Reiki has been studied more than most complementary practices, including in randomised trials with sham controls, which makes it unusually assessable.
The findings are worth setting out because they're reasonably consistent and because they illustrate a general point about how such practices are evaluated.
What the practice involves
A practitioner places hands on or near a recipient, in a sequence of positions, generally for a session lasting under an hour. The recipient lies clothed and still.
The stated mechanism involves transmission of a form of energy, channelled by the practitioner, which is understood to promote healing.
Practitioners receive attunement from a teacher, which is understood to confer the ability.
The trial design
What makes this assessable is that a sham condition is possible. Someone with no training can perform the same hand positions, for the same duration, in the same setting.
The recipient cannot distinguish the two, since nothing observable differs.
Which permits a genuine comparison between the specific practice and everything surrounding it.
What the trials found
The general pattern across reviews of this literature: recipients frequently report benefits — reduced anxiety, reduced pain, improved wellbeing — and these benefits are not reliably greater than in the sham condition.
Systematic reviews have concluded that evidence is insufficient to support reiki as an effective treatment for any condition, while noting that studies frequently report improvements in both arms.
Which is the important finding. Something is happening; it appears not to be the thing practitioners describe.
What's producing the effect
The candidates are the same contextual factors identified in placebo research generally.
Lying still for an hour. Uninterrupted, with no demands, in a quiet room. For many people this is genuinely unusual and produces measurable relaxation on its own.
Attention from another person. Sustained, focused, non-judgemental attention. Research on therapeutic relationships suggests this has effects independent of technique.
Touch. Where hands rest on the body, this is a form of non-threatening physical contact, which has documented physiological effects.
Expectation. Recipients arrive expecting benefit, from a practitioner presenting confidence, in a setting designed to signal care.
Ritual structure. Defined beginning and end, sequence, and the sense of something being done.
Each of these is real. Together they account for a considerable amount, and the trials suggest they account for essentially all of the observed effect.
The practitioner response
Common objections and their weight.
That the sham practitioner may inadvertently channel energy. If this is possible, the practice cannot be tested, and an untestable claim isn't a claim about the world.
That effects are subtle and long-term. Trials with longer follow-up have been conducted and haven't produced different conclusions.
That the studies were too small. Some are, and the pattern across the literature is consistent, which is more informative than any single study.
That it works alongside conventional treatment rather than instead. This is a reasonable position and it changes the claim considerably — from a treatment to a supportive practice, which is compatible with the findings.
Where this leaves it
The honest position has two parts and they're both true.
People who receive reiki frequently feel better afterwards. That's a real experience and the studies confirm it.
The improvement appears to come from the context rather than from anything specific to the practice. Which means the same benefit is available from anything providing those elements.
Whether that matters depends on the setting. In a hospice, where the purpose is comfort, a practice reliably providing relaxation and attention has value regardless of mechanism. Several healthcare settings offer it on exactly these grounds.
Where it matters is if it's presented as treatment for a condition requiring medical care, or if it displaces effective treatment. That's where harm occurs, and it's documented.
The general lesson
This case is instructive beyond the specific practice.
A treatment that produces improvement is not necessarily working through its stated mechanism, and testing against a sham that preserves the context is how you find out.
The frequent response — that improvement proves efficacy — misses that improvement occurs in the sham arm too. Everybody feels better; the question is what made the difference, and the answer here appears to be nothing.
Which is worth knowing without being dismissive. An hour of quiet attention is a good thing to receive, and it's available without any theory attached.
General information only. Complementary practices should not replace assessment or treatment by a qualified healthcare professional.
The hospice use
Worth addressing because it is where the practice is most defensible and where the objections have least force.
Several palliative care settings offer this and comparable practices, on the stated basis of comfort rather than treatment.
In that context, the mechanism question matters considerably less. The purpose is that somebody who is dying receives an hour of quiet, attentive, unhurried contact from a person whose entire attention is on them. That is a good thing to receive and the trials confirm that recipients report benefit.
What would be objectionable is presenting it as affecting the disease. Presented as comfort care, it is doing exactly what it does, and nobody is being misled.