Tai Chi for health and well-being: A bibliometric analysis of published clinical studies between 2010 and 2020

Friday, 2 July 2021

The objective of this bibliometric review was to identify the volume, breadth, and characteristics of clinical studies evaluating Tai Chi published between January 2010 and January 2020. Five English and four Chinese language databases were searched. Following independent screening, 1018 eligible publications representing 987 studies were identified, which was a three-fold increase from the previous decade. Most common were randomized controlled trials (548/987, 55.5 %), followed by systematic reviews (157/987, 15.9 %), non-randomized controlled clinical studies (152/987, 15.4 %), case series (127/987, 12.9 %) and case reports (3/987, 0.3 %) that were conducted in China (730/987, 74.0 %), followed by the United States of America (123/987, 12.5 %) and South Korea (20/987, 2.0 %). Study participants were mostly in the adult (55.2 %) and/or older adult (72.0 %) age groups. The top ten diseases/conditions were hypertension, chronic obstructive pulmonary disease, diabetes, knee osteoarthritis, heart failure, depression, osteoporosis/osteopenia, breast cancer, coronary heart disease and insomnia. A quarter of the studies enrolled healthy participants to evaluate the effects of Tai Chi on health promotion/preservation, balance/falls, and physiological/biomechanical outcomes. Yang style Tai Chi was the most popular, followed by Chen and Sun style. Tai Chi was mostly commonly delivered face-to-face by a Tai Chi instructor in group settings for 60 min, three times a week, for 12 weeks. Most studies (93.8 %) reported at least one outcome in favor of Tai Chi. Adverse events were underreported (7.2 %). Over half fell short of expected intervention reporting standards, signalling the need for Tai Chi extensions to existing guidelines.

1. Introduction

Tai Chi (also known as Tai Chi Chuan/Quan or Taiji) is commonly described as a meditative or internal martial art, or a mind-body exercise that originated in China. Tai Chi was developed by the famous martial artist Chen Wang-Ting towards the end of Ming Dynasty (17th Century A.D.).1 Tai Chi comprehensively incorporates Chinese folk and military martial arts, breathing and meditative techniques, Chinese philosophy of yin and yang, and traditional Chinese medicine theory.2 There are different styles of Tai Chi practiced in modern society, which can be broadly classified into traditional styles (e.g., Chen, Yang, Wu/Hao, Wu, and Sun) according to the General Administration of Sport of China.2 Chen style Tai Chi is the original style, and all the other styles are developed from traditional Chen style.1

The multiple potentially therapeutic components of Tai Chi were deconstructed by Wayne3 as follows1: awareness, mindfulness, and focused attention2; intention, belief and expectation3; structural integration, dynamic form and function4; active relaxation5; strengthening and flexibility6; natural, freer breathing7; social support, interaction and community8; embodied spirituality, philosophy and ritual. These components might work independently and/or synergistically with the potential to impact a wide range of health outcomes. As such, Tai Chi can be considered a complex intervention.

An increasing number of clinical studies have documented the safety and potential benefits of Tai Chi. In the past decades, more clinical studies have been conducted to explore the safety and potential benefits of Tai Chi for various health and well-being outcomes in healthy populations or people with different diseases/conditions. Since the peer-reviewed literature on Tai Chi continues to grow, it is necessary to summarize and analyze the overall characteristics and trends of clinical studies on Tai Chi over the past decade. Moreover, our previous bibliometric review involving 507 clinical studies published between 1958 and 2013 found that most studies were performed in China (62.5 %) and published in Chinese (52.3 %).4 However, many reviews of Tai Chi do not include studies published in Asian languages, which is necessary for a more complete evaluation of available evidence. Hence, this study provides an update of our previous bibliometric review by mapping clinical studies published between 2010 and 2020. The objectives of this study were to explore the volume, breadth, and characteristics of clinical evidence on Tai Chi in the past decade, and to identify gaps and priorities for further clinical research.

2. Materials and methods

2.1. Eligibility criteria

Eligible studies were clinical studies including systematic reviews (SR), randomized clinical trials (RCT), non-randomized controlled clinical studies (CCS) (quasi-randomized clinical trial or observational studies such as cohort or case-control studies), case series (CS) and case reports (CR), in which Tai Chi was assessed as an intervention in healthy participants and/or people with any disease/condition. Any type of Tai Chi, regardless of the style, form, or training regimen was included. Studies published between 2010 and 2020 were eligible. Considering publication lags in different databases might be a potential reason for the decline in publication rate identified in our previous bibliometric review for the recent years, we included studies overlapping in our previous bibliometric review.

Anecdotes and study protocols were excluded. Reports published in abstracts, posters, and studies lacking basic information on Tai Chi interventions were not eligible for inclusion. We excluded non-systematic reviews, network analysis and overview of SRs of Tai Chi. Also excluded were studies of Tai Chi in combination with other modalities when the effects of the Tai Chi intervention(s) could not be ascertained. Studies that applied a single movement of Tai Chi, Tai Chi gait, or wheelchair/seated Tai Chi were excluded. Studies that applied Tai Chi pushing hands, Tai Chi sword, Tai Chi knife, Tai Chi soft ball or other forms practiced with instruments were also excluded.

2.2. Information sources and searches

Electronic searches were conducted on the following databases: PubMed, Cochrane Library, EMBASE, Medline, Web of Science, China National Knowledge Infrastructure (CNKI), Chinese Scientific Journal Database (VIP), Sino-Med, and Wanfang Database, from 1st January 2010 to 31st January 2020. The search strategies were developed using the terms from our previous bibliometric analysis and refined through team discussion. The search terms included “Taiji”, “Tai Ji”, “Tai-ji”, “Tai Chi”, “Tai Chi Chuan”, “Tai Chi Quan”, or “Taijiquan”. Limitation to language was not applied. Supplementary file S1 Table outlines two examples of the detailed search syntaxes used to search English and Chinese databases.

2.3. Selection of sources of evidence

The search results from English databases were exported into EndNote (version X9), and those from Chinese databases into NoteExpress (version 3.2) after which, duplicates were removed. To increase consistency among the reviewers, a screening manual based on the previous bibliometric review was used and the reviewers performed a calibration exercise before beginning the study selection process. The titles and abstracts and then full text were independently screened by two of four reviewers (GYY, WLH, MXJ and LNZ) who worked in pairs. Disagreements on study selection were resolved by consensus and discussion.

2.4. Data extraction

A structured data extraction form based on the previous bibliometric analysis was developed and piloted prior to data extraction into REDCap software. Information was extracted about the publication, author, funding, study design, sample, participant characteristics, Tai Chi intervention (e.g., rationale, style, frequency, duration, delivery, instructor background), control/comparison groups, types of outcomes assessed and outcome measures, and a summary of the main findings. The findings were also coded as “possibly positive” if the estimate effects of at least one outcome, regardless of primary or secondary outcomes, statistically favored Tai Chi; “negative” if all the results did not statistically favor Tai Chi; and “unclear” if there was insufficient information to judge whether the estimate effects statistically favored Tai Chi, or the conclusions were inconclusive.

To enhance consistency among reviewers, a data extraction manual was developed, and all reviewers performed a calibration exercise prior to data extraction. Thirteen reviewers (GYY, AS, WLH, LNZ, MXJ, HZ, XWZ, CHZ, YJC, ND, MW, ZA, PZ and FLB) extracted data that was then verified by one reviewer (GYY). Any uncertainty or disagreements were resolved by group consensus and discussion.

2.5. Synthesis of results

The data was exported into Excel. The unit of analysis was the clinical study, not publication. Studies with more than one publication were identified following data extraction and the data combined. The diseases/conditions were classified into categories according to International Statistical Classification of Diseases and Related Health Problems 11th Revision (ICD 11).5 Descriptive statistics were used to summarize the findings. The results are presented in tables and visual representations including figures and diagrams, as well as in a narrative format.

3. Results

3.1. Search results

The database search identified 11,969 publications. After removing duplicates, 6234 titles and abstracts, followed by 1397 full texts were screened. A total of 987 studies reported in 1018 publications were included (Fig. 1).

Fig. 1

Fig. 1. PRISMA Flow diagram.

Supplementary file S1 Text presents the references for the included studies. Among the 1018 publications, 568 (55.8 %) were published in Chinese language, of which 462 were published in 232 different Chinese language journals, 96 were dissertations from 37 different Chinese university repositories, and 10 were from conference proceedings. A further 445 (43.7 %) publications were published in 235 different English language journals, 2 were published in 2 different Portuguese language journals, 2 were published in 2 different Persian language journals and 1 was published in a Korean language journal. Eleven publications were in leading medical journals with a 5-year impact factor greater than 10, including 3 that were published in JAMA Internal Medicine, 2 in the New England Journal of Medicine, and one in The BMJ (S2 Table).

3.2. General characteristics of included studies

Most studies were RCTs (548/987, 55.5 %), followed by 157 (15.9 %) SRs, 152 (15.4 %) CCSs, 127 (12.9 %) CSs, and 3 (0.3 %) CRs. The studies were conducted in 31 different countries and in all six of the World Health Organization (WHO) regions. An overwhelming majority of studies (730/987, 74.0 %) were conducted in China, followed by the United States of America (123/987, 12.5 %) and South Korea (20/987, 2.0 %) (Fig. 2). Just under half (461/987, 46.7 %) of the studies reported funding information, of which 28 (6.1 %) received no funding. Of the funded studies, 70.5 % (325/461) were funded by government, 88 (19.1 %) by academic institutions, 11 (2.4 %) by charities, and 7 (1.5 %) by industry or private companies. The number of studies published each year rose from 67 in 2010 to a peak of 127 in 2016 (Fig. 3). The annual average was 98.7 studies.

Fig. 2

Fig. 2. Number of clinical studies on Tai Chi conducted in different countries. Abbreviation: SR, systematic review; RCT, randomized clinical trial; CCS, non-randomized controlled clinical studies (quasi-randomized clinical trial or observational studies such as cohort or case-control studies); CS, case series; CR, case report.

Fig. 3

Fig. 3. Study designs over time in the numbers of published clinical studies on Tai Chi. Abbreviation: SR, systematic review; RCT, randomized clinical trial; CCS, non-randomized controlled clinical studies (quasi-randomized clinical trial or observational studies such as cohort or case-control studies); CS, case series; CR, case report.

3.3. Participant characteristics

The 830 primary studies involved 69,088 participants of which 11.9 % dropped out or were lost to follow-up. The mean sample size was 86 for RCTs, 97 for CCSs, and 53 for CSs. Most clinical studies enrolled adults aged 60 years and older (598/830, 72.0 %) and/or adults aged 25–59 years (458/830, 55.2 %). Young people age 10–24 years were included in 16.4 % (136/830) of studies and children aged 0–9 years in only 0.5 % (4/830) studies. Gender information was reported in 719 (86.6 %) studies and 57.7 % were female.

Most studies (723/987, 73.3 %) included participants with certain diseases/conditions to investigate the effects of Tai Chi for health promotion/preservation (102/987, 14.1 %), treatment (487/987, 49.3 %) or rehabilitation (132/987, 13.4 %). A quarter (241/987, 24.4 %) enrolled only healthy participants and evaluated the effects Tai Chi on health promotion/preservation (169/987, 17.1 %), balance control/falls prevention/proprioception (66/987, 6.7 %), or exercise physiology/biomechanics (6/987, 0.01 %). There were 23 (2.3 %) studies that involved both healthy participants and those with certain diseases/conditions.

A total of 105 diseases/conditions were investigated in the 987 included studies. Among them, the majority were diseases/conditions of the circulatory system and the musculoskeletal system, based on ICD 115 (Table 1). The top 10 most investigated diseases/conditions were hypertension, chronic obstructive pulmonary disease (COPD), diabetes, knee osteoarthritis, heart failure, depression, osteoporosis or osteopenia, breast cancer, coronary heart disease, and insomnia (Table 2).

Table 1. Clinical trials on Tai Chi (n = 987) organised by prevalence of disease categories classified based on ICD-11 classifications.

Disease/conditions (ICD-11 codes)No. of study (%)
11 Diseases of the circulatory system 121 (12.3 %)a
15 Diseases of the musculoskeletal system 78 (7.9 %)a
08 Diseases of the nervous system 76 (7.7 %)a
06 Mental, behavioural or neurodevelopmental disorders 74 (7.5 %)a
05 Endocrine, nutritional and metabolic diseases 58 (5.9 %)
12 Diseases of the respiratory system 49 (5.0 %)a
02 Neoplasms 28 (2.8 %)a
21 Symptoms, signs or clinical findings, not elsewhere classified 21 (2.1 %)a
16 Diseases of the genitourinary system 21 (2.1 %)
24 Factors influencing health status or contact with health services 13 (1.3 %)
13 Diseases of the digestive system 11 (1.1 %)
22 Injury, poisoning and certain other consequences of external causes 6 (0.6 %)
26 Supplementary Chapter Traditional Medicine Conditions 4 (0.4 %)
10 Diseases of the ear or mastoid process 2 (0.2 %)
14 Diseases of the skin 1 (0.1 %)
09 Disease of the visual system 1 (0.1 %)
17 Conditions related to sexual health 1 (0.1 %)

Abbreviations: ICD-11, International Classification of Diseases, Eleventh Revision.


Some systematic reviews involved more than one type of diseases or conditions.

Table 2. Top 10 diseases/conditions included in clinical studies on Tai Chi (n = 987).

Study design (number of studies)
Disease/conditionSRaRCTCCSCSCRTotal (%)
Hypertension 10 35 12 12 0 69 (7.0)
COPD 9 38 5 1 0 53 (5.4)
Diabetes 12 29 9 3 0 53 (5.4)
Knee osteoarthritis 3 20 3 1 0 27 (2.7)
Heart Failure 6 16 2 2 0 26 (2.6)
Depression 2 18 4 1 0 25 (2.5)
Osteoporosis or osteopenia 5 15 5 0 0 25 (2.5)
Breast cancer 5 19 1 0 0 25 (2.5)
Coronary heart disease 2 17 3 1 0 23 (2.3)
Insomnia 0 15 2 4 0 21 (2.1)

Abbreviations: SR, systematic review; RCT, randomized clinical trial; CCS, non-randomized controlled clinical studies (quasi-randomized clinical trial or observational studies such as cohort or case-control studies); CS, case series; CR, case report; USA, United States of America; UK, United Kingdom; COPD, chronic obstructive pulmonary disease.


Some systematic reviews involved more than one type of diseases or conditions.

3.4. Characteristics of Tai Chi interventions

In the 830 primary studies, the reporting of the Tai Chi interventions varied widely. Almost all the studies reported the duration/follow-up time (815/830, 98.2 %) and most reported the Tai Chi style and/or form (668/830, 80.4 %), frequency (708/830, 85.3 %), session duration (764/830, 92.0 %), learning method (592/830, 71.3 %) and practicing method (569/830, 68.6 %). Fewer studies reported the details about the background of instructor(s) (449/830, 54.1 %), composition of each session (410/830, 49.4 %), learning/practicing process (304/830, 36.6 %), and extent to which the intervention varied (146/830, 17.6 %) (S1 Fig.).

The reported duration of the entire intervention ranged from one to 192 weeks. The most common duration was 12 weeks (272/815, 33.4 %), followed by 24 weeks (145/815, 17.8 %) and 8 weeks (92/815, 11.2 %). In 96 (11.8 %) studies the duration was over 24 weeks, while nine (1.1 %) studies investigated the immediate effects of Tai Chi from one session. A total of 96 (11.6 %) studies reported follow-up information, and the follow-up varied from one week to six years following the intervention.

The duration and frequency of each session ranged from 10−240 min, and 1 to over 14 times per week, respectively. The most common were 60 min sessions (411/764, 53.8 %), three times per week (224/708, 31.6 %). Seven studies reported changing the session duration, for example, starting with 15 min in the first 4 weeks and increasing to 20−40 min in the following 8 weeks. In one CS study, an intensive training with 240-min session per day for seven days was applied to investigate the effects of Tai Chi among long-term practitioners.6

Among the popular traditional Tai Chi styles (i.e., Chen, Yang, Wu, Wu/Hao, Sun), Yang style was applied most (487/830, 58.7 %), followed by Chen style (57/830, 6.9 %) and Sun style (19/830, 2.3 %). A variety of known, traditional, modified/simplified, or unspecified forms under each style were also reported (Table 3). Under Yang style, the Simplified 24-form Yang style Tai Chi was most popular (347/487, 71.3 %), followed by the modified/simplified form varying from 5 to 21 movements (98/487, 20.1 %). Other known or traditional forms of Yang style ranging from 32 to 108 movements were applied in fewer (25/487, 5.1 %) studies. Under Chen and Sun styles, the modified/simplified form accounts for 24.6 % (14/57) and 89.5 % (17/19), respectively. Some studies applied more than one style or form of Tai Chi. Of the 146 (17.6 %) studies that reported the extent to which the Tai Chi intervention varied, only 2.7 % were fully standardized with no variations allowed. Instead, most allowed some variation (75.3 %) or were personalized (21.9 %), that is tailored to the individual participants in the study.

Table 3. Tai Chi styles and forms applied in 830 clinical studies of RCTs, CCSs, CSs and CRs.

Tai Chi styleNo. of studyFrequencya
Yang style 487 58.7 %
 24-form Yang style 347 41.8 %
 32-form Yang style 2 0.2 %
 36-form Yang style 1 0.1 %
 37-form Yang style 3 0.4 %
 42-form Yang style 12 1.4 %
 64-form Yang style 3 0.4 %
 85-form Yang style 2 0.2 %
 108-form Yang style 2 0.2 %
 Modified form of Yang style 98 11.8 %
 Unspecified forms of Yang style 29 3.5 %
Chen style 57 6.9 %
 24-form Chen style 8 1.0 %
 37-form Chen style 1 0.1 %
 42-form Chen style 4 0.5 %
 48-form Chen style 2 0.2 %
 74-form Chen style 1 0.1 %
 Modified Chen style 14 1.7 %
 Unspecified forms of Chen style 27 3.3 %
Sun style 19 2.3 %
 Modified Sun style 17 2.0 %
 Unspecified forms of Sun style 2 0.2 %
Wu style 11 1.3 %
 Modified Wu style 6 0.7 %
 Unspecified forms of Wu style 5 0.6 %
Wu/Hao style 2 0.2 %
 Modified Wu/Hao style 1 0.1 %
 Unspecified forms of Wu/Hao style 1 0.1 %
Others 267 32.2%
 Unspecified style and form 162 19.5 %
 Unspecified style in modified form 105 12.7 %

Abbreviation: RCT, randomised clinical trial; CCS, non-randomized controlled clinical studies (quasi-randomized clinical trial or observational studies such as cohort or case-control studies); CS, case series; CR, case report.


Some studies applied more than one type or form of Tai Chi.

Most studies delivered Tai Chi through face-to-face lessons from instructors (565/592, 95.4 %) and in group classes only (394/569, 69.2 %) or a combination of group classes and home practice (108/569, 19.0 %). In 14 studies, participants practiced Tai Chi in a group, however, it is unclear whether the sessions took place were supervised by an instructor. Six studied (6/569, 1.1 %) evaluated internet-delivered lessons. In 40 studies (40/569, 7.0 %) Tai Chi was only practiced at home with support materials (e.g., DVDs, printed documents and books).

The composition of each session (e.g. ‘10 min of warm-up and a review of Tai Chi principles; 30 min of Tai Chi practice; and 10 min of cool down’ (7)) was reported in approximately half (410/830, 49.4 %) of the studies, and the learning/practicing process (e.g. ‘the first 10 weeks emphasized the mastery of single forms through multiple repetitions; later weeks focused on repetitions to enhance balance and increase locomotion’ (8)) was reported in 36.6 % studies.

Of the 449 (54.1 %) studies that reported information about instructor(s), 136 (30.2 %) briefly reported information pertaining to the background of the instructor(s) such as their Tai Chi training, qualifications, experience, including any additional training, experience or knowledge specific to the target population. Some Tai Chi instructor(s) had additional training or qualifications (e.g., nursing, physiotherapy, rehabilitation therapy, exercise therapy, Chinese medicine, physical education, and mental health care). In 17.1 % (77/449) of studies, multiple instructors were reported as delivering the intervention.

3.5. Characteristics of controls and comparison groups

In more than half of the 830 primary studies, the Tai Chi intervention was used alone or used alone in at least one Tai Chi group (471/830, 56.7 %). The control groups in the 700 RCTs and CCSs included blank control (29.0 %), conventional medication (27.3 %), aerobic exercise and other exercises (20.4 %), health education (13.7 %), usual care or usual rehabilitation program (11.1 %), diet and lifestyle guidance (8.4 %), waitlist (4.4 %), acupuncture and related therapies (3.4 %), Qigong (2.7 %), Chinese herbal medicine (2.6 %), psychological intervention (2.6 %), recreational or non-exercise activities (2.0 %), yoga (1.4 %), breathing technique (1.4 %), balance training (1.4 %), massage/tuina (1.3 %), physiotherapy (0.7 %), and others therapies.

Modalities used in combination with Tai Chi include conventional medication (23.0 %), health education (9.0 %), diet and lifestyle guidance (8.3 %), usual care or usual rehabilitation program (7.2 %), acupuncture and related therapies (4.1 %), Chinese herbal medicine (4.0 %), aerobic exercise and other physical exercise (3.6 %), psychological intervention (3.5 %), massage/tuina (1.4 %), physiotherapy (1.4 %), Qigong (1.2 %), breathing technique (1.2 %), music (0.8 %), and other therapies.

Eighty-six (12.3 %) of the 700 RCTs and CCSs evaluated more than one Tai Chi group and compared different styles, forms, frequencies, durations, delivery modes (e.g., group-based vs individual-based Tai Chi) or settings (e.g., in water vs on land).

3.6. Outcomes and main findings

In the 987 included clinical studies, physical performance outcomes (e.g., strength, balance, cardiopulmonary function, biomarker, etc.) were most commonly evaluated (752/987, 76.2 %), followed by psychological outcomes (e.g., stress, depression, mood, etc.) (233/987, 23.6 %), symptoms (e.g., pain, fatigue, etc.) (230/987, 23.3 %), and quality of life (QoL) (using generic or disease-specific measures) (213/987, 21.6 %). Less common were health-related events (e.g., falls, fracture, angina, stroke, hospitalization, death, etc.) (8.3 %), safety/adverse events (7.2 %), compliance and adherence (2.5 %), cognitive function (2.3 %), satisfaction with Tai Chi (1.4 %), and sleep quality (1.0 %).

Most primary studies (788/830, 94.9 %) reported at least one positive result in favor of Tai Chi, while 3.0 % of studies reported no beneficial effects, and in 2.0 % of studies the evidence was insufficient to make a conclusion. Similarly, among the 157 SRs, 138 (87.9 %) reported at least one positive result, while 1.9 % reported no beneficial effects and 10.2 % of studies did not make a conclusion due to insufficient evidence. Of the 71 studies that reported on the safety profile of Tai Chi, 40 reported no intervention related adverse events and 31 studies reported minor, mostly musculoskeletal related complaints (e.g., knee pain, transient low back) that were attributed to Tai Chi. No serious adverse events were reported.

Dropout/withdrawal information was reported in less than half (354/830, 42.7 %) of the primary studies, among which 41 (0.3 %) reported no dropouts or withdrawal. Reported reasons for dropout/withdrawal in the Tai Chi groups included time conflict, family/work commitment, illness, deterioration of health, hospital admission, injury accidents, transportation/travel issues, loss of interest in Tai Chi, relocation, death, and refusal of follow-up assessment.

4. Discussion

This comprehensive review updates our previous bibliometric analysis on the volume, breadth, and characteristics of clinical studies on Tai Chi for various health and well-being outcomes in healthy population or people with different diseases/conditions.4 The number of published clinical studies on Tai Chi continues to grow with a three-fold increase in publications compared to the previous decade.4 The Tai Chi studies were mainly funded by governments and several studies were published in leading medical journals, indicating a broad high-level interest in Tai Chi research.

The number of countries in which Tai Chi has been investigated also increased. Compared to our previous findings, an additional ten countries have now undertaken clinical research on Tai Chi, making the total 31 countries. Despite the strong cultural roots of Tai Chi, this finding confirms the feasibility and transferability of Tai Chi to non-Chinese populations. It is noteworthy that over half of the studies were conducted in China and published in Chinese language. This finding is consistent with our previous bibliometric review.4 As such, language restrictions must still be cautiously applied when searching the literature for clinical studies on Tai Chi interventions.9

An increasing number of diseases, conditions and outcomes were also investigated that highlights a growing interest in evaluating the multidimensional and multisystem effects of Tai Chi interventions. Consistent with our previous findings, the most commonly investigated diseases/conditions aligned those that most commonly impacting middle age and older adults. Indeed, most study participants were in the 60 years and older age groups. However, in China, Tai Chi is part of the national educational curriculum.10 As such, there are further opportunities for studies to investigate the benefits of Tai Chi on mental and physical health in young people and children, along with its role in health promotion and disease prevention.

A variety of traditional forms of Tai Chi or modified/simplified forms of each style were reported. Yang style remained the most common style, although there is a growing body of clinical studies evaluating Chen style and Sun style. Its popularity might reflect the national level promotion of Yang style Tai Chi since 1950s.2 To make Tai Chi easier to learn, practice and remember, the General Administration of Sport of China issued a book titled ‘Simplified Tai Chi’ in 1956, to introduce the Simplified 24-form Yang style Tai Chi which they modified from the traditional longer Yang style Tai Chi by removing difficult and duplicate movements.2 Importantly, the larger number of studies investigating Yang style does not imply that it confers greater health benefits. Nevertheless, although the principles and essential components of different styles and forms of Tai Chi are similar, there may well be differences that warrant further investigation.

The wide range of variability in the Tai Chi training methods, frequencies, and durations also warrants further investigation. Of note, 10.4 % of included primary studies had more than one Tai Chi group with the aim of evaluating the potential benefits, advantages and/or acceptability of the different styles, forms, frequencies, duration, delivery modes and settings for the target population. Consistent with our previous findings, we found that Tai Chi interventions were mostly commonly delivered as 60 min per session, three or two times per week, for 12 or 24 weeks. It is important to note, however, that the ‘intensity’ of Tai Chi practice can vary whilst maintaining the same duration and frequency, yet only a small portion of clinical studies reported the extent of the variation in intervention among participants. Therefore, more studies that evaluate the dose-response effects of Tai Chi for people with different health status are indicated to inform evidence-based decisions about service delivery.

Despite the increasing number of publications, the transparency and clarity of reporting the intervention details remains a concern.4 Many studies fell short of expected reporting standards such as the Template for Intervention Description and Replication (TIDieR) checklist.11 Most notably, less than half of the studies reported sufficient information about the instructor(s) delivering the intervention, the procedures and composition of each session, and any tailoring or modifications that were applied. The other key concern was that the reporting rate of adverse events was very low at 7.2 %. The reporting rate was substantially lower than that of our previous bibliometric review (20.7 %),4 as well as another review of English-language RCTs of Tai Chi (33 %).12 Systematic reviews that have evaluated the reporting of RCTs evaluating Tai Chi interventions have also identified poor reporting quality.13, 14, 15 In all, these findings suggest there is a need to develop reporting guidelines that are specific for clinical trials of Tai Chi interventions.

4.1. Limitation of the study

This review has limitations which should be considered prior to interpreting the findings. Firstly, the aim of this study was to map the general trends and characteristics of clinical studies on Tai Chi as a whole; however, whilst no language restrictions were applied, only English and Chinese language databases were searched, so some studies may have been missed. Secondly, the methodological quality of included studies was not evaluated, therefore the reported ‘possibly positive’ or ‘possibly negative’ effects of Tai Chi should be interpreted with caution. Also, we did not conduct an extensive data analyses regarding the effects of Tai Chi for specific diseases/conditions, therefore no recommendations on the effects and safety of Tai Chi for specific populations can be made. Finally, this review did not include cross-sectional studies, which account for a large portion of the evidence body about the prevalence of Tai Chi use, and the search strategy was not designed to identify longitudinal cohort studies that may have recorded Tai Chi use among other things. Therefore, it is likely that studies reporting associations between Tai Chi practiced over many years and long-term outcomes have been missed.

5. Conclusions

Over the past decade there has been a continuing growth in the publication of clinical studies on Tai Chi. These studies were conducted in more countries and investigated Tai Chi for more populations, diseases, conditions, and outcomes. Wide variations in the information reported remains a concern, as it reduces confidence in the existing evidence. Specific guidance aimed at improving the reporting of Tai Chi clinical studies is warranted if we are to gain a deeper understanding of the short and long-term benefits and safety of Tai Chi, and how best to optimize its implementation not only for preventing and treating disease, but also for promoting health and wellbeing.

Author contributions

GYY: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Writing – Original Draft, Writing – Review & Editing, Visualization, Project administration. AS: Investigation, Formal analysis, Writing – Review & Editing, Visualization. WLH, LNZ, MXJ: Investigation, Validation, Resources. ND, ZA, YJC, HZ, CHZ, XWZ, FLB, MW and XZ: Investigation. JPL: Conceptualization, Methodology, Writing – Review & Editing. PMW, CE and DC: Methodology, Writing – Review & Editing. HK, JH and AB: Writing – Review & Editing. All authors proofread the manuscript.


Declaration of Competing Interest

The authors report no declarations of interest. As a medical research institute, NICM Health Research Institute receives research grants and donations from foundations, universities, government agencies, individuals, and industry. Sponsors and donors also provide untied funding for work to advance the vision and mission of the Institute. The authors declare no competing financial interests. PMW is the founder and sole owner of the Tree of Life Tai Chi Center. PMW’s interests were reviewed and managed by the Brigham and Women’s Hospital and Partner’s Health- Care in accordance with their conflict-of-interest policies.


We thank Seungyeon Yeon, a PhD candidate from NICM Health Research Institute, Western Sydney University, for assisting the study selection and data extraction of potential studies published in Korean, and Junko Yoshikawa, a PhD candidate from Nanjing University of Traditional Chinese Medicine for assisting the study selection of potential studies published in Japanese.

PMW was supported by National Institute of Health [grant number K24 AT009282].


Appendix A. Supplementary data


The following is Supplementary data to this article:

Download : Download Word document (23KB)Download : Download Word document (149KB)Download : Download Word document (30KB)Download : Download Word document (13KB)





These authors contributed equally to this work.


G.Y. Yang, L.Q. Wang, J. Ren, et al.
Evidence base of clinical studies on Tai Chi: a bibliometric analysis
PLoS One, 10 (3) (2015), Article e0120655
World Health Organization
ICD-11 International classification of diseases for mortality and morbidity statistics (ICD-11 MMS)
Available from:
R.W. Geib, H. Li, E. Oggero, et al.
Using computerized posturography to explore the connection between BMI and postural stability in long-term tai chi practitioners
Biomed Sci Instrum, 47 (2011), pp. 288-293
C.C. Wang, C.H. Schmid, R.A. Fielding, et al.
Effect of tai chi versus aerobic exercise for fibromyalgia: comparative effectiveness randomized controlled trial
BMJ-Br Med J (2018), p. 360
F. Li, P. Harmer, K. Fitzgerald, et al.
Tai chi and postural stability in patients with Parkinson’s disease
NEJM - New Engl J Med, 366 (6) (2012), pp. 511-519
J. Hunter, S. Arentz, J. Goldenberg, et al.
Choose your shortcuts wisely: COVID-19 rapid reviews of traditional, complementary and integrative medicine
Integr Med Res, 9 (3) (2020), p. 100484
J.Y. Yang
A practical exploration on the reform of Taijiquan elective courses in universities based on the Chinese health physical education curriculum model
J Phys Educ (06) (2020), pp. 111-117
T.C. Hoffmann, P.P. Glasziou, I. Boutron, et al.
Better reporting of interventions: template for intervention description and replication (TIDieR) checklist and guide
BMJ (Clin Res Ed), 348 (2014), p. g1687
P.M. Wayne, D.L. Berkowitz, D.E. Litrownik, J.E. Buring, G.Y. Yeh
What do we really know about the safety of tai chi? A systematic review of adverse event reports in randomized trials
Arch Phys Med Rehabil, 95 (12) (2014), pp. 2470-2483