Twenty-One Nurses, No Control Group: What a Reiki Pilot Study Can and Cannot Tell Us

Key Takeaways

• A 2025 pilot study of Reiki with twenty-one Japanese nurses found that psychological stress measures improved significantly, while physical measures did not.
• The study had no control group at all, which places a hard limit on what its results can be used to claim.
• Reading a study honestly means naming both what it found and what its design cannot rule out.

Key Terms Explained

• 靈氣 (Reiki) / Universal Energy — the traditional Japanese practice of connecting with universal energy so that life settles.
• 手当て (Teate) / Hands-On — the ordinary Japanese word for placing the hands on someone, and the root of how Reiki is given.
• 気付き (Kizuki) / Noticing, Awareness — becoming aware of something that was already there, a word that appears often in Japanese descriptions of what a session brings.

The Study That Was Done in My Own Country

Most of the Reiki research I read comes from the United States, Turkey, or Brazil. So when a study appears with Japanese participants, done in Japan, I read it more slowly.

This one was published in the Journal of Holistic Nursing in 2025, by Yu Morimitsu, Miwa Nakahira and Akiko Nishikawa. Twenty-one nurses received a Reiki session. Before and after, the researchers measured pulse, respiration rate, blood pressure and salivary alpha-amylase activity. They also used the Profile of Mood States, Second Edition, to measure mood dimensions and overall mood disturbance. Afterwards, they interviewed the participants, and sent a web questionnaire the following day.

The result, stated plainly: psychological stress reactions improved significantly. Physical stress measures showed no significant difference. No adverse events were reported.

The qualitative half produced two categories, positive effects and negative effects. Among the positive codes were "warm," "body feeling more comfortable," and "stress awareness."

That last one stopped me. 気付き (Kizuki) / Noticing — the participants described becoming aware of stress they had been carrying without registering it. Not that the stress went away. That they noticed it.

This is the third study I have written about here in as many days, and it is the one with the weakest design. I want to be direct about that rather than skip past it.

Related: Ten Minutes of Reiki for 1,724 First Responders: What a Big Community Study Does and Doesn't Show explains this in detail.

What This Design Can and Cannot Support

It Had No Control Group

This is the central limitation, and everything else follows from it.

Twenty-one nurses received Reiki. There was no second group who received a sham treatment, no group who simply rested for the same period, and no group who received nothing. Every participant got the intervention.

That means the study cannot separate Reiki from the other things that happened in that room. Lying down for a set period. Someone giving you their full attention. Being asked, twice, how you feel. Any of these might move a mood score.

The two studies I looked at earlier this week both had comparison groups — one with a sham condition, one comparing against usual care. This one has neither. The results are therefore not evidence that Reiki caused the improvement. They are evidence that the improvement happened.

Related: 10-Minute Reiki Study: What the Chicago Data Shows explains this in detail.

The Sampling Was Not Random

Participants were recruited through nonprobability snowball sampling, meaning existing participants pointed the researchers toward further participants.

This is a practical method for reaching a specific professional group, and there is nothing improper about it. But it tends to gather people who are open to the thing being studied. Nurses who think Reiki is nonsense are unlikely to be passed along the chain.

Related: What 1,284 Reiki Sessions Actually Show: An Honest Read of a New Study on the Subjective Experience explains this in detail.

Twenty-One People Is a Pilot, and It Says So

The word "pilot" is in the title, and it is doing real work there. A pilot study exists to find out whether a larger study is feasible and worth running — whether the measurements work, whether participants complete the protocol, whether anything unexpected occurs.

Judging a pilot as though it were a definitive trial is unfair to the researchers, who never claimed it was one.

The Split Result Is the Interesting Part

Set aside the design limits for a moment, and one finding stands out: the psychological measures moved and the physical ones did not.

Salivary alpha-amylase, pulse, respiration and blood pressure all stayed statistically unchanged. Mood measures improved. If something purely psychological was at work — expectation, rest, attention — this is roughly the pattern you would predict.

I do not find that deflating. It is close to what I would say from twenty years of practice if someone asked me honestly what a session does. People get up feeling calmer and more settled. I have never claimed their blood pressure changed, and this study did not find that it did.

Why I Would Rather Read the Weak Study Than Skip It

There is a habit among people who practise something to cite only the research that supports it. I understand the pull. But it produces a body of writing that no thoughtful reader trusts, and I would rather be trusted.

My background before Reiki was in IT, and in that world a result without a control condition is not a result yet. That instinct has not left me, and I do not think it should. When I read that twenty-one self-selected nurses felt better after lying down for a Reiki session, my first thought is that I would expect twenty-one self-selected nurses to feel better after lying down for almost anything.

And yet I do not think the study is worthless, for two reasons.

The first is that it recorded no adverse events. Safety findings survive weak designs better than efficacy findings do, because nothing about the absence of a control group would hide harm if harm had occurred.

The second is the qualitative half. The interviews were not trying to prove that Reiki works. They were asking what the experience was like, and "stress awareness" came back as a theme. That is a description worth having, and no controlled trial design would have surfaced it.

I think this is the honest position: the numbers here cannot tell us whether Reiki did anything beyond rest and attention. The interviews tell us something about what receiving it was like for nurses in Japan. Those are different kinds of knowledge, and only one of them needed a control group.

FAQ

Q: Does this study show that Reiki reduces stress in nurses? A: It shows that psychological stress measures improved in twenty-one nurses who received a Reiki session, and that physical measures did not change. Because there was no control group, it cannot show that Reiki was the cause of that improvement rather than rest, attention, or expectation. The researchers described it as a pilot study, which is the appropriate weight to give it.

Q: Why do the physical measurements matter if people felt better? A: They matter mostly for honesty about what is being claimed. If someone says Reiki lowers blood pressure or changes a physiological stress marker, this study does not support that. It found no significant change in pulse, respiration, blood pressure or salivary amylase. Feeling calmer is worth having on its own, and it does not need a physiological finding attached to it.

Q: Should nurses use Reiki for their own stress at work? A: That is entirely a personal decision, and Reiki is not medical care or a treatment for workplace stress. What this study reports is that the participants described the session as warm and comfortable, that some became more aware of stress they were carrying, and that no adverse events occurred. Anyone considering it should treat it as something that sits alongside rest, support and proper working conditions rather than in place of any of them.

Key Insights to Remember

• A study without a control group can still be worth reading, but only if you are precise about which parts of it survive that limitation. Safety findings and participant descriptions survive. Claims about cause do not. Reading research honestly is mostly a matter of knowing which sentences you are allowed to carry away.
• The split between improved psychological measures and unchanged physical ones is what I would have expected, and I find that reassuring rather than disappointing. It matches what I would say about a session from my own practice: people get up calmer. Anything more than that would be a claim I could not stand behind.
• The most interesting result came from the interviews, not the instruments. Participants described becoming aware of stress they had not registered before — 気付き (Kizuki) / Noticing. No measurement in the study was designed to capture that, and no controlled design would have found it. Some things about a practice are only visible when you ask people what it was like.

Sources

• The Effects of Reiki on the Stress of Japanese Nurses: Mixed Methods Pilot Study — PubMed
• The Effects of Reiki on the Stress of Japanese Nurses: Mixed Methods Pilot Study — Journal of Holistic Nursing (SAGE)
• Reiki: What You Need To Know — National Center for Complementary and Integrative Health (NCCIH)