I listen to NPR nearly every morning, and on Saturdays that means Living on Earth, at least on my local station. This Saturday I tuned in partway through a segment on “forest bathing,” also known as Shinrin-Yoku. As the host, Steve Curwood, describes it, forest bathing is a practice popular in Japan and China “in which practitioners spend meditative time breathing in nature.” The interviewee, Moshe Sherman, throws out several red flags typical of the evidence-challenged alternative medicine crowd, but he also makes some pretty specific health claims that he says are backed up by empirical evidence.

I’m going to argue that a lot of this is baloney. I want to be clear up front, though, that I’m not saying that walking in the woods and meditating in nature aren’t good for you. What I am saying is that the evidence that there’s something special about Shinrin-Yoku, something that provides benefits beyond those of exercise and relaxation, is lousy.

Sherman is “…a medical Qi Gong therapist who leads regular forest bathing outings.” In an interview with the Allegheny Front‘s Kara Holsopple , he describes the outings thus:

…when we first bathe, we want to look all around and just take it in. So we breathe in the healing energy. We also bring it in through our eyes, we bring it in through our ears, hearing the sounds we bring in through feeling whether it’s feeling the weather or touching a leaf…in the urban environment we often have to put up what I call energetic shields to protect us.

I would charitably assume that Sherman is being metaphorical here, if he hadn’t said later in the interview,

I work with Chi or energy, which is a little more abstract, but I think of each tree and each plant as being a unique life form with its own consciousness.

Chi doesn’t exist, so this is someone with a loose grasp of reality. It doesn’t really matter if he has some weird beliefs, though; what matters is whether his claims are supported by evidence. So what are his claims?

..[forest bathing] really impacts so many systems of the body, cardiovascular system, the respiratory system, and most profoundly, I think, to me right now is how it impacts the immune system. If you do one session of forest bathing for even 20 minutes, we see an increase in what’s called the NK cells are the natural killer cells. And these are the cells that protect us from viruses and even from tumor formations.

As evidence, he refers to a systematic review:

…it has been widely researched mostly in Asia and also in Europe. And what brought the most recent attention to the United States is a review that was published in summer of 2017 that looked at 64 of those research studies which are all based on empirical evidence…

The review he cites is that of Margaret Hansen, Reo Jones and Kirsten Tocchini in the International Journal of Environmental Research and Public Health. IJERPH is published by MDPI, which is, I think it’s fair to say, controversial. Briefly, they’ve had some high-profile failures, they were included in Jeffrey Beall’s list of predatory publishers but successfully appealed and were removed, and the Open Access Scholarly Publishers Association investigated them and found them legit. The full story is a rabbit hole I don’t want to go down; Wikipedia has a summary.

The review article does include studies reporting the benefits Sherman claims, but it also points out that these studies have substantial shortcomings*:

Of the studies included within the PP [Physiological and Psychological] section, and irrespective of study aims, there was a trend towards small sample sizes, gender and age homogeneity, and skewed ratios of females to males/vice versa, which by either methods of convenience, purpose and/or imparted bias to the research…

Self-reported manual BP reports are subject to reliability issues of measurement and bias. Moreover, the lack of additional objective analysis in the short-term forest-environment exposure are identified confounding variables, as well as the fact the participants took prescribed antihypertensive medications throughout the trials in the intervention and control groups respectively…

…inherent to the research aim of investigating the previous successes of nature-based therapy is an implicit bias toward the functionality and reliable successes of the research outcomes.

As with several SY [Shinrin-Yoku] studies, specific to psychological indices, self-reported measures in the form of questionnaires, such as the SD [Semantic Differential] method, provided a bulk of the data. Therefore, leaving a question about the consistency, validity and reliability of the reported psychological outcomes…

Limitations of this review include the biases among the authors of the studies and articles culled, as well as the conceivable restrictions of SY [Shinrin-Yoku] as an evidence based practice within the traditional principles of Western medicine…publication bias must be acknowledged as most of the studies reviewed demonstrate a positive correlation among SY practice and NT [Nature Therapy] with favorable physiological and psychological outcomes. In addition, original study sample sizes were often limited to less than 20 participants, with the inclusion criteria of primarily of healthy, young, male university students, making results difficult to generalize across entire populations. Other limitations within the studies include their inability to distinguish physical and psychological effects purely based on the participants’ surroundings versus the participants’ level of activity while present in either an urban or nature setting.

Not surprisingly, Sherman doesn’t mention any of this in the interview (to be fair, it’s possible that he did and that it was edited out). The bottom line is that most of the studies reviewed are of poor quality, with small sample sizes, homogeneous subjects, and improper or no controls. Most of the reported benefits are either subjective or nonspecific (or both). Subjective benefits are worthless (as data) without proper placebo controls. By nonspecific, I mean things like cortisol and adrenaline levels, blood pressure, and heart rate, which are generally improved by exercise and relaxation. Sure, forest bathing probably does improve those measures, but few of the reviewed studies included controls that would distinguish specific benefits of forest bathing from general benefits that could be achieved by any activity that combined exercise and relaxation.

The claim I found most striking, though, was that

If you do one session of forest bathing for even 20 minutes, we see an increase in what’s called the NK cells or the natural killer cells.

The Hansen et al. review article doesn’t directly address this, but an earlier review article in the same journal, by Chorong Song, Harumi Ikei, and Yoshifumi Miyazaki, does:

Three studies by Li et al. [32,33,34] demonstrated a boosting effect of forest therapy on weakened immune functions.

References 32, 33, and 34 are Li et al. 2007, 2008a, and 2008b, respectively (see Stable links below). The three studies followed essentially the same experimental procedure, the first two each with twelve male subjects, the third with thirteen female subjects. I’ll quote just the first, and I’ll point out some differences between that one and the other two:

The subjects experienced a three-day/two-night trip in three different forest fields. On the first day, subjects walked for two hours in the afternoon in a forest field; and on the second day, they walked for two hours in the morning and afternoon, respectively, in two different forest fields. Blood was sampled on the second and third days, and NK activity; proportions of NK, T cells, granulysin, perforin, and granzymes A/B-expressing cells in PBL [peripheral blood lymphocytes] were measured.

Unfortunately, the first and third studies employed no useful controls: the various measures were simply compared before, during, and after the forest bathing in the same twelve or thirteen people. The second study did employ a control:

The forest bathing trip significantly increased NK activity and the numbers of NK, perforin, granulysin, and granzyme A/B-expressing cells and significantly decreased the concentration of adrenaline in urine. The increased NK activity lasted for more than 7 days after the trip. In contrast, a city tourist visit did not increase NK activity, numbers of NK cells, nor the expression of selected intracellular anti-cancer proteins, and did not decrease the concentration of adrenaline in urine.

But again, it was the same group of people. It’s not clear to if the two groups of men (in the first and second studies) are completely non-overlapping: the age ranges are very similar, and one group was selected from “three large companies” in Tokyo, the other from “four large companies” in Tokyo. Either way, this is not how randomized controlled trials are done. Having the same group of subjects experience both treatments (all in the same order) is a crappy control, if for no other reason than because the effect of the second treatment may be influenced by the first.

All three studies found an increase in natural killer cells, exactly as Sherman said (though I don’t know where he got the 20 minutes). However, all three had small sample sizes, and all three were poorly controlled. More importantly, though, only the second study made any effort to compare the effects of forest bathing with an equivalent amount of exercise and relaxation in another context. Even if the results of the second study are reliable, would it be so strange to find that time in the forest was more relaxing than time in the city?

This is the weakness that runs through all of these studies: they don’t ask whether forest bathing has specific benefits, i.e., those that couldn’t be obtained through other activities that combine exercise and relaxation. Furthermore, claims to “enhance immune response” in a general way in healthy people are almost always nonsense. It’s not even clear that doing so would be a good thing. Even if we accept that forest bathing reliably increases natural killer cell activity, which we can’t really tell from these studies because of small samples and poor controls, and even if we think that this response is specific to forest bathing, there’s no indication in any of these papers that increased natural killer cell activity improves health outcomes.

Walking in the woods is probably good for you. I do a good bit of it myself (245 miles last year). You get some exercise, and in most cases some relaxation as well. Is it better for you than, say, walking on the beach, going for a run then soaking in a hot tub, or doing yoga and meditating? There’s no evidence that it is. If you like meditating in the forest, by all means, do it. But there’s no need to carry pre-scientific, mystical baggage like energy shields and chi with you. There’s really no need to call it “therapy” at all.

 

* It’s also nearly unreadable because of the proliferation of abbreviations: SY, NT, CNT, EBM, GS, FT, EBP, PP, SM, F, CNS, HRV, SCL, IgA, HR, PR, SBP, DBP, HRV, LVF, RVF, PNS, HF, SNS, LF, ANS, HTN, CAD, COPD, CV, RCT, CBT, POMS, ROS, SVS, BDI, HAM-D17, STAI, CWP, VAS, NDI, and on and on, leading to such gems as “Significant research has revealed the effects of SY and NT on specific physiological disease states, including HTN, CAD, COPD, and Diabetes Mellitus Type II (DMII).” This does not inspire my confidence in the rigor of the peer review this paper received.

 

Stable links:

Hansen, M. M., Jones, R. & Tocchini, K. 2017. Shinrin-yoku (Forest bathing) and nature therapy: A state-of-the-art review. Int. J. Environ. Res. Public Health 14, 851. doi: 10.3390/ijerph14080851

Li, Q. et al. 2007. Forest bathing enhances human natural killer activity and expression of anti-cancer proteins. Int. J. Immunopathol. Pharmacol. 20, 3–8. doi: 10.1177/03946320070200S202

Li, Q. et al. 2008a. Visiting a forest, but not a city, increases human natural killer activity and expression of anti-cancer proteins. Int. J. Immunopathol. Pharmacol. 21, 117–127. doi: 10.1177/039463200802100113

Li, Q. et al. 2008b. A forest bathing trip increases human natural killer activity and expression of anti-cancer proteins in female subjects. J. Biol. Regul. Homeost. Agents 22, 45–55. ncbi.nlm.nih.gov/pubmed/18394317

Song, C., Ikei, H. & Miyazaki, Y. 2016. Physiological effects of nature therapy: A review of the research in Japan. Int. J. Environ. Res. Public Health 13, 781. doi: 10.3390/ijerph13080781