Qigong and tai chi: what systematic reviews say

Author: Raymond Hackland · Last reviewed: 18 October 2026 · Reading time: about 8 minutes

In short

• What this is: A guide to the big picture of research on tai chi and Qigong, drawn from systematic reviews (which pool individual trials) and overviews of reviews (which pool the reviews themselves), including the reviews on Liu Zi Jue.
• What the evidence says: There is a lot of research, and much of it is weak. The strongest finding, rated high certainty in a Cochrane review, is that tai chi reduces the number of older people who fall. A 2026 evidence map also reports high-certainty reviews for blood pressure in hypertension and bone density after menopause, but we could not read those reviews ourselves. Many other results are promising but rated only low to moderate certainty; the Liu Zi Jue reviews in chronic lung disease did not formally rate certainty at all, and their authors point to methodological bias. Serious side effects are rarely reported, but harms are often not reported at all.
• What it means for your practice: Tai chi and Qigong look like reasonable, gentle forms of movement for many people. They are not a substitute for medical care, and results from one style or one group of patients don’t automatically apply to another, including to the Six Healing Sounds.

What science says

Why reviews, and why reviews of reviews. Label: Well studied (as a research method). A single trial can be small, flawed or simply lucky. A systematic review searches for all trials on a question, assesses their quality and, if suitable, combines them in a meta-analysis. Because there are now hundreds of reviews on tai chi and Qigong, researchers have also written overviews of reviews, also called umbrella reviews, and evidence maps. Several of them grade their findings for certainty, from high to very low, depending on risk of bias, consistency and precision [1, 2]. For how we turn this into our own labels, see our evidence labels.

The big picture for tai chi. Label: Early evidence (overall) · Well studied (for a few specific outcomes). A 2022 overview found 210 systematic reviews of tai chi published between 2010 and 2020 [1]. From the best and most recent reviews it extracted 114 effect estimates covering 59,306 adults. Compared with control groups, 66 of these estimates reported clinically important benefits. Only 8 of the 114 (7.0%) were rated high certainty, 43 (37.7%) moderate, 36 (31.6%) low and 27 (23.7%) very low, mostly because of risk of bias in the underlying trials [1]. Benefits were most consistently reported for Parkinson’s disease, falls risk, knee osteoarthritis, low back pain, stroke and heart and circulatory conditions, including high blood pressure. Except for stroke, the overview found no evidence on preventing disease [1]. An earlier evidence map of 107 reviews saw potentially positive effects in a similar range of areas, but the reviews themselves cautioned that firm conclusions could not be drawn [3].

Tai chi and Qigong together. Label: Early evidence (including blood pressure and bone density). A 2026 evidence map looked only at reviews from 2014 to 2024 that formally rated the certainty of their findings for tai chi or Qigong: 26 reviews in all [2]. Two reported high-certainty evidence, for lower blood pressure in people with hypertension and for bone mineral density after menopause. We could read only the map’s summary, not those two reviews, and the 2022 overview, which we read in full, rated tai chi’s effect on blood pressure in essential hypertension as moderate to low certainty [1]. That is why we label these results Early evidence. Sixteen of the 26 reviews reported at least one moderate-certainty finding, but much of the literature remained low or very low certainty [2]. A 2022 overview of 58 reviews noted that most of the studies were carried out in China [4].

Falls in older people. Label: Well studied. A Cochrane review of exercise for preventing falls in older people living in the community found that tai chi reduces the number of people who fall by about 20% (high-certainty evidence, 8 studies, 2,677 participants) and may reduce the overall rate of falls by about 19% (low certainty) [5]. The often-quoted 23% fewer falls at high certainty refers to exercise programmes of all types, not to tai chi specifically [5]. Programmes built on balance and functional exercises had the strongest evidence overall [5]. Falls risk is also among the outcomes where the 2022 overview found benefits most consistently [1].

Chronic obstructive pulmonary disease (COPD). Label: Early evidence. A Cochrane review of 12 studies with 984 participants found that, compared with usual care, people doing tai chi walked about 30 metres further in six minutes and had better lung function after the programme. Breathlessness after the programme was similar in both groups, and no clear change in quality of life was seen; adding tai chi to other breathing or exercise programmes showed no extra benefit. The walking-distance result itself was rated very low certainty, and the quality of evidence overall ranged from very low to moderate. No adverse events were reported [6].

Liu Zi Jue. Label: Early evidence. Clinical research on Liu Zi Jue, the Six Healing Sounds, is mostly about people with stable COPD. A 2020 review of 14 randomised trials with 920 patients, in which the control groups received usual care, reported improvements in breathlessness, walking distance, lung function and quality of life. The authors added that methodological bias and possible placebo effects mean further research is needed to confirm these findings [7]. A 2021 review of 16 trials with 1,039 patients [8] and a 2022 review of 40 studies with 3,137 participants [9] reported similar improvements; the 2022 review also reported lower anxiety and depression scores, and concluded that Liu Zi Jue “may be an effective adjuvant therapy”, that is, an addition to standard care [9]. These reviews cover overlapping search periods, so they probably share many trials (we did not check the overlap); three reviews are not three independent confirmations. As our article on the history of Liu Zi Jue notes, the authors themselves point to methodological weaknesses.

Safety. Label: Early evidence. A review of 153 randomised trials of tai chi found that reported side effects were usually minor and mostly musculoskeletal, with no serious adverse events attributed to the practice, but only about a third of the trials reported on side effects at all [10]. In the 2026 evidence map, 18 of the 26 reviews reported data on adverse events and none reported serious ones, though harms were reported inconsistently [2]. Our article on who should be careful with breathing practices explains when extra care makes sense.

Why the quality is so often low

Label for this section: Well studied (these are standard, well-documented problems in research on mind-body practices).

The reviews above name the same problems again and again [1, 2, 6, 7]:

• Risk of bias. In the 2022 overview, concerns about bias affected about 81% of the effect estimates [1]. Typical issues are unclear randomisation, missing allocation concealment and outcome assessors who knew who practised [1, 6].
• No real placebo. You can’t practise tai chi without knowing it. Trials usually compare it with usual care or add it to another programme, so expectation and attention may explain part of the effect; the authors of the 2020 Liu Zi Jue review name possible placebo effects explicitly [6, 7].
• Small and short trials. In the Cochrane COPD review, study size ranged from 10 to 206 participants [6].
• Different styles and doses. Trials use different forms, numbers of movements and programme lengths, so the “best” protocol can’t be identified [6].
• Weak reviews. The 2022 overview found many reviews limited by narrow searches, language bias, poor handling of differences between trials, poor reporting and no registered protocol [1].
• Patchy reporting of harms. Side effects are often not recorded systematically [2, 10].

What is reasonably supported, and what isn’t

Label for this section: Early evidence (overall) · Well studied (where marked high certainty).

• Reasonably supported (high certainty, confirmed in the review): fewer older people falling with tai chi [5].
• Promising, but mostly moderate certainty or lower, or not yet checked by us: lower blood pressure in hypertension and better bone density after menopause with tai chi or Qigong [1, 2]; knee osteoarthritis, low back pain, Parkinson’s disease, depression, cognitive performance, COPD [1, 3, 6]; Liu Zi Jue for COPD, where the reviews did not formally rate certainty [7, 8, 9].
• Not supported by these reviews: preventing disease in general (with stroke as the only exception the 2022 overview found) [1]; replacing medical treatment; and any claim that a particular sound or form works on a particular organ.

Where popular claims go beyond the evidence

Label for this section: Speculative.

• “Science has settled it: Qigong works.” The reviews show a large but mostly low-to-moderate quality literature, with a few high-certainty findings for specific outcomes [1, 2]. “Settled” overstates it.
• “If it helps COPD, it helps everything.” Results apply to the people and outcomes studied. Even within COPD, adding tai chi to other exercise showed no extra benefit [6].
• “More studies means stronger evidence.” Many reviews of the same trials don’t add certainty: the 2022 overview found 210 reviews of tai chi, yet rated only 7% of the effect estimates it extracted as high certainty [1].
• “No side effects.” Serious adverse events are rarely reported, but many trials never looked properly [2, 10].

How this relates to the Six Healing Sounds, Qigong and neigong

Label for this section: Early evidence (research on Liu Zi Jue) · Tradition only (qi, neigong concepts).

Liu Zi Jue is one of the few breathing-centred Qigong forms with its own systematic reviews, and they are encouraging but limited: mostly COPD, mostly usual-care controls, and mostly from China [7, 8, 9]. In two of the reviews, 12 of 14 and 15 of 16 included trials were published in Chinese [7, 8]. The trials studied the traditional form, usually alongside standard medical care, in people with a specific diagnosis. Their results can’t simply be transferred to healthy people, to other conditions, or to a music-assisted practice. Results for tai chi, a movement-based practice, don’t transfer automatically to the Six Healing Sounds either.

Neigong, the “internal work” of cultivating breath, attention and qi, has hardly been studied as such; trials measure outcomes such as walking distance or blood pressure, not qi. Qi is a traditional concept, and research neither confirms nor refutes it.

Resonance Heritage Qi is a personal, music-assisted practice inspired by Liu Zi Jue: connected to it, not a replacement for the traditional form or for study with a teacher of that tradition. None of the reviews above studied it, and we make no claims that it produces any of the effects these reviews report.

Sources

1. Yang GY, Hunter J, Bu FL, Hao WL, Zhang H, Wayne PM, Liu JP, 2022. Determining the safety and effectiveness of Tai Chi: a critical overview of 210 systematic reviews of controlled clinical trials. Systematic Reviews 11:260. https://doi.org/10.1186/s13643-022-02100-5
2. Mak SS, Allen JE, Begashaw MM, Beroes-Severin JM, Miake-Lye IM, Shekelle PG, 2026. An evidence map of Tai Chi and qigong, 2014–2024: a systematic review. Journal of Integrative and Complementary Medicine, published online 16 August 2026 ahead of print. https://doi.org/10.1177/27683605261476938
3. Solloway MR, Taylor SL, Shekelle PG, Miake-Lye IM, Beroes JM, Shanman RM, Hempel S, 2016. An evidence map of the effect of Tai Chi on health outcomes. Systematic Reviews 5:126. https://doi.org/10.1186/s13643-016-0300-y
4. Lee SH, Jeon Y, Huang CW, Cheon C, Ko SG, 2022. Qigong and Tai chi on human health: an overview of systematic reviews. American Journal of Chinese Medicine 50(8):1995–2010. https://doi.org/10.1142/S0192415X22500859
5. Sherrington C, Fairhall NJ, Wallbank GK, Tiedemann A, Michaleff ZA, Howard K, Clemson L, Hopewell S, Lamb SE, 2019. Exercise for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews 1:CD012424. https://doi.org/10.1002/14651858.CD012424.pub2
6. Ngai SP, Jones AY, Tam WW, 2016. Tai Chi for chronic obstructive pulmonary disease (COPD). Cochrane Database of Systematic Reviews 6:CD009953. https://doi.org/10.1002/14651858.CD009953.pub2
7. Xiao L, Duan H, Li P, Wu W, Shan C, Liu X, 2020. A systematic review and meta-analysis of Liuzijue in stable patients with chronic obstructive pulmonary disease. BMC Complementary Medicine and Therapies 20:308. https://doi.org/10.1186/s12906-020-03104-1
8. Gao P, Tang F, Liu W, He K, Mo Y, 2021. Effect of liuzijue qigong on patients with stable chronic obstructive pulmonary disease: a systematic review and meta-analysis. Medicine (Baltimore) 100(41):e27344. https://doi.org/10.1097/MD.0000000000027344
9. Xu S, Zhang D, He Q, Ma C, Ye S, Ge L, Zhang L, Liu W, Chen Z, Zhou L, 2022. Efficacy of Liuzijue Qigong in patients with chronic obstructive pulmonary disease: a systematic review and meta-analysis. Complementary Therapies in Medicine 65:102809. https://doi.org/10.1016/j.ctim.2022.102809
10. Wayne PM, Berkowitz DL, Litrownik DE, Buring JE, Yeh GY, 2014. What do we really know about the safety of tai chi?: A systematic review of adverse event reports in randomized trials. Archives of Physical Medicine and Rehabilitation 95(12):2470–2483. https://doi.org/10.1016/j.apmr.2014.05.005

Source quality note: [1]–[4] are overviews of systematic reviews or evidence maps; [1] and [2] rated the certainty of the evidence with formal methods, [4] summarised reviews without such ratings. [5] and [6] are Cochrane reviews, which follow particularly strict methods. [7]–[9] are systematic reviews of Liu Zi Jue trials, almost all from China, and probably overlapping; [9] rated trial quality with a simpler scale (the modified Jadad scale). [10] is a systematic review of how side effects were reported. [2] was published online ahead of print in August 2026. Figures were checked against the full texts of [1], [3] and [5]–[8] and [10]; for [2], [4] and [9] only the published abstracts were available. Numbers are as the authors report them.