstrate anything in mild depression!
Mild depression is not hard to treat. Mild depression is easy to treat. Almost anything will get mild depression a little better. Thus, it is remarkable to do a study in this very hard-to-research condition.
Without further ado, I will walk you through a masterpiece of research study design, and will all come out smelling like a flower, although not a rose, in the end.1
Lavender oil preparation Silexan is effective in mild-to-moderate major depression: a randomized, placebo- and reference-controlled trial
My readers are savvy by this point, I've hammered into you repeatedly the crucial nature of evaluating Table One. Table one is not laid out in the way that I like. However, I will include it below.
My readers will recall that the point of looking at table one is to make sure the randomization succeeded. There should be no statistical differences between the study groups. I've spent time in the newsletter explaining how to look at the differences between two groups before, and this study includes three. But don't worry, we're not supposed to compare all three of them. I'll explain as we go along!
In the second half of Table One, more non-differences!