J Acupunct Meridian Stud 2009;2(3):171181

RE V I E W A RTI CL E

The Effect of Tai Chi on Psychosocial Well-being: A Systematic Review of Randomized Controlled Trials
Wei Chun Wang1, Anthony Lin Zhang2, Bodil Rasmussen3, Li-Wei Lin3, Trisha Dunning3, Seung Wan Kang4, Byung-Joo Park5, Sing Kai Lo1*
1 2

Faculty of Health, Medicine, Nursing and Behavioural Sciences, Deakin University, Australia Discipline of Chinese Medicine, RMIT University, Australia 3 School of Nursing, Deakin University, Australia 4 Seoul National University Medical Research Center, Korea 5 Seoul National University College of Medicine, Korea

Received: Mar 12, 2009 Revised: Jul 01, 2009 Accepted: Jun 18, 2009 KEY WORDS:
critical appraisal; mental health; oriental medicine; therapy

Abstract Objective: This systematic review aimed to critically appraise published clinical trials designed to assess the effect of Tai Chi on psychosocial well-being. Data Sources: Databases searched included MEDLINE, CINAHL, EMBASE, HEALT, PsycINFO, CISCOM, the Cochrane Central Register of Controlled Trials of the Cochrane Library, and dissertations and conference proceedings from inception to August 2008. Review Methods: Methodological quality was assessed using a modified Jadad scale. A total of 15 studies met the inclusion criteria (i.e. English publications of randomized controlled trials with Tai Chi as an intervention and psychological well-being as an outcome measure), of which eight were high quality trials. The psychosocial outcomes measured included anxiety (eight studies), depression (eight studies), mood (four studies), stress (two studies), general mental health three studies), anger, positive and negative effect, self-esteem, life satisfaction, social interaction and self-rated health (one study each). Results: Tai Chi intervention was found to have a significant effect in 13 studies, especially in the management of depression and anxiety. Although the results seemed to suggest Tai Chi is effective, they should be interpreted cautiously as the quality of the trials varied substantially. Furthermore, significant findings were shown in only six high quality studies. Moreover, significant between group differences after Tai Chi intervention was demonstrated in only one high quality study (the other three significant results were observed in non-high quality studies). Two high quality studies in fact found no significant Tai Chi effects. Conclusion: It is still premature to make any conclusive remarks on the effect of Tai Chi on psychosocial well-being.

*Corresponding author. Faculty of Health, Medicine, Nursing and Behavioural Sciences, Deakin University, 221 Burwood Highway, Burwood, VIC 3125, Australia. E-mail: singkai.lo@deakin.edu.au 2009 Korean Pharmacopuncture Institute

172

W.C. Wang et al [32,33], stress [34,35], anxiety [36,37] and other psychosocial aspects [3540]. Nevertheless, the results were inconsistent: non-significant results were found in some studies [41,42] but significant in others [32,43]. More importantly, all the publications regarding TC and mental well-being are individual studies. Systematic reviews that address mental health outcomes have not been conducted. Therefore, we initiated this review to critically appraise and summarize studies that examined the effect of TC on psychosocial well-being and to address the apparent gap in the literature.

1. Introduction
Complementary and Alternative Medicine (CAM) has gained popularity in Western societies [13]. The cost for CAM services had increased 45% between 1990 and 1997 in the USA [4] and 62% between 1993 and 2000 in Australia [5]. The number of people using CAM is still growing [1]. This is because CAM can provide medical efficacies that conventional medicine may not be able to achieve [6]. Ernst [7] suggested several advantages that made CAM sometimes more preferable than conventional medicine. These include an emphasis on health rather than disease, a solution to chronic disease states, personalization, and empathy, which are more compatible with patients values and beliefs regarding the nature and meaning of health and illness [3,8]. Tai Chi (TC) is a traditional Chinese exercise/ movement that was thought to coordinate the body, mind and spirit. Through controlled breathing, interaction and moving of the body, one can reach body relaxation and mind calm and Tian Ren He Yi (a theory that states mankind is an integral part of nature) [9]. TC is also regarded as a mind-body exercise; it improves an individuals capacity to undertake more demanding aerobic exercises. Furthermore, TC is a low to middle impact exercise [10,11] with various forms such as Yang, Wu, or Su styles that improves flexibility and muscle strength. With suitable modifications, TC can be practiced by people of almost any age, at any time, in any place, without equipment; and is accepted by many populations including the elderly and people with different disabilities [12]. The effect of TC on disease prevention and health promotion has been well studied, including cardiovascular diseases [1317], the respiratory system [18,19], arthritis [20,21], and improves balance and lowers falls risk in older people [2224]. Systematic reviews have also been conducted to examine the effect of TC; however, all reviews focused primarily on physical status such as physical health outcomes in patients with chronic conditions [25], improved balance [26], falls prevention [27], and improved aerobic capacity [28]. Psychosocial well-being is essential to a persons quality of life [29]. For all individuals, mental health is as vital a strand of life as their physical and social health [30]. Mental health is largely reflected by ones psychosocial well-being such as levels of depression, anxiety, stress and life dissatisfaction, which may impact negatively on ones capacity to live a full and productive life [31]. Therefore, the improvement of mental wellness is critical to individuals, communities and societies. The present study seeks to investigate whether peoples psychosocial well-being can be enhanced through TC intervention. Some studies have explored the effect of TC on mental health status, such as mood disturbance

2. Methods
2.1. Search strategy
An electronic literature search was conducted using seven sources: (i) MEDLINE; (ii) CINAHL; (iii) EMBASE; (iv) PsycINFO; (v) CISCOM; (vi) the Cochrane Central Register of Controlled Trials (CENTRAL) of the Cochrane Library; and (vii) dissertations and conference proceedings. The databases were searched from their inception to August 2008. The search keywords used included different combinations of: Tai Chi, Tai Chi Chuan, Tai Chi, Taijiquan, psychosocial, stress, anxiety, depression, mood, and sleep disorder. Only studies that met the inclusion criteria were reviewed, which included: (i) employing a randomized controlled design; (ii) using TC as the intervention and other forms of activities or waitinglist as control; and (iii) using psychosocial well-being such as depression, stress, anxiety, and sleep disorder as outcome measures. We excluded non-English publications in this review due to the concerns of the reporting qualities of studies published in other languages [44].

2.2. Reliability assessment

In this review, reliability refers to the difference between reviewers in the processes of literature searching, extraction, and appraisal. Two researchers conducted the literature search independently. Articles were screened and selected after reading the title and abstract. The full text was obtained for further assessment when the abstract did not provide enough information to make a judgment. Disagreements between the two researchers were resolved by consensus or through discussion with a third researcher.

2.3. Methodological quality assessment

The same two reviewers independently graded each study using a modified Jadad scale [45]. The modified Jadad used is an 8-item scale designed to assess

Tai Chi on psychosocial well-being randomization, blinding, withdrawals/dropouts, inclusion/exclusion criteria, adverse effects and statistical analysis. The score for each article can range from 0 (lowest quality) to 8 (highest quality). Scores of 4 to 8 represent good to excellent (high quality). The Jadad scale has been widely used because it is simple, short, reliable and valid. Again, disagreements between the two researchers were resolved by consensus or through discussion with a third researcher.
Potentially relevant articles identified (n = 160)

173

Reviews, letters to editor, abstracts or other publications excluded (n = 16) Studies retrived for more detailed evaluation (n = 144) Duplicated and Non-English studies excluded (n = 21) Potentially appropriate studies to be included in the review (n = 123) Not psychosocially related studies excluded (n = 83) Potentially more appropriate studies for the review (n = 40) Non TC & RCT & Quasi-experimental studies excluded (n = 25) RCTs with usable outcome measures (n = 15) Figure Flow chart of trial selection process.

2.4. Data extraction and analyses

For each eligible publication, information extracted and recorded included: (i) name of the author(s); (ii) year of publication; (iii) study design including intervention and control group information; (iv) country in which the study was carried out; (v) TC style; (vi) duration of intervention; (vii) sample size; (viii) participants demographic characteristics; (ix) primary and all other outcome measures; (x) instruments used for assessment of outcomes; (xi) results; and (xii) effect size and 95% confidence interval. Data were analyzed using Cochrane Review Manager software (RevMan) 5.0 [46]. The effect size was calculated for each of the outcome measures based upon the available data reported in the randomized clinical trials (RCTs). Meta-analysis was not performed because of the heterogeneity of the study conditions and outcome measures used in the included trials.

3. Results
3.1. Search results
Using the pre-defined keywords, a total of 687 publications pertaining to the practice of TC and psychosocial well-being were found. After reading the titles and abstracts, 527 publications were excluded, leaving 160 for full paper evaluation. Of these 160 articles, 145 were excluded because of non psychosocial outcomes (n = 83), other types of publications (e.g. reviews, n = 16), quasi-experimental designs (n = 14), non-English publications (n = 13), duplication (n = 8), non RCTs (n = 6), and non TCs (n = 5). Finally, 15 papers that specifically examined the effect of TC on mental health outcomes were critically appraised. Figure illustrates the trial selection process.

3.2. Study characteristics

Table presents the descriptive information of each study reviewed. The trials were conducted between 1992 and 2008 in the United States, the United Kingdom, Australia, Hong Kong and Taiwan.

Six studies focused on people aged 59 years and over. In total, there were 1,229 participants aged over 18 years. Nine trials had less than 25 participants in one or more of the study groups. Participants included patients with moderate heart failure, chronic symptomatic hip or knee osteoarthritis, HIV/AIDS, depression and healthy sedentary adults. The TC intervention was mainly a modified Yang style (i.e. a modification of Yang style, that to be performed slowly with a slow breathing technique). The intensity of TC varied between once and three times, 40 to 90 minutes per week for 6 to 48 weeks. Home practice was required only in some studies but was not monitored. Most studies randomly assigned participants into two groups (i.e. a parallel clinical trial), with an intervention group receiving TC and a control group receiving no treatment [33,37,47,4851]; or three [32,41,43] and four [52] groups with other types of activities such as health education [53,54], walking [42], meditation and reading [52]. The psychosocial outcomes measured in the studies varied. Anxiety and depression were measured by eight studies; mood by four studies; stress by two studies; SF-12/36 mental health scores by three studies; anger, positive and negative effect, self-esteem, life satisfaction, social interaction and self-rated health by one study.

174

Table Participants conditions/ diseases Age (yr) Results Both TC and control groups showed a significant reduction in anxiety scores, but there was no difference between the groups. Depression scores showed significant reductions in TC group, no significant changes in control group Significant difference for mood improvement found in BDI (Beck Depression Inventory) scores between groups 4690 55 minutes per M = 69.4 session, 2 times per week; home practice encouraged Anxiety (SCL-90) Depression (SCL-90) Intervention & TC style Outcome measures Chronic heart failure Effect size (95% CI)

Summary of studies of Tai Chi on psychosocial well-being

First author, year 32, 33 (7, 6)

Modified Jadad

Duration of n intervention (Withdrew) (wk)

Barrow, 2007

16

Bhatti, 1998

26, 25 (1, 0)

Back pain

1865 90 minutes per session, 1 time per week; participants were required to practice 15 minutes per day 45 minutes per session, 3 times per week Mood (POMS) Anxiety (STAI) Anger (STAXI) Positive and negative effect (PANAS) Self-esteem (RSES) Life satisfaction (LSES) M = 72.6 45 minutes SD = 4.2 per session, 3 times per week; 18-form Yang Style Depression (CESD) Depression (BDI)

Brown, 1995

16

24, 34, 28, 18, 31 (total 9)

Sedentary people

4069

Female in the TC group experienced reductions in mood disturbance and general mood. Female TC group achieved a significantly greater decrease in anger Male in the moderate intensity walking group reported increased positive affect TC has a significant positive effect on reducing depressive symptoms compared with control group. Social support might partly contribute to the beneficial effect of TC

Female: POMS total: 31.8 (41.56, 22.04) POMS depression 6.1 (5.81, 4.10) LSES MOOD 1.2 (0.67, 1.73) Male: PANAS positive affect 1.0 (1.93, 0.07) CESD: 23.1 (34.21, 11.99)

Chou, 2004

12

7, 7 (0, 0)

Depression

W.C. Wang et al

Fransen, 2007

12

55, 56, 41 (3, 8, 0)

Hip or knee osteoarthritis

M = 70 60 minutes per session, 2 times per week; modified 24-form Sun Style Depression (CESD-D) Anxiety (STAI) Sleep (PSQI) Exercise groups Depression experience significant 3.0 (7.95, 1.95) time improvements to Anxiety depression and anxiety 4.8 (1.62, 11.22) significant differences in the PSQI between group Both exercise groups improved in the overall health perception subscale compared with control. The POMS showed significant main effect for time in confusion bewilderment and tensionanxiety. Enhanced psychologic coping and improved social interactions (emerged from the qualitative data) Perceived health status; self-rated health No significant difference in perceived health status. No significant difference in self-rated health Depression, anxiety, stress (DASS21) SF-12 (mental health component) No significant difference among groups on SF-12 MCS, depression, anxiety, or stress Depression 0.1 (3.6, 3.1) Anxiety 0.7 (1.5, 2.9) Stress 0.01 (3.0, 3.0) SF-12-MCS 0.03(2.9, 2.3)

Tai Chi on psychosocial well-being

Frye, 2007

12

31, 30, 23 (8, 2, 2)

Sedentary people

5282 40 minutes M = 69.2 per session; modified Yang Style

Galantino, 2005

13, 13, 12 (0, 0, 0)

HIV/AIDS

2060 60 minutes per session, 2 times per week Mood(POMS)anxiety, social interaction (by qualitative recorded),

Greenspan, 2007

48

148, 143 (11, 11)

Old people transitioning to frailty

> = 70

Starting at 60 minutes and progressing to 90 minutes per session, 2 times per week; Six simplified forms 60 minutes per session, 2 times per week; Yang Style

Hartman 2000

12

19, 16 (1, 1)

Lower extremity osteoarthritis

4981 M = 68

Mood, tension, satisfaction with general health (AIMSII)

TC group experienced significant improvements in level of tension and satisfaction with general health status

Mood 0.5 (1.43, 0.43) Tension 1.9 (3.29, 0.51) Satisfaction 1.5 (2.37, 0.63) (Contd.)

175

176

Table Participants conditions/ diseases Age (yr) Results M = 70 40 minutes per session, 3 times per week Depression (BDI) SF-36mental health Intervention & TC style Outcome measures Healthy adults

Continued Effect size (95% CI)

First author, year 59, 53 (7, 3)

Modified Jadad

Duration of n intervention (Withdrew) (wk)

Irwin, 2007

16

For severity of depressive Depression score symptoms (BDI), an 0.35 (2.52, 1.82) overall time effect was SF-36-Mental health found, with significant 1.3 (2.68, 5.34) improvements in the TC and Health Education groups. For SF-36 measure, time effects (from baseline to week 25) for mental health were found in TC group The mood states were significantly improved after all four treatments. TC group also experienced more reduction in state of anxiety than did the reading group

Jin, 1992

24, 24, 24, 24 (0, 0, 0, 0)

Healthy adults

M = 36.2 1 hour; the Anxiety Long Form or (STAT Y-I), Yang Style or Mood (POMS) the Wu variation of the Yang Style

Sun, 1996

12

10, 10 (0, 0)

Healthy Hmong older adults

6079 1 time per M = 66.8 week

Stress

Significant difference among groups on selfperceived stress score and stress level by body temperature Anxiety TC group experienced time significant improvements in trait anxiety and state anxiety

Self-perceive stress score 10.3 (9.05, 11.55) Stress level by body temperature 8.7 (7.02, 10.38) Trait anxiety 9.0 (12.0, 5.98) state anxiety 7.6 (10.69, 4.51)

Tsai, 2003

12

44, 44 (7, 5)

Blood pressure at high-normal or stage 1 hypertension Rheumatoid arthritis

3565 50 minutes M = 52.0 per session, 3 times per week; Yang style M = 49.5 60 minutes per session,

W.C. Wang et al

Wang, 2008

12

10, 10 (0, 0)

Anxiety/ depression

TC group improved significantly more

EQ-5D 16.3 (3.9, 36.5)

Tai Chi on psychosocial well-being

177

3.3. Methodological characteristics

CES-D 5.1 (8.93, 1.27) SF-36 MCS 7.1 (2.5,16.76)

(EQ-5D), Depression (CES-D) SF-36-mental summary score (MCS)

Depression (CESD)

Studies were classified into two quality categories: (i) high quality with a modified Jadad score of 4 and above; and (ii) non-high quality with a modified Jadad score of 3 and below. After the review, eight studies were classified as high quality trials (Table). Only five of the eight high quality trials had detailed explanation of how randomization was performed and provided adequate report on the assignment [32,33,41,54,55]. Participants were not blinded to treatment allocation because TC was a physical intervention. Ten studies reported the dropout rates, ranging from 2% to 26%, but five studies did not clearly explain the reasons for withdrawals/ dropouts [42,4952]. Only three studies indicated the acceptability of the withdrawal/dropout rate [41,47,54]. Four non-high quality trials did not report the group similarities at baseline [42,43,50,52]. Most trials followed the intention-to-treat principle except three non-high quality studies [43,48,49].

TC has a significant positive effect on reducing depressive symptoms compared with control group. Social support might partly contribute to the beneficial effect of TC.

than the control group on the vitality subscale of SF-36 and the CES-D

3.4. Effect of Tai Chi 3.4.1. Depression

Nine studies examined depression using several depression scales. The Beck Depression Inventory [56] was used by two studies [48,54]. The Centre for Epidemiology Studies Depression (CES-D) Index [57] was used in four trials [43,49,51,55]. The SCL-90-R Depression and Anxiety [58] Scale, the Depression, Anxiety and Stress Scale (DASS21) [59], the Profile of Mood States (POMS) [60], and the EuroQol (EQ-5D) [61] were used in one study each [38,39,42,50]. Using between group comparisons after intervention, significant TC effect on reducing depression was reported by one high quality [55] and two other trials [48,49]. Moreover, significant within group effect after TC intervention was documented by two high quality [47,54] and four other studies [42,43,49,51]. The effect size on mean score change ranged from 5.1 to 23.1. However, one high quality trial found no significant impact of TC on depression for both between and within group comparisons [41].

(functional class I or II)

12

7, 7 (0, 0)

Depression

M = 72.6 45 minutes SD = 4.2 per session, 3 times per week; 18-form Yang style

2 times per week; Yang style

M = mean, SD = standard deviation.

3.4.2. Anxiety
Eight studies examined the association between TC and anxiety. Various anxiety measures were employed. The State-Trait Anxiety Inventory (STAI) [62] was used by four studies [37,42,43,52]. The SCL90-R Depression and Anxiety [58], the POMS, the DASS21 [59], and the EQ-5D [61] were used in one study each [29,38,42,50].

Chou, 2008

178 The between and within group differences after intervention were found to be non-significant in two high quality trials [41,55]. The remaining two high and four non-high quality studies demonstrated significant within group effects. The only effect size documented was a mean score change of 9.0 (95% CI 12.0, 5.98), which was reported by a high quality study [37].

W.C. Wang et al that perceived health status and self-rated health were not sensitive to TC intervention.

4. Discussion
In the present review, 13 studies reported a beneficial effect of TC on at least one of the psychosocial well-being variables examined. The effect appeared to be most significant on depression (nine studies), anxiety (six studies) and mood (four studies). While these results might be encouraging, they should be interpreted cautiously. This is so because, of the 13 studies, only six were high quality trials. It is also worth remembering that two high quality studies found no significant TC effects [41,53]. Moreover, significant between-groups difference (i.e. parallel group comparison) after TC intervention was observed in only four studies [48 50,55], in which only one was classified as a high quality trial [50]. Significant improvement in mental health outcomes was detected in 10 studies performing within group comparisons (i.e. before and after the TC intervention). The TC group did demonstrate significant reductions in post test measures on well validated assessment tools, which was not observed in the control group. But these 10 trials did not find any significant difference between the TC and the control group. Hence, the effect of TC intervention can be described as inconclusive when the assessment was based on whether TC exercise was more effective than a control condition. On the other hand, our review showed that only nine trials (60%) provided either an estimate of the effect size or information needed to compute the effect size (for at least one outcome variables measured). The remaining six trials failed to provide sufficient information for obtaining effect sizes. Of the 22 effect sizes reported, only 11 (50%) were statistically significant. It is therefore reasonable to conclude that more evidence is needed to justify the effectiveness of TC on mental health and psychosocial well-being. The above remarks we made are related to issues by and large attributable to the shortcomings of the RCTs [69,70]. In this review, only 53% (8 out of 15) of studies that were critically appraised could be classified as high quality trials using the modified Jadad score of 4 and above. The overall quality score varied between 2 and 6 (out of 8 points), with a mean of 3.7 points, indicating an overall insufficient methodological quality. It is important to note that a quality assessment tool that has been specifically designed for manual therapy such as TC intervention is not available, thus, the modified Jadad scale was used in the current review.

3.4.3. Mood
Mood was examined in four trials, two of which were high quality studies [32,33]. POMS [60] was used in three studies [29,39,47]. The AIMS II scale [63,64] and the Positive and Negative Affect Schedule (PANAS) [65] were used in one study each [30,39]. Significant improvement on mood was demonstrated by three trials. Only one study reported the effect size on mean score change [42], which was 1.2 (95% CI 0.67, 1.73).

3.4.4. General mental health

Two high quality trials [54,55] measured general mental health using the SF-36 mental health component [66]. Both studies found non-significant effect of TC on general mental health, in a between-group setting after the TC intervention. However, Irwin (2007) found a pre- and post-significant difference in the TC group [54]. Another high quality trial [41] used SF-12 [67], but results were not significant using either between or within group comparisons.

3.4.5. Stress
Stress was examined by two studies, including one high quality [41] and one non-high quality [50] trial. Self-perceived stress [68] was used by Sun et al [50], who found significant TC within group effect on reducing stress [50]. The effect size was a mean score change of 10.3 (95% CI 9.05, 11.6). However, no significant reduction in stress, either between or within group, was found by Fransen et al [41] who used the DASS21 [59].

3.4.6. Other outcome measures

Other mental health outcomes examined included sleep [43], tension [33], anger [42], life satisfaction [42], satisfaction with general health [33], selfesteem [42] and perceived health status and selfrated health [53]. Among several low quality and one high quality [33] studies, TC was suggested to be effective for many mental problems such as sleep disturbance, tension, anger, self-esteem, dissatisfaction with life and health status. However, a high quality trial by Greenspan et al [53] found

Tai Chi on psychosocial well-being Although the modified Jadad scale was developed for pharmaceutical trials [45] it has been used widely in reviews of non-pharmaceutical trials such as acupuncture [71,72] and TC [44,73]. Inspection of the present scoring procedure showed that the adequacy of blinding was problematic for all the trials reviewed, even though blinding is important to prevent performance and detection bias [74]. However, blinding is often difficult in trials using exercise as the intervention [71,75]. All trials in the present review were scored zero points for their blinding procedure (i.e. none of the trials was double-blinded). Nevertheless, the blinding of outcome assessors is critical and possible for RCTs using exercise as the intervention in which the outcome assessor is required to not be involved in intervention administration [76]. Research has stressed that lack of blinding of outcome assessors can result in systematic differences in outcome assessment [77]. Noticeably, the current review showed that even studies with a high modified Jadad score were subject to this methodological issue. Future studies will therefore need to address this issue. Eight high quality and two other trials have provided the descriptions concerning study withdrawals and dropouts. However, the number of withdrawals and dropouts were not reported by the remaining five trials. Moreover, only three trials [41,47,54] reported the acceptability of the withdrawal and dropout rate. Since withdrawals and dropouts are essential elements directly related to bias reduction, future RCTs are advised to examine in detail the characteristics of the participants leaving the study before the end of the trial. It is well documented that following the intention-to-treat analysis procedures would limit the bias associated with non random loss of participants [78]. Interestingly enough, none of the studies have attempted to use an imputation method, such as the EM algorithm [79,80], to estimate the missing observations before data analysis, even though the intention-to-treat principle was reported to have been used in 12 studies. The EM algorithm has been proven to be a powerful method of imputing missing data in health related studies [81]. Researchers are advised to consider using such an approach in future clinical trials. Likewise, four trials did not describe the group similarities at baseline although it is essential for RCTs because group comparisons at the end of the study would not be meaningful if similarities of baseline data are not warranted [82]. Future RCTs should also focus more on meeting these criteria. From reviewing the 15 trials, we noticed that there were other aspects of study design and methodological criteria that are of concern. For example,

179 four studies [4951,55] had fairly small sample sizes (n < 25) per group, which could lead to low statistical power and wide confidence interval [83,84]. On the other hand, the duration of the interventions varied from 6 to 48 weeks with eight trials lasting for 12 weeks. It is not clear when, if ever, the positive effect of TC exercise was exhibited; i.e. the duration and follow-up period both can be important. Li et al [85] found that improvements in sleep quality was observed only after 24 weeks of TC practice, and only for adults over 60 years of age. It is not impossible that the 15 studies reviewed may not have sufficiently long intervention course or follow-up periods to demonstrate the effectiveness of TC for particular types of mental health measures. The heterogeneity of the participants in different age groups with various diseases/conditions at the baseline of the trials would have had impact on the study results. In addition, the miscellaneous instructional techniques of TC training may also limit the conclusions and generalizations that can be made about the effectiveness of TC on psychosocial well-being. The present review excluded non-English studies because of the concerns of the reporting qualities of trials that have been published in nonEnglish journals. However, the exclusion of nonEnglish studies could lead to language and citation biases as many TC literatures may have been published in Chinese and other Asian journals. In conclusion, there are some signs that TC can lead to improved psychosocial well-being. However, the evidence is still not strong enough for us to make any conclusive remarks, and no precise and accurate estimates of the effective size can be summarized. More well-designed RCTs on homogenous populations utilizing specific TC instructional technique and style with an appropriate follow-up period of time are required to evaluate the effect of TC on mental health. This could well be a fundamental re-starting point, which may lead to the modifications of existing TC styles for people with various mental disorders.

References
1. McCabe P. Complementary and alternative medicine in Australia: a contemporary overview. Complement Ther Clin Pract 2005;11:2831. Bensoussan A, Lewith GT. Complementary medicine research in Australia: a strategy for the future. Med J Aust 2004; 181:3313. Chesney MA, Straus SE. Complementary and alternative medicine: the convergence of public interest and science in the United States. Med J Aust 2004;181:3356. Eisenberg DM, Davis RB, Ettner SL, Appel S, Wilkey S, Van Rompay M, et al. Trends in alternative medicine use in the United States, 19901997: results of a follow-up national survey. JAMA 1998;280:156975.

2.

3.

4.

180
5. MacLennan AH, Wilson DH, Taylor AW. The escalating cost and prevalence of alternative medicine. Preventive Medicine 2002;35:16673. Coulter ID, Willis EM. The rise and rise of complementary and alternative medicine: a sociological perspective. Med J Aust 2004;180:587789. Ernst E. Complementary medicine: scrutinising the alternatives. Lancet 1993;341:1626. Astin JA. Why patients use alternative medicine: results of a national study. JAMA 1998;279:154853. Liu JY. The effect of Tai Chi Chuan on body-mind. Tamkang Journal 1975;13:21723. Lan C, Chen SY, Lai JS. The exercise intensity of Tai Chi Chuan. Med Sport Sci 2008;52:129. Lan C, Chen SY, Lai JS. Relative exercise intensity of Tai Chi Chuan is similar in different ages and gender. Am J Chin Med 2004;32:15160. Chen KM, Lin JN, Lin HS, Wu HC, Chen WT, Li CH, et al. The effects of a Simplified Tai-Chi Exercise Program (STEP) on the physical health of older adults living in long-term care facilities: a single group design with multiple time points. Int J Nurs Stud 2008;45:5017. Cheng TO. Tai Chi: the Chinese ancient wisdom of an ideal exercise for cardiac patients. Int J Cardiol 2007;117: 2935. Yeh GY, Wood MJ, Lorell BH, Stevenson LW, Eisenberg DM, Wayne PM, et al. Effects of tai chi mind-body movement therapy on functional status and exercise capacity in patients with chronic heart failure: A randomized controlled trial. Am J Med 2004;117:5418. Ades PA, Wu G. Benefits of tai chi in chronic heart failure: body or mind? Am J Med 2004;117:6112. Wolf SL, OGrady M, Easley KA, Guo Y, Kressing RW, Kutner M. The influence of intense Tai Chi Training on physical performance and hemodynamic outcomes in transitionally frail, older adults. J Gerontol 2006;61A:1849. Taylor-Piliae RE, Haskell WL, Stotts NA, Froelicher ES. Improvement in balance, strength, and flexibility after 12 weeks of Tai Chi exercise in ethnic Chinese adults with cardiovascular disease risk factors. Altern Ther Health Med 2006;12:5058. Hong Y, Li JX, Robinson PD. Balance control, flexibility, and cardiorespiratory fitness among older Tai Chi practitioners. Br J Sports Med 2000;34:2934. Lan C, Chen SY, Lai JS, Wong MK. Heart rate responses and oxygen consumption during Tai Chi Chuan practice. Am J Chin Med 2001;29:40310. Brisme JM, Paige RL, Chyu MC, Boatright JD, Hagar JM, McCaleb JA, et al. Group and home-based Tai Chi in elderly subjects with knee osteoarthritis: a randomized controlled trial. Clin Rehab 2007;21:99111. Song R, Lee EO, Lam P, Bae SC. Effects of tai chi exercise on pain, balance, muscle strength, and perceived difficulties in physical functioning in older women with osteoarthritis: a randomized clinical trial. J Rheumatol 2003;30: 203944. Lin MR, Hwang HF, Wang YW, Chang SH, Wolf SL. Communitybased tai chi and its effect on injurious falls, balance, gait, and fear of falling in older people. Phys Ther 2006; 86:1189201. Wolf SL, Sattin RW, Kutner M, OGrady M, Greenspan AI, Gregor RJ. Intense Tai Chi exercise training and fall occurrences in older, transitionally frail adults: a randomized, controlled trial. J Am Geriatr Soc 2003;51:1693701. Voukelatos A, Cumming RG, Lord SR, Rissel C. A randomized, controlled trial of tai chi for the prevention of falls: the central Sydney tai chi trial. J Am Geriatr Soc 2007;55: 118591.

W.C. Wang et al
25. Wang C, Collet JP, Lau J. The effect of Tai Chi on health outcomes in patients with chronic conditions: a systematic review. Arch Intern Med 2004;164:493501. 26. Komagata S, Newton R. The effectiveness of Tai Chi on improving balance in older adults: an evidence-based review. J Geriatr Phys Ther 2003;26:916. 27. Wu G. Evaluation of the effectiveness of Tai Chi for improving balance and preventing falls in the older population: a review. J Am Geriatr Soc 2002;50:74654. 28. Taylor-Piliae RE, Froelicher ES. The effectiveness of Tai Chi exercise in improving aerobic capacity: a meta-analysis. J Cardiovasc Nurs 2004;19:4857. 29. Testa MA, Simonson DC. Assessment of quality-of-life outcomes. N Engl J Med 1996;334:83540. 30. World Health Organization. World Health Report 2001 Mental Health: New Understanding, New Hope. In. Geneva: WHO; 2001. 31. Ruini C, Ottolini F, Rafanelli C, Tossani E, Ryff CD, Fava GA. The relationship of psychological wellbeing to distress and personality. Psychother Psychosom 2003;72:26875. 32. Galantino ML, Shepard K, Krafft L, LaPerriere A, Ducette J, Sorbello A, et al. The effect of group aerobic exercise and Tai Chi on functional outcomes and quality of Life for persons living with acquired immunodeficiency syndrome. J Altern Complement Med 2005;11:108592. 33. Hartman CA, Manos TM, Winter C, Hartman DM, Li B, Smith JC. Effects of Tai Chi training on function and quality of life indicators in older adults with osteoarthritis. J Am Geriatr Soc 2000;48:155359. 34. Sandlund ES, Norlander T. The effects of Tai Chi Chuan relaxation and exercise on stress responses and well-being: An overview of research. Int J Stress Manag 2000;7:13949. 35. Taylor-Piliae RE, Haskell WL, Waters CM, Froelicher ES. Change in perceived psychosocial status following a 12-week Tai Chi exercise programme. J Adv Nurs 2006;54:31329. 36. Bond DS, Lyle RM, Tappe MK, Seehafer RS, DZurilla TJ. Moderate aerobic exercise, Tai Chi, and social problemsolving ability in relation to psychological stress. Int J Stress Manag 2002;9:32943. 37. Tsai JC, Wang WH, Chan P, Lin LJ, Wang CH, Tomlinson B, et al. The beneficial effects of Tai Chi Chuan on blood pressure and lipid profile and anxiety status in a randomized controlled trial. J Alt Comp Med 2003;9:74754. 38. Liu Y, Mimura K, Wang L, Ikuta K. Psychological and physiological effects of 24-style Taijiquan. Neuropsychobiol 2005;52:2128. 39. Chan AS, Ho YC, Cheung MC, Albert MS, Chiu HFK, Lam LCW. Association between mind-body and cardiovascular exercises and memory in older adults. J Am Geriatr Soc 2005; 53:175460. 40. Wang YT, Taylor L, Pearl M, Chang LS. Effects of Tai Chi exercise on physical and mental health of college students. Am J Chin Med 2004;32:4539. 41. Fransen M, Nairn L, Winstanley J, Lam P, Edmonds J. Physical activity for osteoarthritis management: A randomized controlled clinical trial evaluating hydrotherapy or Tai Chi classes. Arthritis Care Res 2007;57:40714. 42. Brown DR, Wang Y, Ward ANN, Ebbeling CB, Fortlage L, Puleo E, et al. Chronic psychological effects of exercise and exercise plus cognitive strategies. Med Sci Sports Exer 1995;27:76575. 43. Frye B, Scheinthal S, Kemarskaya T, Pruchno R. Tai Chi and low impact exercise: effects on the physical functioning and psychological well-being of older people. J App Gerontol 2007;26:43353. 44. Yeh GY, Wang C, Wayne PM, Phillips RS. The effect of tai chi exercise on blood pressure: a systematic review. Prev Cardiol 2008;11:829.

6.

7. 8. 9. 10. 11.

12.

13.

14.

15. 16.

17.

18.

19.

20.

21.

22.

23.

24.

Tai Chi on psychosocial well-being

45. Oremus M, Wolfson C, Perrault A, Demersa L, Momolia F, Moride Y. Interrater Reliability of the Modified Jadad Quality Scale for Systematic Reviews of Alzheimers Disease Drug Trials. Dement Geriatr Cogn Disord 2001;12:2326. 46. Higgins J, Green S. Cochrane handbook for systematic reviews of interventions: the Cochrane Collaboration; 2008. 47. Barrow DE, Bedford A, Ives G, OToole L, Channer KS. An evaluation of the effects of Tai Chi Chuan and Chi Kung training in patients with symptomatic heart failure: a randomised controlled pilot study. Postgrad Med J 2007;83:71721. 48. Bhatti TI, Gillin JC, Atkinson JH, Jordan JE, Golshan S, Garfin SR, et al. Tai Chi Chih as a treatment for chronic low back pain: a randomized, controlled study. Altern Ther 1998;4:901. 49. Chou KL, Lee PWH, Yu ECS, Macfarlane D, Cheng YH, Chan SSC, et al. Effect of Tai Chi on depressive symptoms amongst Chinese older patients with depressive disorders: a randomized clinical trial. Int J Geriatr Psych 2004;19:11057. 50. Sun WY, Dosch M, Gilmore GD, Pemberton W, Scarseth T. Effects of a Tai Chi Chuan Program on Hmong American older. Educ Gerontol 1996;22:1617. 51. Chou KL. Effect of Tai Chi on depressive symptoms amongst Chinese older patients with major depression: The role of social support. Med Sport Sci 2008;52:14654. 52. Jin P. Efficacy of Tai Chi, brisk walking, meditation, and reading in reducing mental and emotional stress. J Psychosom Res 1992;36:36170. 53. Greenspan AI, Wolf SL, Kelley ME, OGrady M. Tai Chi and perceived health status in older adults who are transitionally frail: a randomized controlled trial. Phys Ther 2007; 87:52535. 54. Irwin MR, Olmstead R, Oxman MN. Augmenting immune responses to Varicella Zoster virus in older adults: a randomized, controlled trial of Tai Chi. J Am Geriatr Soc 2007; 55:5117. 55. Wang C. Tai Chi improves pain and functional status in adults with rheumatoid arthritis: results of a pilot singleblinded randomized controlled trial. Med Sport Sci 2008;52: 21829. 56. Beck AT, Steer RA, Carbin MG. Psychometric properties of the Beck Depression Inventory: twenty-five years of evaluation. Clin Psychol Rev 1988;8:77100. 57. Radloff LS: The CES-D Scale: a self-report depression scale for research in the general population. App Psychol Meas 1977;1:385401. 58. Derogatis LR. SCL-90-R, administration, scoring, and procedures manual for the revised version. Baltimore: Johns Hopkins University, School of Medicine; 1977. 59. Lovibond PF, Lovibond SH. The structure of negative emotional states: comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behav Res Ther 1995;33:33543. 60. Chuang HT, Devins GM, Hunsley J, Gill MJ. Psychosocial distress and well-being among gay and bisexual men with human immunodeficiency virus infection. Am J Psych 1989;146: 87680. 61. Hurst NP, Kind P, Ruta D, Hunter M, Stubbings A. Measuring health-related quality of life in rheumatoid arthritis: Validity, responsiveness and reliability of EuroQol (EQ-5D). Rheumatology 1997;36:5519. 62. Spielberger CD. State-Trait Anxiety Inventory: a comprehensive bibliography. Palo Alto, CA: Consulting Psychologists Press; 1989. 63. Meenan RF, Anderson JJ, Kazis LE, Egger MJ, Altz-Smith M, Samuelson CO, et al. Outcome assessment in clinical trials evidence for the sensitivity of a health status measure. Arthritis Rheum 1984;27:134452.

181
64. Meenan RF, Gertman PM, Mason JH. Measuring health status in arthritis. Arthritis Rheum 1980;23:14652. 65. Watson D, Clark LA, Tellegen A. Development and validation of brief measures of positive and negative affect: the PANAS scales. J Pers Soc Psychol 1988;54:106370. 66. McHorney CA, Ware JE Jr, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care 1993;31:24763. 67. Gandhi SK, Salmon WJ, Zhao SZ, Lambert BL, Gore PR, Conrad K. Psychometric evaluation of the 12-Item ShortForm Health Survey (SF-12) in osteoarthritis and rheumatoid arthritis clinical trials. Clin Ther 2001;23:108098. 68. Gilmore GD, Campbell MD, Becker BL. Needs assessment strategies for health education and health promotion. Indianapolis, IN: Benchmark; 1989. 69. Kuramoto AM. Therapeutic benefits of Tai Chi exercise: research review. WMJ 2006;105:426. 70. Verhagen AP, Immink M, van der Meulen A, Bierma-Zeinstra SMA. The efficacy of Tai Chi Chuan in older adults: a systematic review. Fam Pract 2004;21:10713. 71. Leo RJ, Ligot JJSA. A systematic review of randomized controlled trials of acupuncture in the treatment of depression. J Affect Disord 2007;97:1322. 72. Kwon YD, Pittler MH, Ernst E. Acupuncture for peripheral joint osteoarthritis: a systematic review and meta-analysis. Rheumatology 2006;45:13317. 73. Yeh GY, Wang C, Wayne PM, Phillips R. Tai Chi exercise for patients with cardiovascular conditions and risk factors: a systematic review. J Cardiopulm Rehab Prev 2009;29: 15260. 74. Moher D, Pham B, Jones A, Cook DJ, Jadad AR, Moher M, et al. Does quality of reports of randomised trials affect estimates of intervention efficacy reported in meta-analyses? Lancet 1998;352:60913. 75. White P, Lewith G, Berman B, Birch S. Reviews of acupuncture for chronic neck pain: pitfalls in conducting systematic reviews. Rheumatology 2002;41:122431. 76. Boutron I, Estellat C, Guittet L, Dechartres A, Sackett DL, Hrbjartsson A, Ravaud P. Methods of blinding in reports of randomized controlled trials assessing pharmacologic treatments: a systematic review. PLoS Med 2006;3:e425. 77. Hrbjartsson A. What are the main methodological problems in the estimation of placebo effects? J Clin Epidemiol 2002;55:4305. 78. Lachin JM. Statistical considerations in the intent-to-treat principle. Controlled Clin Trials 2000;21:16789. 79. Moon TK. The expectation-maximization algorithm. Signal Processing Magazine, IEEE 1996;13:4760. 80. Ramakrishnan V, Wang Z. Analysis of data from clinical trials with treatment related dropouts. Commun Stat: Sim Comp 2005;34:34353. 81. Jenkinson C, Heffernan C, Doll H, Fitzpatrick R. The Parkinsons Disease Questionnaire (PDQ-39): evidence for a method of imputing missing data. Age Ageing 2006;35:497502. 82. Lai JS, Lan C, Wong MK, Teng SH. Two-year trends in cardiorespiratory function among older Tai Chi Chuan practitioners and sedentary subjects. J Am Geriatr Soc 1995;43: 12227. 83. Bland JM. Sample size in guideline trials. Fam Prac 2000;17:S17S20. 84. Kirby A, Gebski V, Keech AC. Determining the sample size in a clinical trial. Med J Aust 2002;177:2567. 85. Li F, Fisher KJ, Harmer P, Irbe D, Tearse RG, Weimer C. Tai Chi and self-rated quality of sleep and daytime sleepiness in older adults: a randomized controlled trial. J Am Geriatr Soc 2004;52:892900.