The following is a guest post by Earle Holland, who, for almost 35 years, was the senior science and medical communications officer at Ohio State University. He’s been a member of our editorial team since January.
It started as a Facebook post from a friend of a friend, a comment excitedly anticipating the arrival of a box of essential oils, the key ingredients of so-called aromatherapy. Some supporters of alternative medicine tout the benefits of aromatherapy, that it does more than simply improve moods and reduce stress, but can actually improve immunity and reduce pain.
Even the Mayo Clinic, widely respected throughout the medical community, explains on its website that studies of the use of lavender oil “may help make needle sticks less painful” and “reduce pain for children undergoing tonsillectomy.”
And as the Facebook discussion progressed, the comments became more and more effusively supportive, one respondent gleefully said that she had bought the “Family Physician Kit,” suggesting an actual medicinal effect.
In truth, it was just so much light-hearted banter among friends, nothing serious at all. But as those of us who’ve spent lifetimes in the battles between science and pseudo-science well know, such conversations can often lead to more fallacy than fact among the public.
Seven years ago, I had reported on research by a team of world-class scientists looking at the claims of aromatherapy proponents. These experts had spent more than three decades finding connections between psychological stress and problems with immunity, wound-healing, vaccine effectiveness, cancer outcomes and a host of other concerns. They had found dozens of situations where stress had increased the production of hormones and other biochemical markers that had an actual physiological effect on health outcomes.
They had turned their attention to testing two popular oils – lemon and lavender – to see if either caused changes that they could measure, differences which normally connote a change in human health. They used a standard test for both pain and for wound-healing, all the while taking blood samples throughout the tests. They looked for changes in stress hormones like cortisol, norepinephrine and other catecholamines, as well as levels of cytokines like interleukin-6 and interleukin-10, both known to vary with immune changes.
There were 56 healthy volunteers in the tests, some whom had favorable views about aromatherapy and some who had no opinion, and all completed several standard psychological tests to gauge stress levels during the research sessions. At the time, I wrote:
“While one of two popular aromas touted by alternative medicine practitioners – lemon – did appear to enhance moods positively among study subjects, the other – lavender – had no effect on reported mood, based on three psychological tests.
“Neither lemon nor lavender showed any enhancement of the subjects’ immune status, nor did the compounds mitigate either pain or stress, based on a host of biochemical markers.
“In some cases, even distilled water showed a more positive effect than lavender.”
The study was published in the journal Psychneuroendocrinology and had been supported by the National Institutes of Health. The aromatherapy industry, understandably, was not pleased.
I offered a link to that study to the participants in that Facebook discussion.
Soon after, one of the posters offered an opposing link, one pointing to a study of the use of such oils to reduce stress in the Vanderbilt University Medical Center’s Emergency Department. [Editor’s note: we could find no link to a published paper on the study, so we are linking to the report from the Facebook discussion, which in turn links to this magazine article about the study.] It looked at the use of these oils to lower perceived stress among the staff working in that ER. The two nurses conducting the study had asked staff to complete survey instruments reporting their stress levels before and after the use of the oils. Their findings seemed to clearly support the value of the oils in reducing stress:
“Before the use of essential oils, 41 percent of staff members surveyed felt work-related stress very often. After the use of essential oils, only 3 percent felt work related stress very often.”
The surveys reported similar improvements in staff’s feelings of being overwhelmed, in their energy levels, and in their perceived ability to handle stress.
So this was a case of competing studies with opposite findings, right?
And that’s the point in bringing this whole issue up. In no way were the two studies comparable. The first looked for objectively measurable biochemical markers known to change when stress is involved. Many, many previous studies covering a host of conditions had shown their effectiveness in measuring the impact of stress, as well as the health outcomes from that stress.
The Vanderbilt study was a self-reporting survey, with individuals stating their subjective perceptions of their status. There was no control group to compare with. There was no mention of the ER staff being blinded to the use of the oils. And there were no biochemical markers taken which could actually show a change in stress responses, nothing to rule out that the observed changes were anything more than a placebo effect.
And that’s the main problem with situations like this. The public is confronted with two pieces of research, studies which may be radically different in their design and ability to produce high-quality evidence. But most people are ill-equipped to gauge the differences. All too many are unwilling to investigate reports beyond a simplistic understanding.
It’s just much easier to assume that all studies are equal.