Tai Chi and walking are both moderate-intensity physical activity (PA) that can be easily practiced in daily life. The objective of the study was to determine the effects of these two PAs on weight loss, metabolic syndrome parameters, and bone mineral density (BMD) in Chinese adults. We randomized 374 middle-aged subjects (45.8 ± 5.3 years) into 12-week training (45 minutes per day, 5 days per week) of Tai Chi
Along with the increasing prevalence of obesity and sedentary lifestyles worldwide, the metabolic syndrome (MetSyn) has become a global public health problem [
Evidences have shown that exercise is beneficial for preventing MetSyn [
Tai Chi (also called Tai Chi Chuan or Taiji), a traditional Chinese mind-body exercise, is popular in Chinese population, particularly middle-aged and elderly people. Its physiological and psychosocial benefits on health outcomes have been well addressed [
A 3-arm parallel-group C-RCT was designed. The whole research project has been briefly reported elsewhere [
Subjects were recruited from some large housing estates in the Shatin district of Hong Kong. These recruitment sites were classified as 9 geographic areas. Advertisements in flyers, surface mails, and bulletin boards were used for recruitment. The target subjects were aged 36 to 60. To screen eligible participants, all preliminary registered subjects were evaluated by health consultants using professional assessment form. Only inactive (no lifestyle physical activity or structured exercise experiences for at least 6 months [
Based on the cluster randomized trial design, fixed three clusters per arm (9 geographic areas/3 arms) were determined in advance. The planned number of participants was 360, 120 in each arm, 40 in each cluster. Based on the conventional assumptions of two-sided 5% significance level and assuming the intracluster correlation was 0.01, according to the formula for fixed number of clusters each of fixed size by Donner and Klar [
Participants recruited from one geographical area were considered as one cluster. The randomization was carried out at geographic area level for avoiding contamination. Nine clusters were then randomized to either Tai Chi, walking, or control arms, with the allocation ratio of 1 : 1. An independent statistician conducted the randomization using Excel to generate the allocation sequence. Another independent researcher critically carried out the allocation according to the sequence, to shield relevant investigators who might admit participants to the trial from knowing the upcoming assignments.
A modified 32-short form Yang-style Tai Chi Chuan was adopted in the Tai Chi group. Tai Chi integrates physical and spiritual elements to slowly and gently move
Individual information indluding sociodemographic characteristics (age, sex, residential area, and housing estate), medical history, and medication were obtained at baseline by face-to-face interviews. Participants were also required to provide a 1-week diet food frequency data, which were collected by the Food Frequency Questionnaire (FFQ). Dietary nutrients (energy, protein, and total fat) were then calculated. The MetSyn parameters include waist circumference, blood pressure (BP), FBG, total cholesterol, HDL-C, LDL-C, and triglycerides [
Weight, height, and waist circumference were measured to the nearest 0.1 kilogram or centimeter based on a standard protocol. BMI was then calculated (BMI = weight (kg)/height (m2)). All measurements were conducted twice by the trained research assistant; the average value was used. Waist circumference was measured by a standardized Gulick tape. To avoid the contraction of abdominal muscles, the measurement was recorded after the participant exhaled gently in normal breathing. The measurement position was at a level midway between the lower rib margin and the iliac crest. Obesity and central obesity were defined as BMI ≥ 25 kg/m2 and waist circumference ≥ 80 cm for women and ≥ 90 cm for men, respectively [
Body mass and BMD were measured by Hologic QDR-2000 dual-energy X-ray densitometer (Hologic, Bedford, MA, US) at baseline and 12 weeks. The machine provides values for lean mass, fat mass, BMD, and total body water assuming that water constitutes 73.2% of lean mass. The BMD were measured at the hip (femoral neck, intertrochanteric area, and the total hip), the spine (L1–L4), and the total body. The values from total body were used in current analysis. The stability of the machine and the long-term precision of the measurement have been identified acceptable where the coefficient of variance (CV%) was less than 0.42% in a previous study using the same machine [
Overnight fasting (10–12 h) venous blood samples were obtained at 8–10 am for the measurement of glucose and lipid concentrations. Blood withdrawal was conducted by professional nurses. Participants who had acute inflammation or taking anti-inflammation drugs (i.e., aspirin or antibiotics) at the time of completing 12-week program were required to postpone 1 week for blood withdrawal. The blood samples were centrifuged at 3000 ×g for 15 min under condition of 4°C. The serum was isolated within 2 h after collection and divided into several aliquots and stored at −85°C until analysis. Serum was measured by enzymatic methods and serum ACE (Gcell, Beijing Strong Biotechnologies, Inc.) by colorimetric assay. The FBG, total cholesterol, high-density lipoprotein cholesterol (HDL-C), and triglycerides were measured with a modified hexokinase enzymatic method (7020 clinical analyzer, Hitachi, Tokyo, Japan).
Statistical analysis was performed by SPSS 20.0 (SPSS Institute) software. Data were analyzed according to an intention-to-treat (ITT) principle. All participants who attended baseline assessment were involved in the ITT. Baseline differences between three groups were compared by one-way ANOVA and Pearson chi-square test for continuous and categorical variables, respectively. Post hoc procedures were further conducted for the pairwise comparison. One-way analysis of covariance (ANCOVA) was used to compare the mean changes of outcomes from baseline to after intervention (12-week) between three groups. To compare the between-group differences in mean change from baseline to 12-week, repeated measures of ANCOVA were implemented. The time × group interaction effects between intervention group and control group, as well as between the two intervention groups, were examined. The variables that were significantly different between groups at baseline were adjusted as covariates in the ANCOVA models. Bivariate correlations between main outcomes at baseline, as well as the intentional weight (and body mass) loss, and changes in BMD and MetSyn parameters at 12 weeks were examined by Spearman’s correlation test.
A total of 398 subjects from 9 locations were successfully enrolled in the study. 24 subjects (3 in the Tai Chi group, 2 in the walking group, and 19 in the control group) quit after randomization because they were unwilling to participate in the allocated group. Finally 374 subjects attended baseline assessment. Detailed study flow and intervention allocating were indicated in the Supplementary Figure
The mean age of participants in Tai Chi, walking, and cotrol groups was
Baseline characteristics of participants and comparisons among groups.
Tai Chi group | Walking group | Control group |
| |
---|---|---|---|---|
Geographic areas ( |
3 | 3 | 3 | |
Participants ( |
129 | 121 | 124 | |
Age group, |
0.068 | |||
36–40 | 18 (14.0) | 14 (11.6) | 30 (24.2) | |
41–50 | 89 (69.0) | 82 (67.8) | 71 (57.3) | |
51–60 | 22 (17.1) | 25 (20.7) | 23 (18.5) | |
Gender, |
0.863 | |||
Female | 77 (59.4) | 75 (62.0) | 78 (62.9) | |
Male | 52 (40.3) | 46 (38.0) | 46 (37.1) | |
Dietary intake | ||||
Energy, kCal/day, mean (SD) | 2094.9 (838.8) | 2159.4 (1135.1) | 2216.2 (898.6) | 0.608 |
Protein, g/day, mean (SD) | 93.7 (20.4) | 100.4 (28.1) | 88.1 (38.7) | 0.494 |
Fat, g/day, mean (SD) | 88.5 (25.5) | 84.6 (29.1) | 85.9 (25.1) | 0.909 |
Obesity, BMI ≥ 25 kg/m2, |
0.446 | |||
Yes | 41 (31.8) | 32 (26.4) | 31 (25.0) | |
No | 88 (68.2) | 89 (73.6) | 93 (75.0) | |
Central adiposity, waist circumference ≥ 80 cm, |
0.004 | |||
Yes | 44 (34.1) | 42 (34.7) | 22 (17.7) | |
No | 85 (65.9) | 79 (65.3) | 102 (82.3) | |
Hypertension, |
0.180 | |||
Yes | 13 (10.1) | 20 (16.5) | 12 (9.7) | |
No | 116 (89.9) | 101 (83.5) | 112 (90.3) | |
Metabolic syndrome, |
0.662 | |||
Yes | 16 (12.4) | 13 (10.7) | 11 (8.9) | |
No | 113 (87.6) | 108 (89.3) | 113 (91.1) | |
Presence of any 1 of 5 risk factors of metabolic syndrome, |
0.064 | |||
Yes | 83 (64.3) | 72 (59.5) | 62 (50.0) | |
No | 46 (35.7) | 49 (40.5) | 62 (50.0) | |
Presence of any 2 of 5 risk factors of metabolic syndrome, |
0.142 | |||
Yes | 44 (34.1) | 30 (24.8) | 30 (24.2) | |
No | 85 (65.9) | 91 (75.2) | 94 (75.8) |
bThe presence of any 3 of 5 risk factors constitutes a diagnosis of metabolic syndrome: (1) central adiposity; (2) triglycerides ≥ 1.7 mmol/L; (3) high-density lipoprotein cholesterol < 1.0 mmol/L in males; <1.3 mmol/L in females; (4) systolic BP ≥ 130 and/or diastolic BP ≥ 85 mm Hg; (5) blood fasting glucose ≥ 100 mg/dL.
Table
Changes of weight, body mass, bone mineral density, and metabolic syndrome parameters from baseline to 12 weeks in each group.
Measurea | Tai Chi group ( |
Walking group ( |
Control group ( |
|
---|---|---|---|---|
Weight (kg) | ||||
Baseline | 61.1 (11.2) | 61.1 (11.0) | 59.9 (11.0) | 0.661 |
12-week | 60.6 (11.0) | 60.3 (10.8) | 60.0 (11.0) | 0.933 |
Mean change from baseline | −0.50 (−0.80 to −0.21) | −0.76 (−0.97 to −0.55) | 0.1 (−0.12 to 0.31) | <0.001 |
Total body fat mass (kg) | ||||
Baseline | 18.6 (49.8) | 18.8 (52.8) | 18.1 (60.9) | 0.580 |
12-week | 18.1 (51.9) | 18.2 (49.7) | 18.2 (60.7) | 0.980 |
Mean change from baseline | −0.47 (−0.79 to −0.15) | −0.59 (−0.92 to −0.26) | 0.09 (−0.07 to 0.26) | 0.002 |
Total body lean mass (kg) | ||||
Baseline | 39.4 (90.9) | 39.5 (92.1) | 39.2 (85.3) | 0.968 |
12-week | 39.4 (92.4) | 39.3 (92.5) | 39.1 (86.1) | 0.987 |
Mean change from baseline | −0.03 (−1.01 to 0.96) | −0.11 (−0.64 to 0.42) | −0.10 (−0.24 to 0.04) | 0.714 |
Waist circumference (cm) | ||||
Baseline | 80.9 (8.8) |
81.1 (9.4) |
76.1 (9.8) | <0.001 |
12-week | 77.7 (9.1) | 77.4 (9.6) | 76.5 (9.8) | 0.585 |
Mean change from baseline | −3.3 (−3.9 to −2.6) | −3.3 (−4.5 to −2.9) | 0.4 (−0.0 to 0.8) | <0.001 |
Systolic blood pressure (mm Hg) | ||||
Baseline | 114.8 (15.4) | 112.9 (16.0) | 113.3 (14.3) | 0.561 |
12-week | 112.3 (14.6) | 110.4 (14.3) | 112.0 (13.9) | 0.560 |
Mean change from baseline | −2.6 (−4.3 to −0.8) | −2.4 (−4.2 to −0.7) | −1.4 (−3.0 to 0.2) | 0.509 |
Diastolic blood pressure (mm Hg) | ||||
Baseline | 74.4 (11.7) | 74.9 (13.2) | 75.3 (11.0) | 0.929 |
12-week | 72.9 (11.5) | 71.7 (12.9) | 72.3 (11.9) | 0.704 |
Mean change from baseline | −1.8 (−3.2 to −0.3) | −3.2 (−4.7 to −1.7) | −2.9 (−4.4 to −1.5) | 0.397 |
Fasting blood glucose (mmol/L) | ||||
Baseline | 4.8 (0.5) |
4.8 (0.4) |
4.5 (0.4) | <0.001 |
12-week | 4.6 (0.5) | 4.6 (0.5) | 4.5 (0.5) | 0.355 |
Mean change from baseline | −0.17 (−0.22 to −0.11) | −0.21 (−0.28 to −0.14) | 0.01 (−0.06 to 0.09) | <0.001 |
Total cholesterol (mmol/L) | ||||
Baseline | 5.1 (0.9) | 5.1 (0.9) | 5.0 (0.9) | 0.363 |
12-week | 5.1 (1.0) | 4.9 (0.9) | 4.8 (0.8) | 0.064 |
Mean change from baseline | −0.07 (−0.16 to 0.03) | −0.13 (−0.23 to −0.03) | −0.18 (−0.27 to −0.09) | 0.235 |
HDL-C (mmol/L) | ||||
Baseline | 1.5 (0.4) | 1.7 (0.5) |
1.5 (0.4) | 0.035 |
12-week | 1.5 (0.4) | 1.6 (0.5) | 1.4 (0.4) | 0.033 |
Mean change from baseline | −0.03 (−0.08 to 0.03) | −0.09 (−0.14 to −0.05) | −0.10 (−0.15 to −0.06) | 0.087 |
LDL-C (mmol/L) | ||||
Baseline | 3.0 (0.8) | 2.9 (0.8) | 2.9 (0.8) | 0.645 |
12-week | 3.0 (0.9) | 2.9 (0.8) | 2.9 (0.7) | 0.528 |
Mean change from baseline | −0.01 (−0.11 to 0.09) | −0.03 (−0.11 to 0.06) | −0.03 (−0.12 to 0.06) | 0.932 |
Triglycerides (mmol/L) | ||||
Baseline | 1.4 (1.0) | 1.2 (0.7) | 1.3 (1.0) | 0.121 |
12-week | 1.4 (0.9) | 1.1 (0.6) | 1.2 (0.9) | 0.017 |
Mean change from baseline | 0.01 (−0.08 to 0.09) | −0.05 (−0.13 to 0.03) | −0.10 (−0.20 to 0.00) | 0.204 |
Bone mineral density (mg/cm2) | ||||
Baseline | 1037.3 (79.6) | 1039.3 (85.0) | 1034.1 (82.3) | 0.887 |
12-week | 1036.9 (79.1) | 1037.6 (85.0) | 1033.8 (83.5) | 0.930 |
Mean change from baseline | −0.39 (−6.57 to 5.80) | −1.65 (−5.16 to 1.85) | −0.33 (−2.43 to 1.78) | 0.480 |
HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol.
aValues were presented as mean (SD) for 12-week measurement and mean (95% CI) for mean change from baseline; mean change from baseline = 12-week evaluation − baseline evaluation.
bOne-way ANOVA was used to compare the mean difference at baseline and at 12 weeks between groups; univariate ANCOVA was used to compare the mean change difference between groups; variables with significant different between groups at baseline were adjusted as covariates.
Between-group differences of body weight, body mass, metabolic syndrome parameters, and bone mineral density after intervention.
Measure | Tai Chi versus controla | Walking versus controla | Tai Chi versus walkinga | |||
---|---|---|---|---|---|---|
Between-group difference (95% CI) |
|
Between-group difference (95% CI) |
|
Between-group difference (95% CI) |
|
|
Weight (kg) | −0.60 (−0.94 to −0.25) | 0.008 | −0.85 (−1.20 to −0.50) | <0.001 | 0.26 (−0.09 to 0.60) | 0.164 |
Total body fat mass (kg) | −0.56 (−0.96 to −0.17) | 0.004 | −0.68 (−1.08 to −0.28) | <0.001 | 0.12 (−0.28 to 0.51) | 0.353 |
Total body lean mass (kg) | 0.08 (−0.86 to 1.01) | 0.926 | −0.01 (−0.96 to 0.94) | 0.858 | 0.08 (−0.85 to 1.02) | 0.878 |
Waist circumference (cm) | −3.7 (−4.4 to −2.9) | <0.001 | −4.1 (−5.0 to −3.2) | <0.001 | 0.5 (−0.6 to 1.5) | 0.489 |
Systolic blood pressure (mm Hg) | −1.2 (−3.6 to 1.1) | 0.311 | −1.1 (−3.5 to 1.3) | 0.283 | 0.4 (−0.5 to 1.3) | 0.824 |
Diastolic blood pressure (mm Hg) | 1.2 (−0.9 to 3.2) | 0.568 | −0.3 (−2.4 to 1.8) | 0.340 | 1.4 (−0.7 to 3.5) | 0.201 |
Fasting blood glucose (mmol/L) | −0.18 (−0.27 to −0.09) | <0.001 | −0.22 (−0.31 to −0.13) | <0.001 | 0.04 (−0.05 to 0.13) | 0.366 |
Total cholesterol (mmol/L) | 0.11 (−0.02 to 0.24) | 0.095 | 0.05 (−0.09 to 0.18) | 0.269 | 0.06 (−0.07 to 0.20) | 0.453 |
HDL-C (mmol/L) | 0.07 (0.01 to 0.14) | 0.330 | 0.01 (−0.06 to 0.08) | 0.705 | 0.06 (−0.004 to 0.13) | 0.060 |
LDL-C (mmol/L) | 0.02 (−0.11 to 0.15) | 0.850 | 0.002 (−0.13 to 0.14) | 0.882 | 0.02 (−0.11 to 0.15) | 0.708 |
Triglycerides (mmol/L) | 0.10 (−0.02 to 0.23) | 0.085 | 0.05 (−0.17 to 0.08) | 0.693 | 0.06 (−0.06 to 0.18) | 0.265 |
Bone mineral density (mg/cm2) | −0.06 (−6.16 to 6.03) | 0.477 | 1.33 (−7.52 to 4.86) | 0.295 | 1.27 (−4.85 to 7.39) | 0.573 |
HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol.
From baseline to 12 weeks, on average, Tai Chi and walking groups significantly lost 0.50 and 0.76 kg of body weight and 0.47 and 0.59 kg of fat mass, respectively (all
Participants in two intervention groups also had significantly greater decreases in waist circumference and FBG than those in the control group (Table
Mean differences of body weight, total fat mass, waist circumference, and fasting blood glucose from baseline to 12 weeks by male and female in three groups.
The correlation coefficients between changes in body weight, fat mass, lean mass, waist circumference, BMD, and other MetSyn parameters at 12 weeks in the two intervention groups were shown in Table
Correlation coefficients of weight loss, fat mass, lean mass, waist circumference with BMD, and metabolic syndrome parametersa.
Δ BMD | Δ waist circumference | Δ SBP | Δ DBP | Δ FBG | Δ TC | Δ HDL-C | Δ LDL-C | Δ triglycerides | |
---|---|---|---|---|---|---|---|---|---|
Weight loss (Δ body weight) | −0.05 | 0.38 |
0.13 |
0.04 | 0.03 | 0.09 | −0.01 | 0.05 | 0.20 |
Δ fat mass | 0.05 | 0.19 |
0.03 | 0.06 | 0.07 | 0.12 |
0.05 | 0.11 |
0.04 |
Δ lean mass | 0.41 |
0.09 | −0.08 | −0.06 | −0.01 | −0.02 | −0.05 | −0.04 | 0.05 |
Δ waist circumference | 0.01 | 1.00 | 0.15 |
0.03 | 0.23 |
0.00 | 0.00 | 0.00 | 0.01 |
BMD: bone mineral density; SBP: systolic blood pressure; DBP: diastolic blood pressure; FBG: fasting blood glucose; TC: total cholesterol; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol.
aCorrelation analyses were conducted between pre-post differences (Δ) of body weight, fat mass, lean mass, waist circumference and pre-post differences (Δ) of bone mineral density (BMD) and other metabolic syndrome parameters. Analysis only included cases in Tai Chi and walking groups. Pre-Post difference (Δ) = post value − pre value.
Results from this exploratory study provide novel information about the effects of 12-week Tai Chi and brisk walking trainings on weight loss, BMD, and MetSyn parameters in middle-aged Hong Kong adults. We found that these two moderate-intensity, short-term PA programs both slightly reduced the body weight and fat mass and had significant improvements on waist circumference and FBG. These two interventions showed no apparent effect on BMD. Furthermore, we found that BMD only correlated with the lean mass; the exercise-induced weight loss, particular loss in fat mass, had no significant associations with the changes in BMD.
The effect of Tai Chi on weight loss is inconsistent in the literature. A study [
Exercise is a well-known lifestyle description for managing type 2 diabetes or impaired glucose tolerance. However, whether Tai Chi has any benefit on glucose control is conflicted in the literature [
No significant change in BMD was observed in our study. A study in Hong Kong elderly people indicated that short-term Tai Chi exercise (e.g., 12 weeks) may not provide sufficient training stimulation in improving bone health [
Our study presents several limitations. First, predetermined number of clusters per arm (9 clusters/3 arms) in our cluster RCT led to imbalance in some variables. Participants in the control group seemed to have better waist circumference and FGB than those in intervention groups at baseline. However, we suggest that the improvement effects in these two variables are unlikely due to worse shape in intervention groups, because the significant within-group improvement was also observed. Nevertheless, the effect sizes might be slightly overestimated. Second, double-blind study design was not available due to the difficulty of administration and operation. Participants in the intervention groups might have higher expectations of the treatment results. This awareness of intervention assignments might introduce some bias into the results. Third, 12 weeks of intervention may not be long enough to observe significant improvement on certain health outcomes; thus, a longer-term follow-up study is suggested.
In summary, our study was the first to comprehensively examine the effects of a type of mind-body exercise and a simple physical exercise on the MetSyn parameters, weight loss, and BMD with a certain large sample size and using a cluster RCT design. We found that these two exercise interventions moderately reduced the body weight and fat mass and improved the waist circumference and FBG, and the exercise-induced weight loss did not impact the BMD. We suggest that Tai Chi and walking are both feasible and promising daily moderate PA for middle-aged adults. Findings from this study provide referable information for current public health initiatives to health aging and future community-based moderate PA and lifestyle intervention programs.
The study protocol was approved by the Joint Chinese University of Hong Kong-New Territories East Cluster Clinical Research Ethics Committee. All participants returned a signed statement of informed consent.
All the authors have declared no competing interests.
Stanley Sai-Chuen Hui, Jean Woo, and Timothy Chi-Yui Kwok designed the study. Stanley Sai-Chuen Hui carried out the experiments. Yao Jie Xie and Stanley Sai-Chuen Hui analyzed the data and wrote the paper. All authors reviewed the paper.
This study was entirely supported by the Health and Health Service Research Fund (HHSRF) from the Food and Health Bureau, Hong Kong (HHSRF 02030511).