Qigong’s evidence problem starts with its definition

In 2019 a panel of rheumatologists, therapists and patients sat down to grade every treatment anyone had proposed for osteoarthritis. Their report, issued jointly by the American College of Rheumatology and the Arthritis Foundation, put tai chi in its top recommendation tier. Then it reached qigong and stopped. Other mind-body practices, the American College of Rheumatology panel wrote, “could not be assessed due to insufficient evidence, as well as a lack of standard definitions of certain interventions (hypnosis, qi gong).”

Read that sentence twice, because it contains two different problems. One is ordinary: not enough good studies. The other is unusual and more serious: the panel could not settle on what the thing being studied was. You cannot grade an intervention you cannot define, no matter how many trials pile up.

That is the honest headline for qigong, and it is not a dismissal. Qigong is a real family of practices with real history and a real, if thin, research record. It is also a word that covers a gentle breathing exercise done in a chair and a practitioner holding their hands near a stranger’s head. Those are not the same claim, and almost everything confusing about qigong’s reputation follows from treating them as one.

Two practices share one word

The split researchers use is internal versus external, and it is worth learning before anything else.

Internal qigong is a person doing the exercises themselves: slow movement, deep breathing, directed attention. A 2025 pilot study at the University of California, Davis, describes this form as movement-based and similar to tai chi. Everything else in this article refers to that.

External qigong is a practitioner treating someone else. The Davis researchers describe it as a form of non-touch energy healing and group it with reiki and therapeutic touch under the heading of biofield therapies, which rest on the premise that a subtle energy surrounds and permeates the body and can be engaged to assist healing. In their trial the practitioner’s hands stayed at least six inches from the participant’s body. The energy-healing modalities are covered in their own article in this library; what matters here is that external qigong belongs with them and not with the movement practices.

Why this boundary needs policing: the external literature produces numbers that look spectacular. A 2007 systematic review in The Journal of Pain found five randomized trials of external qigong for pain, every single one favouring qigong, and pooled two of them to a difference of 36.3 millimetres on a 100-millimetre pain scale, from a combined 80 people. That is better than a third of the entire scale, from a treatment in which nobody is touched. An effect that large, from a sample that small, with no negative trials anywhere in the set, is the fingerprint of publication bias rather than of a powerful remedy. The reviewers, from a group generally sceptical of complementary medicine, nonetheless called the evidence encouraging and asked for more studies, which tells you the problem sits in the trials and not in the reviewers.

Where the boundary cannot be sourced is instructive. The federal government’s own qigong page, which the National Center for Complementary and Integrative Health last updated in February 2022, never defines internal or external qigong. The words appear on that page only inside the titles of cited studies. The split is standard in the research literature and absent from the agency a consumer would check first.

The category was assembled in the 1950s

Qigong is routinely sold as five thousand years old. The techniques are genuinely ancient. The category is not, and the difference explains the definitional mess.

David Palmer, the leading academic historian of the subject, dates the individual techniques to Chinese texts of the sixth century BC or earlier. The single word qigong, used as an umbrella for all of them, dates to the middle of the twentieth century, when health authorities in the newly founded People’s Republic gathered scattered religious, martial and medical exercises and reorganized them as a secular, scientific system. A government bureau drew the boundaries of the category, in the 1950s, for administrative reasons.

What followed was a cycle of suppression and revival that no other practice in this field has lived through. Qigong was banned during the Cultural Revolution from 1966 to 1976, then reappeared in the late 1970s. Through the 1980s and 1990s, with most organized religion destroyed or tightly controlled, it became the main legal channel for China’s spiritual traditions. Hundreds of groups formed. Leading Chinese scientists and senior Communist Party officials backed the movement and helped it grow. The two largest organizations, Zhong Gong, founded in 1987, and Falun Gong, founded in 1992, each claimed tens of millions of followers at their peak.

Then it ended. Authorities tried tighter regulation in the mid-1990s amid accusations of quackery and unease about charismatic masters. On April 25, 1999, roughly ten thousand Falun Gong practitioners stood silently around Zhongnanhai, the Communist leadership compound in Beijing. On July 22 that year the government banned the group as an evil cult. Most popular organizations were dismantled afterward, and only five sets of exercises stayed authorized, all drawn from the medical daoyin tradition.

Keep the political history in proportion. It is not an argument about whether breathing exercises help anyone. It is the reason a research panel in 2019 could not find a standard definition: a category assembled by a bureaucracy, inflated into a mass movement, then dismantled, does not leave behind an agreed curriculum.

Is tai chi’s evidence also qigong’s evidence?

This is the question everything turns on, because tai chi has the better trial record and qigong is frequently described as its parent. The federal qigong page says that tai chi, practised for health, counts as a form of qigong. A widely cited 2010 review in the American Journal of Health Promotion went further, examining 77 randomized trials of both and arguing that the two have shared roots, shared proposed mechanisms and shared expected benefits, and should be recognized as similar and equivalent.

That is a thesis, argued by its authors, and should be attributed rather than repeated as settled. Three things cut against it.

The first is indexing. The National Library of Medicine, which catalogues the world’s biomedical literature, maintains Tai Ji and Qigong as two separate subject headings. Search either term in PubMed and your query maps to a different controlled vocabulary term. The largest biomedical index on earth does not treat them as one intervention.

The second is that the reviews disagree with each other about whether to pool them. Some merge tai chi with Baduanjin, a well-known qigong form, into a single experimental category and report combined results, which cannot support a statement about either practice alone. Others keep them apart and get different answers. An overview of 27 systematic reviews in chronic respiratory disease, covering about 37,000 participants with searches to November 2022, concluded that in adults with chronic obstructive pulmonary disease qigong improved breathlessness, exercise endurance, lung function and quality of life, while tai chi and yoga improved endurance only. Separating them changed the result.

The third is the rheumatology panel itself. It had tai chi directly in front of it, graded it strongly recommended, and still could not evaluate qigong. If the evidence transferred, that would have been the moment to say so.

The defensible conclusion is narrower than either camp wants. Tai chi’s trial record, which the tai chi article in this library sets out in detail, is not available to qigong by inheritance. The reason is definitional before it is empirical: qigong names a range of practices wide enough that a claim beginning “qigong improves” is not well formed until somebody says which qigong.

What the qigong-specific reviews found

Strip out everything borrowed and a modest literature remains.

A search of Cochrane’s library, the most conservative source of systematic reviews in medicine, turns up a single review of qigong as a treatment in its own right: qigong for the primary prevention of cardiovascular disease, published in June 2015. It found 11 completed trials with 1,369 participants and then did something telling: it performed no meta-analysis at all, because the trials were small and at significant risk of bias. Risk of bias could not even be judged for nine of the trials, which were published in Chinese with insufficient methodological detail and whose authors could not be reached. Reviewers reported very low confidence in the results. The detail that should stay with you is this: the one trial they judged to be at low risk of selection and detection bias found no statistically significant effect on cardiovascular risk factors, and was too small to settle anything either way.

A 2011 overview collected every systematic review of qigong in any condition it could find: ten of them, covering a wide range of illnesses. Five concluded qigong was effective and five were inconclusive. The primary studies, and several of the reviews themselves, carried a high risk of bias. The authors judged it unwise to draw firm conclusions. Nothing comparable appears to have replaced it in the indexed literature since.

A 2015 meta-analysis of qigong for chronic pain, from an evidence-based medicine centre at Lanzhou University, is useful mostly for its structure: it analysed internal and external qigong as separate treatments across ten trials rather than merging them, and still ended by asking for trials with more methodological rigour.

The federal health agency’s condition-by-condition summary reads as a catalogue of hedges, and quoting it selectively is how qigong gets oversold. Its own characterisations: research on pain is conflicting; benefit in fibromyalgia rests on a few small studies; research in Parkinson disease is small; a 2021 blood pressure review’s authors indicated a lack of firm evidence; only a small amount of cognitive research has been done; and research on qigong for COVID-19 is extremely limited.

One line in that summary points the other way, and it should never be quoted alone. The agency reports that a 2019 survey of reviews found sufficient evidence supporting qigong for balance training and fall prevention. That sits beside a rheumatology panel in the same era that could not assess qigong at all, and beside the agency’s own entry noting mixed results in older adults generally. Different questions, different grain, opposite directions.

The trial most often cited as proof deserves its sample size in the same sentence. A 2012 pilot randomized 14 people with fibromyalgia, roughly seven per group, to six weeks of home qigong or to a sham version using the same body movements. Reported improvements were large. The design is genuinely clever, because a sham built from identical movements is an attempt to isolate what the movement is not doing, and the authors call it a pilot and ask for a larger sample. Fourteen people settle nothing. The sham-control problem that runs through this whole field is handled at length in the companion article on acupuncture.

Why every reviewer attaches a caveat

Readers who notice that every synthesis of this literature ends with a complaint about quality deserve to know where that comes from.

Part of it is specific and measurable. A 1998 analysis in the journal Controlled Clinical Trials examined trial abstracts indexed in Medline and classified them by country of origin. In the acupuncture arm of that work, every trial from China, Japan, Hong Kong and Taiwan reported a positive result. In a second arm covering interventions other than acupuncture, 75 percent of trials published in England found the test treatment superior, against 99 percent in China, 97 percent in Russia and the Soviet Union, 95 percent in Taiwan and 89 percent in Japan. No trial published in China or Russia reported a treatment to be ineffective.

Three limits have to travel with that finding or it becomes a slur. It is not about qigong: the first study examined acupuncture and the second pooled interventions other than acupuncture, and neither isolated any movement practice. Its evidence base stops in June 1995, which is three decades ago, before modern trial registration and substantial changes in journal standards. And the authors proposed publication bias, a decision about what gets printed, as a possible explanation. They did not allege fabrication and did not conclude the trials were badly run. Their recommendation was that reviewers think carefully about how to handle data from these countries, which is precisely what Cochrane did in 2015 when it declined to pool nine Chinese-language qigong trials.

Safety, including the part that gets left out

The physical safety record is good and straightforwardly reported. The federal health agency states that qigong appears to be a safe form of activity, with many studies reporting no negative side effects, including among older adults and people with chronic illness. In one review of qigong for neck pain, side effects turned up in under 10 percent of adults and were comparable to those in other exercise groups. On pregnancy the agency is plain that there is no research on safety, and advises talking with a health care provider first.

The part that gets left out concerns the mind, and it needs stating carefully because both silence and alarm would be wrong. There are published case reports of brief psychosis following intensive meditative practice, including in people with no psychiatric history. A 2007 paper in Psychopathology describes one such case and reviews the literature, finding reports that split into relapses of existing psychotic illness and short reactions in people with none. Those authors concluded that meditation can act as a stressor in vulnerable people, and noted that the phenomenon carries a name in culture-bound syndrome taxonomies, qi-gong psychotic reaction, while arguing it is not culture-bound.

What does not exist is a denominator. Case reports have no population behind them, so no rate can be calculated and none should be implied. If you have a history of psychosis, mania or severe dissociation, or if you are in an unstable period, that is a conversation to have with the clinician who treats you before starting an intensive meditative practice. Signs worth a same-day call include new confusion about what is real, sleeplessness that will not break, or a sudden surge of agitation after a long session.

Nobody certifies a qigong teacher

Credentials in this field fall into four buckets: defined in regulation, run as a government certification program, issued by a private third party, or simply undefined. Qigong lands in the fourth.

No state licenses qigong instructors and the federal government does not regulate the practice. There is no national certification standard. Various organizations train and certify teachers using criteria each sets for itself, and nothing stops anyone from using the title. The agency’s advice amounts to asking about a teacher’s training and experience, which is the only consumer protection available and should be described as such rather than dressed up.

The field knows. When tai chi and qigong researchers convened at the Harvard Medical School Osher Center for Integrative Health in autumn 2023, the white paper published from that meeting in 2025 listed teaching competencies, credentialing and licensure among the problems still to be solved, alongside style, dose and contextual effects.

How to read a qigong claim in four steps

Any claim you meet about qigong can be sorted quickly.

1. Which qigong? If the source cannot name the form, the style or the protocol, the claim is not specific enough to be true or false.
2. Who was doing it, and to whom? Self-practice and a practitioner directing energy at a patient are different interventions with different evidence and should never share a sentence.
3. Was it tai chi’s evidence? If the study pooled qigong with tai chi or Baduanjin, it cannot speak about either one.
4. What did the reviewers say about quality in the same breath? In this literature the qualifier usually sits in the same sentence as the finding, and harvesting the first half is the commonest distortion.

Nothing here is a reason to begin, stop, replace or postpone any treatment. If you have heart failure, chronic lung disease, Parkinson disease or uncontrolled blood pressure, the question of whether a gentle movement class fits alongside your care is one for the clinician managing it, who can also tell you which movements to modify.

What this leaves you with

Qigong is a low-risk way to move, breathe and pay attention, and for people who will keep doing it that is worth something on its own. What it is not, on the published record, is a treatment with established effects for any specific disease. The obstacle is not a hostile scientific establishment. It is that a category assembled by a government bureau in the 1950s, covering everything from seated breathing to non-contact healing, has never been pinned down well enough for a guideline panel to grade. Until somebody specifies which qigong, nobody can tell you whether qigong works.

Sources

1. Arthritis & Rheumatology (American College of Rheumatology / Arthritis Foundation). “2019 American College of Rheumatology/Arthritis Foundation Guideline for the Management of Osteoarthritis of the Hand, Hip, and Knee (full text read; qi gong could not be assessed for insufficient evidence and lack of standard definitions). 2020. DOI: 10.1002/art.41142.” pubmed.ncbi.nlm.nih.gov
2. National Center for Complementary and Integrative Health (NIH). “Qigong: What You Need To Know (content last updated February 2022; no definition of internal or external qigong in the body text).” nccih.nih.gov
3. University of Hong Kong Scholars Hub (David A. Palmer, pre-publication version of the entry in Religions of the World: A Comprehensive Encyclopedia, ABC-Clio, 2011). “Qigong (also spelled Qi Gong and Chi-kung): the 1950s creation of the category, the Cultural Revolution ban, the mass movement, and the 1999 suppression.” repository.hku.hk
4. Cochrane Database of Systematic Reviews. “Hartley L, Lee MS, Kwong JS, Flowers N, Todkill D, Ernst E, Rees K. Qigong for the primary prevention of cardiovascular disease. 2015. DOI: 10.1002/14651858.CD010390.pub2.” pubmed.ncbi.nlm.nih.gov
5. JRSM Short Reports. “Lee MS, Oh B, Ernst E. Qigong for healthcare: an overview of systematic reviews. 2011. DOI: 10.1258/shorts.2010.010091.” pubmed.ncbi.nlm.nih.gov
6. The Journal of Pain. “Lee MS, Pittler MH, Ernst E. External qigong for pain conditions: a systematic review of randomized clinical trials. 2007. DOI: 10.1016/j.jpain.2007.05.016.” pubmed.ncbi.nlm.nih.gov
7. BMC Complementary Medicine and Therapies (University of California, Davis). “Joy M, Adams N, Yanuck M, Dossett ML. Experiences with Qi and changes in post-acute sequelae of COVID-19 symptoms with qigong (describes external qigong as a non-touch biofield therapy alongside reiki and therapeutic touch). 2025. DOI: 10.1186/s12906-025-05161-w.” pubmed.ncbi.nlm.nih.gov
8. American Journal of Health Promotion. “Jahnke R, Larkey L, Rogers C, Etnier J, Lin F. A comprehensive review of health benefits of qigong and tai chi. 2010. DOI: 10.4278/ajhp.081013-LIT-248.” pubmed.ncbi.nlm.nih.gov
9. The American Journal of Chinese Medicine. “Bai Z, Guan Z, Fan Y, Liu C, Yang K, Ma B, Wu B. The Effects of Qigong for Adults with Chronic Pain: Systematic Review and Meta-Analysis (internal and external qigong analysed separately). 2015. DOI: 10.1142/S0192415X15500871.” pubmed.ncbi.nlm.nih.gov
10. Controlled Clinical Trials. “Vickers A, Goyal N, Harland R, Rees R. Do certain countries produce only positive results? A systematic review of controlled trials (Medline abstracts to June 1995; publication bias named as a possible explanation). 1998. DOI: 10.1016/s0197-2456(97)00150-5.” pubmed.ncbi.nlm.nih.gov
11. Psychopathology. “Kuijpers HJH, van der Heijden FMMA, Tuinier S, Verhoeven WMA. Meditation-induced psychosis. 2007. DOI: 10.1159/000108125.” pubmed.ncbi.nlm.nih.gov
12. Disability and Rehabilitation. “Heredia-Rizo AM, Martinez-Calderon J, Pina-Pozo F, Gonzalez-Garcia P, Garcia-Munoz C. Effectiveness of mind-body exercises in chronic respiratory diseases: an overview of systematic reviews with meta-analyses (qigong, tai chi and yoga reported separately). DOI: 10.1080/09638288.2023.2233899.” pubmed.ncbi.nlm.nih.gov
13. International Journal of Neuroscience. “Liu W, Zahner L, Cornell M, et al. Benefit of Qigong exercise in patients with fibromyalgia: a pilot study (14 participants, sham-qigong comparison). 2012. DOI: 10.3109/00207454.2012.707713.” pubmed.ncbi.nlm.nih.gov
14. Journal of Integrative and Complementary Medicine (Harvard Medical School Osher Center for Integrative Health). “Yeh GY, Ahn A, Clark J, et al. The Science of Tai Chi and Qigong as Whole Person Health, Part II: Evidence Gaps and Opportunities for Future Research and Implementation. 2025. DOI: 10.1089/jicm.2024.0958.” pubmed.ncbi.nlm.nih.gov