Tai Chi for type 2 diabetes mellitus

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

To assess the effects of Tai Chi for type 2 diabetes mellitus.

Background

Description of the condition

Diabetes mellitus is a metabolic disorder resulting from a defect in insulin secretion, insulin action, or both. A consequence of this is chronic hyperglycaemia (that is elevated levels of plasma glucose) with disturbances of carbohydrate, fat and protein metabolism. Long‐term complications of diabetes mellitus include retinopathy, nephropathy and neuropathy. The risk of cardiovascular disease is also increased. For a detailed overview of diabetes mellitus, please see under 'Additional information' in the information on the Cochrane Metabolic and Endocrine Disorders Group in the Cochrane Library (see 'About', 'Cochrane Review Groups (CRGs)').

Description of the intervention

Exercise or physical activity is one of the principal therapies for type 2 diabetes (Kirk 2007). A systematic review found that exercise can significantly reduce glycosylated haemoglobin A1c (HbA1c) levels by 0.6% (Thomas 2006). The rate of aerobic and resistance exercise necessary to achieve metabolic benefits in clinical trials has sometimes resulted in poor compliance (Brandon 2003), because a large proportion of adults with type 2 diabetes mellitus do not follow recommended physical activity guidelines (Mokdad 2003). A low‐impact, low‐intensity exercise such as Tai Chi may reduce poor compliance in this population and provide a beneficial alternative.

Tai Chi is a traditional Chinese martial art that has been practised for many centuries. The three major components of Tai Chi are movement, meditation and deep breathing (Li 2001a). There are various perspectives on how Tai Chi works. Eastern philosophy holds that Tai Chi unblocks the flow of 'Qi'. When Qi flows properly, the body, mind and spirit are in balance and health is maintained (Cohen 1997). Others believe that Tai Chi works in the same way as other mind‐body therapies, i.e. the connection between the mind and the body can relieve stress, combat disease and enhance physical well‐being (Li 2001a; Qiang 2010). Tai Chi combines deep diaphragmatic breathing and relaxation with movement, including many fundamental postural stances, and Qi is said to flow imperceptibly and smoothly from one to the other through slow and soft activity (Chinese Sport 1983). Physical responses to Tai Chi do not exceed 55% of maximum oxygen intake or 60% of the individual maximum heart rate (Li 2001b).

Adverse effects of the intervention

Exercise may lead to hypoglycaemia, falls, injuries, pain or fatigue.

How the intervention might work

A meta‐analysis showed that exercise significantly improves glycaemic control and reduces visceral adipose tissue and plasma triglycerides, but not plasma cholesterol, in people with type 2 diabetes, independently of weight loss (Thomas 2006). Tai Chi is a low‐impact, low‐intensity exercise, and people with diabetes who exercise regularly have better glycaemia control and cardiovascular outcomes than those who do not exercise (Kuramoto 2006; Li 2001b). Tai Chi also has an impact on muscle mass through slow and gentle movements (Orr 2006; Qin 2005).

An insulin receptor defect is an important risk factor in the pathology of type 2 diabetes (Youngren 2007). Tai Chi exercise may increase insulin sensitivity (Wang 2008). Furthermore, Tai Chi enhances type 1 T helper function along with an increase in blood interleukin (IL)‐12 levels in people with type 2 diabetes mellitus (Yeh 2009).

Why it is important to do this review

Exercise is one of the principal therapies for type 2 diabetes mellitus and has definite effects and few side effects. Exercise interventions significantly improve glycaemic control, as indicated by a decrease in HbA1c. Tai Chi may be especially useful for elderly type 2 diabetes patients. Although Tai Chi may improve insulin sensitivity and lead to better glucose control, the evidence of the effects of Tai Chi on type 2 diabetes are still limited and conflicting. A systematic review of the effects of Tai Chi on type 2 diabetes is warranted.

Objectives

To assess the effects of Tai Chi for type 2 diabetes mellitus.

Methods

Criteria for considering studies for this review

Types of studies

We included randomised controlled clinical trials (RCTs).

Types of participants

Adult (18 years and above) patients with type 2 diabetes mellitus.

Diagnostic criteria

To be consistent with changes in classification and diagnostic criteria for diabetes mellitus over the years, the diagnosis should be established using the standard criteria valid at the time of the beginning of the trial (for example, WHO 1998; ADA 1999; ADA 2008). Ideally, diagnostic criteria should have been described. If necessary, we will use the authors' definition of diabetes mellitus. We planned to subject diagnostic criteria to a sensitivity analysis.

Types of interventions

We planned to investigate the following comparisons of intervention versus control/comparator.

Intervention

• Tai Chi

Comparator

• Sham exercise (e.g. low‐intensity stretching exercises without resistance)

• No intervention

Concomitant interventions had to be the same in both, the intervention and comparator groups to establish fair comparisons.

Exclusion criteria

We not included trials performed on participants with impaired glucose tolerance (IGT) in the analysis. If trials reported the combined results for people with type 2 diabetes and IGT, we attempted to contact the authors of the article in order to obtain the individual patient data for type 2 diabetes. If this was unsuccessful, we excluded the trial.

Types of outcome measures

Primary outcomes

• Glycaemic control.

• Health‐related quality of life.

• Adverse events.

Secondary outcomes

• All‐cause mortality.

• Diabetic complications.

• Insulin resistance.

• Physiological capacity.

• Blood pressure.

• Lipid levels.

• Changes in anthropometric measures and body composition.

• Socioeoconomic effects.

Methods and timing of outcome measurement

• Glycaemic control: glycosylated haemoglobin A1c (HbA1c) measured at the end of the intervention.

• Health‐related quality of life: evaluated by a validated instrument such as SF‐36 or EuroQol, and measured at the end of the intervention.

• Adverse events: such as hypoglycaemic reactions, Tai Chi‐induced injuries measured at the end of the intervention.

• All‐cause mortality defined as dying from any reason during the treatment and measured at the end of the intervention.

• Diabetic complications: defined as retinopathy, nephropathy and neuropathy, and measured at the end of the intervention.

• Insulin resistance: evaluated by the homeostasis model assessment (HOMA) and measured at the end of the intervention.

• Physiological capacity: defined as muscle strength, peak muscle power, balance, and measured at the end of the intervention.

• Blood pressure: systolic and diastolic blood pressure measured at the end of the intervention.

• Lipid levels: as analysed by total cholesterol, LDL‐ and HDL‐cholesterol, triglycerides, and measured at the end of the intervention.

• Changes in anthropometric measures and body composition: body weight, body mass index (BMI), waist circumference, body fat, lean body mass, and measured at the end of the intervention.

• Socioeconomic effects: defined as the decreased of expenditure for type 2 diabetes mellitus and measured at the end of the intervention.

Summary of findings

We present a 'Summary of finding' table to report the following outcomes, listed according to priority.

1. HbA1c.

2. Health‐related quality of life.

3. Adverse events.

4. Diabetic complications.

5. All‐cause mortality.

6. Socioeconomic effects.

7. Blood pressure.

Search methods for identification of studies

Electronic searches

We searched the following sources on 16 August 2016 from inception to the specified date and placed not restrictions on the language of publication.

• Cochrane Central Register of Controlled Trials (issue 7 of 12, July 2016).

• Epub Ahead of Print, In‐Process & Other Non‐Indexed Citations, Ovid MEDLINE(R) Daily and Ovid MEDLINE(R) 1946 to Present.

• Embase 1974 to 2016 Week 33.

• ClinicalTrials.gov

• World Health Organization International Clinical Trials Registry Platform (ICTRP) (www.who.int/trialsearch/).

We also used the following Chinese sources for the identification of trials:

• Chinese Biological Medicine Database (CBM‐disc) (27 April 2015);

• China National Knowledge Infrastructure (CNKI) (27 April 2015);

• Chinese scientific periodical database of VIP INFORMATION (27 April 2015);

• Wanfang Data (27 April 2015).

We continuously applied a MEDLINE (via Ovid) email alert provided by the CMED (Cochrane Metabolic and Endocrine Disorders) group to identify newly published trials using the same search strategy as described for MEDLINE (for details on search strategies see Appendix 1).

Searching other resources

We tried to identify other potentially‐eligible trials or ancillary publications by searching the reference lists of retrieved included trials, (systematic) reviews, meta‐analyses and health technology assessment reports.

Data collection and analysis

Selection of studies

Two review authors (LZ, HL) independently scanned the abstract, title, or both, of every record retrieved, to determine which trialsshould be assessed further. We investigated all potentially‐relevant articles as full text. Where differences in opinion existed, we resolved any discrepancies through consensus or recourse to a third review author (JY). If resolution of a disagreement had not been possible, we planned to add the article to those 'awaiting assessment' and we planned to contact study authors for clarification. We present an adapted PRISMA (Preferred Reporting Items for Systematic Reviews and Meta‐Analyses) flow‐chart showing the process of study selection (Figure 1) (Liberati 2009).

Figure 1.

Study flow diagram.

Data extraction and management

For trials that fulfil inclusion criteria, two review authors (HZ, GS) independently abstracted key participant and intervention characteristics, and reported data on efficacy outcomes and adverse events using standard data extraction templates, with any disagreements to be resolved by discussion, or, if required, by consultation with a third review author (JY). (for details, see Table 9; Appendix 2; Appendix 3; Appendix 4; Appendix 5; Appendix 6; Appendix 7; Appendix 8; Appendix 9; Appendix 10; Appendix 11; Appendix 12; Appendix 13)

Table 1.

Overview of study populations

	Intervention(s) and comparator(s)	Sample sizea	Screened/eligible (N)	Randomised (N)	ITT (N)	Analysed (N)	Finishing trial (N)	Randomised finishing trial (%)	Follow‐upb
(1) Orr 2006	I: Tai Chi	39	72/38	18	2	17	16	89	16 weeks
	C: sham exercise			20	0	20	20	100
	total:			38	1	37	36	95
(2) Chen 2010	I: Tai Chi	39	626/155	56	0	50	50	89	12 weeks
	C: conventional exercise			48	0	44	44	92
	total:			104	0	94	94	90
(3) Kan2004	I: Tai Chi	‐	‐	26	0	26	26	100	3 months
	C: usual exercise			22	0	22	22	100
	total:			48	0	48	48	100
(4) Lam 2008	I: Tai Chi	40	272/60	28	3	24	21	75	6 months
	C: usual exercise			25	0	22	22	88
	total:			53	3	46	43	81
(5) Wang 2009	I: Tai Chi	‐	278/64	34	0	34	34	100	6 months
	C: control			30	0	30	30	100
	total:			64	0	64	64	100
(6) Xiao 2015	I: Tai Chi	‐		16	‐	16	16	100	3 months
	C: control			16	‐	16	16	100
				32	‐	32	32	100
(7) Youngwanichsetha 2013	I: Tai Chi	‐	‐	32	0	32	32	100	12 weeks
	C: usual exercise			32	0	32	32	100
	total:			64	0	64	64	100
(8) Zhang 2008	I: Tai Chi	‐	‐	10	0	10	10	100	14 weeks
	C: control			10	0	9	9	90
	total:			20	0	19	19	95
Grand total	All interventions			220			205
	All comparators			203			195
	All interventions and comparators			423			400

^*Assumed risk was derived from the event rates in the comparator groups

^aAccording to power calculation in study publication or report ^bDuration of intervention and/or follow‐up under randomised conditions until end of study

"‐" denotes not reported

C: comparator; I: intervention; ITT: intention‐to‐treat; N/A: not applicable

We provide information including trial identifier about potentially relevant ongoing trials in the Characteristics of ongoing studies and in the Appendix 5 'Matrix of study endpoints (publications)'. We tried to find the protocol of each included study, either in trial registers or in publications of study designs, or both, and specified the data in the Appendix 6 'Matrix of study endpoints (trial documents)'.

We sent an email request to all authors of included trials to enquire whether they were willing to answer questions regarding their trials. Appendix 14 shows the results of this survey. Thereafter, we sought relevant missing information on the trial from the primary author(s) of the article, if required.

Dealing with duplicate and companion publications

In the event of duplicate publications, companion documents or multiple reports of a primary study, we maximised yield of information by collating all available data and use the most complete data set aggregated across all known publications. In case of doubt, the publication reporting the longest follow‐up associated with our primary or secondary outcomes obtained priority.

Assessment of risk of bias in included studies

Two review authors (HZ, GS) assessed the risk of bias of each included study independently. We resolved disagreements by consensus, or by consultation with a third review author (JY).

We assessed risk of bias using The Cochrane Collaboration's tool for assessment of risk of bias (Higgins 2011a; Higgins 2011b). We assessed the following criteria.

• Random sequence generation (selection bias).

• Allocation concealment (selection bias).

• Blinding (performance bias and detection bias), blinding of participants and personnel assessed separately from blinding of outcome assessment.

• Incomplete outcome data (attrition bias).

• Selective reporting (reporting bias).

• Other potential sources of bias.

We assess outcome reporting bias by integrating the results of 'Examination of outcome reporting bias' (Appendix 7), 'Matrix of study endpoints (trial documents)' (Appendix 6) and 'Outcomes (outcomes reported in abstract of publication)' of the 'Characteristics of included studies' section(Kirkham 2010). This analysis formed the basis for the judgement of selective reporting (reporting bias).

We judged 'Risk of bias criteria' as 'low risk', 'high risk' or 'unclear risk' and evaluated individual bias items as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011a). We presented a 'Risk of bias' graph (Figure 2) and a 'Risk of bias summary' figure (Figure 3).

Figure 2.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

We assessed the impact of individual bias domains on study results at the endpoint and study levels.

For performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessors) and attrition bias (incomplete outcome data) we evaluated risk of bias separately for subjective and objective outcomes (Hróbjartsson 2013). We considered the implications of missing outcome data from individual participants.

We defined the following endpoints as subjective outcomes:

• Health‐related quality of life.

• Adverse events.

We will define the following endpoints as objective outcomes:

• HbA1c.

• All‐cause mortality.

• Complications of diabetes mellitus.

• Blood pressure.

• Weight (kg) or BMI.

• Lipid levels.

• Socioeconomic effects

Measures of treatment effect

We expressed dichotomous data as odds ratios or risk ratios with 95% confidence intervals (CIs). We expressed continuous data as mean differences (MDs) with 95% CIs. We expressed time‐to‐event data as hazard ratios (HRs) with 95% CIs.

Unit of analysis issues

We took into account the level at which randomisation occurred, such as cross‐over trials, cluster‐randomised trials and multiple observations for the same outcome.

Dealing with missing data

We obtained missing data from study authors, if feasible, and carefully evaluate important numerical data such as screened, randomised participants as well as intention‐to‐treat (ITT), and as‐treated and per‐protocol populations. We investigated attrition rates, e.g. drop‐outs, losses to follow up and withdrawals, and critically appraised issues of missing data and imputation methods (e.g. last observation carried forward (LOCF)).

Assessment of heterogeneity

In the event of substantial clinical, methodological or statistical heterogeneity, we did not reported study results as the pooled effect estimate in a meta‐analysis.

We identified heterogeneity (inconsistency) through visual inspection of the forest plots and by using a standard Chi² test with a significance level of α = 0.1. In view of the low power of this test, we also considered the I² statistic, which quantifies inconsistency across trials to assess the impact of heterogeneity on the meta‐analysis (Higgins 2002; Higgins 2003); where an I² statistic of 75% or more indicates a considerable level of heterogeneity (Higgins 2011a).

When we found heterogeneity, we would have attempted to determine possible reasons for it by examining individual study and subgroup characteristics.

We expected the following characteristics to introduce clinical heterogeneity.

• Type of Tai Chi (Yang, Wu, Sun and Chen's Tai Chi styles).

• Intensity of Tai Chi (frequency, duration).

• Age (equal to or less than 65 years, older than 65 years).

• Sex (male or female).

• Body mass (body mass index less than 25 versus over 25).

• Medication (receiving or not receiving medication).

• Complications (with complications or not).

Assessment of reporting biases

If we included 10 trials or more investigating a particular outcome, we planned to used funnel plots to assessed small study effects (Sterne 2011). Several explanations can be offered for the asymmetry of a funnel plot, including true heterogeneity of effect with respect to trial size, poor methodological design (and hence bias of small trials) and publication bias. We therefore interpreted results carefully.

Data synthesis

Unless there was good evidence for homogeneous effects across trials, we summarised primarily low risk of bias data using a random‐effects model (Wood 2008). We interpreted random‐effects meta‐analyses with due consideration of the whole distribution of effects, ideally by presenting a prediction interval. A prediction interval specifies a predicted range for the true treatment effect in an individual study (Riley 2011). In addition, we performed statistical analyses according to the statistical guidelines contained in the latest version of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011a).

Subgroup analysis and investigation of heterogeneity

We planned to carry out the following subgroup analyses and to investigate interactions.

• Age (equal to or less than 65 years versus older than 65 years).

• The type of Tai Chi (Yang, Wu, Sun and Chen's Tai Chi).

• Exercise frequency (less than three times per week versus more than three times per week).

• Duration of intervention (equal to or less than six months versus longer than six months).

Sensitivity analysis

We planned to perform sensitivity analyses to explore the influence of the following factors (when applicable) on effect sizes by restricting the analysis to:

• Published trials.

• Taking into account risk of bias, as specified in the Assessment of risk of bias in included studies section.

• Very long or large trials to establish the extent to which they dominate the results.

• trials using the following filters: diagnostic criteria, imputation, language of publication, source of funding (industry versus other), or country.

We also tested the robustness of the results by repeating the analysis using different measures of effect size (RR, OR etc) and different statistical models (fixed‐effect and random‐effects models).

Appendices

Appendix 1. Search strategies

Cochrane Central Register of Controlled Trials (Cochrane Library)

#1 [mh "Tai Ji"] #2 ("tai chi" or taichi or taichiquan or "tai ji" or taiji or taijiquan):ti,ab,kw #3 {or #1‐#2} #4 [mh "Diabetes Mellitus"] #5 diabet*:ti,ab,kw #6 ("non insulin* depend*" or "noninsulin* depend*" or noninsulindepend* or "non insulindepend*"):ti,ab,kw #7 (NIDDM or MODY or T2DM or T2D):ti,ab,kw #8 {or #4‐#7}

MEDLINE (Ovid SP)

1 Tai Ji/ 2 (tai chi or taichi or taichiquan or tai ji or taiji or taijiquan).tw,ot. 3 1 or 2 4 exp Diabetes Mellitus/ 5 diabet*.tw. 6 (non insulin* depend* or noninsulin* depend* or noninsulin?depend* or non insulin?depend*).tw. 7 (NIDDM or MODY or T2DM or T2D).tw. 8 or/4‐7 9 3 and 8

Embase (Ovid SP)

1 Tai Chi/ 2 (tai chi or taichi or taichiquan or tai ji or taiji or taijiquan).tw,ot. 3 1 or 2 4 exp Diabetes Mellitus/ 5 diabet*.tw. 6 (non insulin* depend* or noninsulin* depend* or noninsulindepend* or non insulindepend*).tw,ot. 7 (NIDDM or MODY or T2DM or T2D).tw. 8 or/4‐7 9 3 and 8 [10:Wong 2006"sound treatment studies" filter – BS version] 10 random*.tw. or clinical trial*.mp. or exp health care quality/ 11 9 and 10

ICTRP Search Platform (Standard search)

tai chi AND diabet* OR taichi AND diabet* OR taichiquan AND diabet* OR tai ji AND diabet* OR taiji AND diabet* OR taijiquan AND diabet* OR chi t ai AND diabet* OR chi tai AND diabet*

ClinicalTrials.gov (Basic search)

("tai chi" OR taichi OR taichiquan OR "tai ji" OR taiji OR taijiquan OR "chi t ai" OR "chi tai") AND (diabetes OR diabetic OR T2DM OR T2D)

Chinese Biological Medicine Database (CBM‐disc)

1 default:Tai Ji ‐limit:‐ 2 default:Tai Ji Quan ‐limit:‐ 3 mesh: tai ji/all the trees/all vice keywords ‐ limit:‐ 4 #1 or #2 or #3 5 mesh: diabetes mellitus, 2 type/all the trees/all vice keywords ‐ limit:‐ 6 #4 and #5

China National Knowledge Infrastructure (CNKI)

1 keywords=Expansion in both Chinese and English(Tai Ji) 2 keywords=Expansion in both Chinese and English(Tai Ji Quan) 3 #1 or #2 4 keywords=Expansion in both Chinese and English(Diabetes mellitus) 5 #3 and #4

Chinese scientific periodical database of VIP INFORMATION

Same as CNKI

Wanfang Data

Same as CNKI

Appendix 2. Description of interventions

	Intervention(s)	Comparator(s)
Orr 2006	Sun and Yang Tai Chi 12‐form	Sham exercise
Chen 2010	Chen‐stysle Tai Chi 13‐form	Aerobic exercise
Kan 2004	Tai Chi 24‐from	Usual exercise
Lam 2008	Yang and Sun style 20‐form	Usual exercise
Wang 2009	Tai Chi	Control
Xiao 2015	Tai Chi	Control
Youngwanichsetha 2013	Tai Chi 18‐form	Control
Zhang 2008	Tai Chi	Control
"‐" denotes not reported C: comparator; I: intervention; N: no; Y: yes

Appendix 3. Baseline characteristics (I)

	Intervention(s) and comparator(s)	Duration of intervention/duration of follow‐up (days, weeks, months, years ...)	Description of participants	Trial period (year to year)	Country	Setting	Ethnic groups (%)	Duration of diabetes (mean/range years (SD), or as reported)
Orr 2006	I: Tai Chi	16 wk ()	Older patients with T2DM (age > 50 years)	‐	Australia	Community people	Caucasian 94.4	8.5 (0‐25)
	C: sham exercise	16 wk ()	Older patients with T2DM (age > 50 years)	‐	Australia	Community people	Caucasian 85	9 (0.7‐50)
Chen 2010	I: Tai Chi	12 wk ()	Obese patients with T2DM	‐	Taiwan	Hospital‐based	Chinese 100	8.5 3.5
	C: conventional exercise	12 wk ()	Obese patients with T2DM	‐	Taiwan	Hospital‐based	Chinese 100	7.8 3.1
Kan 2004	I: Tai Chi	3 mo ()	T2DM	‐	China	‐	Chinese 100	> 3 years
	C: usual exercise	3 mo ()	T2DM	‐	China	‐	Chinese 100	> 3 years
Lam 2008	I: Tai Chi	6 mo ()	T2DM	‐	Australia	Community people	Australians 100	> 6 mo
	C: usual exercise	6 mo ()	T2DM	‐	Australia	Community people	Australians 100	> 6 mo
Wang 2009	I: Tai Chi	6 mo ()	T2DM	2007‐2008	China	Community people	Chinese 100	> 1 year
	C: control	6 mo ()	T2DM	2007‐2008	China	Community people	Chinese 100	> 1 year
Xiao 2015	I: Tai Chi	3 mo ()	Older adults with T2DM	‐	China	‐	Chinese 100	‐
	C: control	3 mo ()	Older adults with T2DM	‐	China	‐	Chinese 100
Youngwanichsetha 2013	I: Tai Chi	12 wk ()	Postpartum patients with T2DM	‐	Thailand	Postpartum patients	Thais 100	2.5 (1.2)
	C: control	12 wk ()	Postpartum patients with T2DM	‐	Thailand	Postpartum patients	Thais 100	2.8 (1.2)
Zhang 2008	I: Tai Chi	14 wk ()	Female patients with T2DM	‐	China	Residents	Chinese 100	4.4 (2.2)
	C: control	14 wk ()	Female patients with T2DM	‐	China	Residents	Chinese 100
‐ denotes not reported C: comparator; I: intervention; mo: months; SD: standard deviation; T2DM: type 2 diabetes; we: weeks

Appendix 4. Baseline characteristics (II)

	Intervention(s) and comparator(s)	Sex (female %)	Age (mean/range years (SD), or as reported)	HbA1c (%)	BMI (mean kg/m² (SD))	Blood pressure (mean mm Hg (SD))	Comedications/Cointerventions	Comorbidities
Orr 2006	I: Tai Chi	88.9	66 (8)	7.1 (0.9)	33.7 (5.0)	‐	‐	Primarily osteoarthritis, hypertension, dyslipidemia, coronary artery disease
	C: sham exercise	85.0	65 (8)	6.9 (0.9)	30.9 (7.2)	‐	‐
Chen 2010	I: Tai Chi	55.4	59 (6)	8.9 (2.7)	33.5 (4.7)	‐	Hypoglycemic drugs and diet plans	‐
	C: conventional exercise	58.3	57 (6)	8.8 (3.6)	33.2 (4.1)	‐	Hypoglycemic drugs and diet plans	‐
Kan 2004	I: Tai Chi	‐	‐	‐	‐	‐	Hypoglycemic drugs and diet plans	‐
	C: usual exercise	‐	‐	‐	‐	‐	Hypoglycemic drugs and diet plans	‐
Lam 2008	I: Tai Chi	46	63 (9)	8.5 (1.2)	32.4 (6.7)	127 / 77	‐	‐
	C: usual exercise	64	61 (12)	9.0 (1.4)	32.0 (6.3)	141 / 80	‐	‐
Wang 2009	I: Tai Chi	47.1	48 (11)	‐	‐	‐	‐	‐
	C: control	46.7		‐	‐	‐	‐	‐
Xiao 2015	I: Tai Chi	‐	65	7.0 (1.8)	‐	‐	‐	‐
	C: control	‐		7.0 (1.6)	‐	‐	‐	‐
Youngwanichsetha 2013	I: Tai Chi	100	35 (6)	7.9 (1.6)	26.8 (3.1)	131 / 78	‐	Postpartum
	C: control	100	36 (4)	8.0 (1.8)	27.5 (3.6)	131 / 81	‐	Postpartum
Zhang 2008	I: Tai Chi	100	57.4 (6.2)	‐	‐	‐	‐	‐
	C: control	100		‐	‐	‐	‐	‐
"‐" denotes not reported BMI: body mass index; C: comparator; HbA1c: glycosylated haemoglobin A1c; I: intervention; SD: standard deviation

Appendix 5. Matrix of study endpoints (publications)

	Endpoint reported in publication		Other than review's primary/secondary outcomes reported in publication	Subgroups reported in publication
	Review's primary outcomes	Review's secondary outcomes
Orr 2006	Insulin resistance (HOMA2), HbA1c	Body composition, mobility, physical function, muscle performance, balance, quality of life	Attitudes towards diabetes, habitual physical activity, food intake	‐
Chen 2010	HbA1c, serum lipid profile	Physical parameters of body weight and BMI	Serum malondialdehyde, C‐reactive protein	‐
Kan 2004	TG, TC, LDL‐C, HDL‐C, BMI	‐	FBG, FInS	‐
Lam 2008	HbA1c, BP	SF‐36, HOMA, TC, TG		‐
Wang 2009	SF‐36	‐	‐	‐
Xiao 2015	HbA1c, balance control	‐	‐	‐
Youngwanichsetha 2013	HbA1c	BP, body‐weight, BMI	FBG	‐
Zhang 2008	TC, HDL‐C, LDL‐C, TC, BP	‐	FBG, GSP, FPI	‐
"‐" denotes not reported; HbA1c: hemoglobin A1c; BMI: body mass index; FBG: fasting blood‐glucose; FInS: fasting insulin; TG: total cholesterol; TC: triglyceride; LDL‐C: low‐density lipoprotein cholesterol; HDL‐C: high‐density lipoprotein cholesterol; HOMA: homeostasis model assessment; SF‐36: short form 36 health survey questionnaire; IL‐6: interleukin‐6; IL‐18: interleukin‐18; BP: blood pressure; GSP: glycated serum protein; FPI: fasting plasma insulin

Appendix 6. Matrix of study endpoints (trial documents)

	Endpointa (P)/(S)/(O)	Sourceb
Orr 2006	(P) insulin resistance (HOMA2), HbA1c / (S) body composition, mobility, physical function, muscle performance, balance, quality of life, attitudes towards diabetes, habitual physical activity, food intake	Australian New Zealand Clinical Trials Registry. Trial ID: ACTRN12605000715673
Chen 2010	‐	‐
Kan 2004	‐	‐
Lam 2008	(P) HbA1c, HOMA, blood pressure, health‐related quality of life / (S) strength, body composition, balance	Australian New Zealand Clinical Trials Registry. Trial ID: ACTRN12606000008527
Wang 2009	‐	‐
Xiao 2015	‐	‐
Youngwanichsetha 2013	‐	‐
Zhang 2008	‐	‐
"‐" denotes no trial document available a(P) Primary or (S) secondary endpoint(s) refer to verbatim statements in the publication, (O) other endpoints relate to outcomes which were not specified as 'primary' or 'secondary' outcomes in the report. bFDA/EMA document/manufacturer's website/design paper/trial protocol document EMA: European Medicines Agency; FDA: Food and Drug Administration (US); HbA1c: hemoglobin A1c; HOMA: homeostasis model assessment.

Appendix 7. Examination of outcome reporting bias according to ORBIT classification

	Outcome	High risk of bias (category A)a	High risk of bias (category D)b	High risk of bias (category E)c	High risk of bias (category G)d
Orr 2006	N/A	N/A	N/A	N/A	N/A
Chen 2010	N/A	N/A	N/A	N/A	N/A
Kan 2004	N/A	N/A	N/A	N/A	N/A
Lam 2008	Waist circumference and balance		X
Wang 2009	N/A	N/A	N/A	N/A	N/A
Xiao 2015	N/A	N/A	N/A	N/A	N/A
Youngwanichsetha 2013	N/A	N/A	N/A	N/A	N/A
Zhang 2008	N/A	N/A	N/A	N/A	N/A
aClear that outcome was measured and analysed; trial report states that outcome was analysed but only reports that result was not significant (Classification 'A', table 2, Kirkham 2010) bClear that outcome was measured and analysed; trial report states that outcome was analysed but no results reported ( Classification 'D', table 2, Kirkham 2010) cClear that outcome was measured; clear that outcome was measured but not necessarily analysed; judgement says likely to have been analysed but not reported because of non‐significant results (Classification 'E', table 2, Kirkham 2010) dUnclear whether the outcome was measured; not mentioned but clinical judgement says likely to have been measured and analysed but not reported on the basis of non‐significant results (Classification 'G', table 2, Kirkham 2010) N/A: not applicable

Appendix 8. Definition of endpoint measurement (I)

	Glycaemic control	Body mass index	Weight	Health related quality of life	All‐cause mortality	Blood pressure	Lipid levels	Diabetic complications	Socioeconomic effects
Orr 2006	HbA1c	kg/m²	kg	SF‐36	N/I	N/I	N/I	N/I	N/I
Chen 2010	HbA1c, FBG	kg/m²	kg	N/I	N/I	N/I	TG, TC, HDL‐C	N/I	N/I
Kan 2004	FBG	kg/m²	kg	N/I	N/I	N/I	TG, TC, HDL‐C, LDL‐C	N/I	N/I
Lam 2008	HbA1c	N/I	kg	SF‐36	N/I	DBP, SBP	TG, TC	N/I	N/I
Wang 2009	N/I	N/I	N/I	SF‐36	N/I	N/I	N/I	N/I	N/I
Xiao 2015	HbA1c	N/I	N/I	N/I	N/I	N/I	N/I	N/I	N/I
Youngwanichsetha 2013	HbA1c, FBG	kg/m²	kg	N/I	N/I	DBP, SBP	N/I	N/I	N/I
Zhang 2008	GSP, FBG	N/I	N/I	N/I	N/I	DBP, SBP	TG, TC, HDL‐C, LDL‐C	N/I	N/I
ND: not defined; N/I: not investigated; HbA1c: hemoglobin A1c; FBG: fasting blood‐glucose; TG: total cholesterol; TC: triglyceride; HDL‐C: high‐density lipoprotein cholesterol; LDL‐C: low‐density lipoprotein cholesterol; DBP: diastolic blood pressure; SBP: systolic blood pressure; GSP: glycated serum protein

Appendix 9. Definition of endpoint measurement (II)

	Insulin resistance	Physiological capacity (muscle strength, peak muscle power, balance)	Mild hypoglycaemia	Moderate hypoglycaemia	Severe hypoglycaemia	Nocturnal hypoglycaemia	Severe/serious adverse events
Orr 2006	HOMA	Balance, physical performance, muscle performance, habitual physical activity	N/I	N/I	N/I	N/I	N/I
Chen 2010	HOMA	N/I	N/I	N/I	N/I	N/I	N/I
Kan 2004	FInS	N/I	N/I	N/I	N/I	N/I	N/I
Lam 2008	HOMA	6 m walk	N/I	N/I	N/I	N/I	N/I
Wang 2009	N/I	N/I	N/I	N/I	N/I	N/I	N/I
Xiao 2015	N/I	6‐minute walk test, balance	N/I	N/I	N/I	N/I	N/I
Youngwanichsetha 2013	N/I	N/I	N/I	N/I	N/I	N/I	N/I
Zhang 2008	FInS	N/I	N/I	N/I	N/I	N/I	N/I
ND: not defined; N/I: not investigated; HOMA: homeostasis model assessment; FInS: fasting insulin

Appendix 10. Adverse events (I)

	Intervention(s) and comparator(s)	Participants included in analysis (N)	Deaths (N)	Deaths (% of participants)	Participants with at least one adverse event (N)	Participants with at least one adverse event (%)	Participants with at least one severe/serious adverse event (N)	Participants with at least one severe/serious adverse event (%)
Orr 2006	I: Tai Chi	17	0	0	1	5.8	0	0
	C: sham exercise	20	0	0	0	0	0	0
Chen 2010	I: Tai Chi	50	0	0	‐	‐	‐	‐
	C: conventional exercise	44	0	0	‐	‐	‐	‐
Kan 2004	I: Tai Chi	26	0	0	‐	‐	‐	‐
	C: usual exercise	22	0	0	‐	‐	‐	‐
Lam 2008	I: Tai Chi	24	0	0	‐	‐	‐	‐
	C: usual exercise	22	0	0	‐	‐	‐	‐
Wang 2009	I: Tai Chi	34	0	0	‐	‐	‐	‐
	C: control	30	0	0	‐	‐	‐	‐
Xiao 2015	I: Tai Chi	16	0	0	‐	‐	‐	‐
	C: control	16	0	0	‐	‐	‐	‐
Youngwanichsetha 2013	I: Tai Chi	32	0	0	0	0	0	0
	C: usual exercise	32	0	0	0	0	0	0
Zhang 2008	I: Tai Chi	10	0	0	‐	‐	‐	‐
	C: control	10	0	0	‐	‐	‐	‐
‐ denotes not reported C: comparator; I: intervention

Appendix 11. Adverse events (II)

	Intervention(s) and comparator(s)	Participants included in analysis (N)	Participants discontinuing trial due to an adverse event (N)	Participants discontinuing trial due to an adverse event (%)	Participants with at least one hospitalisation (N)	Participants with at least one hospitalisation (%)	Participants with at least one outpatient treatment (N)	Participants with at least one outpatient treatment (%)
Orr 2006	I: Tai Chi	17	1	5.8	0	0	0	0
	C: sham exercise	20	0	0	0	0	0	0
Chen 2010	I: Tai Chi	50	‐	‐	‐	‐	‐	‐
	C: conventional exercise	44	‐	‐	‐	‐	‐	‐
Kan 2004	I: Tai Chi	26	‐	‐	‐	‐	‐	‐
	C: usual exercise	22	‐	‐	‐	‐	‐	‐
Lam 2008	I: Tai Chi	24	‐	‐	‐	‐	‐	‐
	C: usual exercise	22	‐	‐	‐	‐	‐	‐
Wang 2009	I: Tai Chi	34	‐	‐	‐	‐	‐	‐
	C: control	30	‐	‐	‐	‐	‐	‐
Xiao 2015	I: Tai Chi	16	‐	‐	‐	‐	‐	‐
	C: control	16	‐	‐	‐	‐	‐	‐
Youngwanichsetha 2013	I: Tai Chi	32	0	0	0	0	0	0
	C: usual exercise	32	0	0	0	0	0	0
Zhang 2008	I: Tai Chi	10	‐	‐	‐	‐	‐	‐
	C: control	10	‐	‐	‐	‐	‐	‐
"‐" denotes not reported C: comparator; I: intervention

Appendix 12. Adverse events (III)

	Intervention(s) and comparator(s)	Participants included in analysis (N)	Participants with at least one hypoglycaemic episode (N)	Participants with at least one hypoglycaemic episode (%)	Participants with at least one severe/serious hypoglycaemic episode (N)	Participants with at least one severe/serious hypoglycaemic episode (%)	Participants with at least one nocturnal hypoglycaemic episode (N)	Participants with at least one nocturnal hypoglycaemic episode (% participants)
Orr 2006	I: Tai Chi	17	0	0	0	0	0	0
	C: sham exercise	20	0	0	0	0	0	0
Chen 2010	I: Tai Chi	50	‐	‐	‐	‐	‐	‐
	C: conventional exercise	44	‐	‐	‐	‐	‐	‐
Kan 2004	I: Tai Chi	26	‐	‐	‐	‐	‐	‐
	C: usual exercise	22	‐	‐	‐	‐	‐	‐
Lam 2008	I: Tai Chi	24	‐	‐	‐	‐	‐	‐
	C: usual exercise	22	‐	‐	‐	‐	‐	‐
Wang 2009	I: Tai Chi	34	‐	‐	‐	‐	‐	‐
	C: control	30	‐	‐	‐	‐	‐	‐
Xiao 2015	I: Tai Chi	34	‐	‐	‐	‐	‐	‐
	C: control	30	‐	‐	‐	‐	‐	‐
Youngwanichsetha 2013	I: Tai Chi	32	0	0	0	0	0	0
	C: usual exercise	32	0	0	0	0	0	0
Zhang 2008	I: Tai Chi	10	‐	‐	‐	‐	‐	‐
	C: control	10	‐	‐	‐	‐	‐	‐
"‐" denotes not reported C: comparator; I: intervention

Appendix 13. Adverse events (IV)

	Intervention(s) and comparator(s)	Participants included in analysis (N)	Participants with a specific adverse event (description)	Participants with at least one specific adverse events (N)	Participants with at least one specific adverse event (%)
Orr 2006	I: Tai Chi	17	Intolerable secondary to pain and fatigue	1	5.8
	C: sham exercise	20	‐	0	0
Chen 2010	I: Tai Chi	50	‐	‐	‐
	C: conventional exercise	44	‐	‐	‐
Kan 2004	I: Tai Chi	26	‐	‐	‐
	C: usual exercise	22	‐	‐	‐
Lam 2008	I: Tai Chi	24	‐	‐	‐
	C: usual exercise	22	‐	‐	‐
Wang 2009	I: Tai Chi	34	‐	‐	‐
	C: control	30	‐	‐	‐
Xiao 2015	I: Tai Chi	16	‐	‐	‐
	C: control	16	‐	‐	‐
Youngwanichsetha 2013	I: Tai Chi	32	‐	0	0
	C: usual exercise	32	‐	0	0
Zhang 2008	I: Tai Chi	10	‐	‐	‐
	C: control	10	‐	‐	‐
‐ denotes not reported C: comparator; I: intervention

Appendix 14. Survey of authors providing information on included trials

	Date trial author contacted	Date trial author replied	Date trial author was asked for additional information (short summary)	Date trial author provided data (short summary)
Orr 2006	30/07/2014, 02/12/2014	05/08/2014, 03/12/2014	Yes, protocol of the study and detail data for BMI and HbA1c	The HbA1c means and SDs were: Tai chi group: Pre‐treatment: 7.09 ± 0.92 % Tai chi group: Post‐treatment: 6.95 ± 0.86 % Control group: Pre‐treatment: 6.85 ± 0.85 % Control group: Post‐treatment: 7.05 ± 0.80 % BMI results: Tai Chi group: Pre‐treatment: 33.70 ± 4.99 kg/m^2 Tai Chi group: Post‐treatment: 33.80 ± 4.60 kg/m^2 Control group: Pre‐treatment: 30.91 ± 7.17 kg/m^2 Control group: Post‐treatment: 31.40 ± 7.35 kg/m^2
Chen 2010	30/07/2014	No	N/A	N/A
Kan 2004	03/08/2014	No	N/A	N/A
Lam 2008	26/06/2014	27/06/2014	N/A	N/A
Wang 2009	03/08/2014	No	N/A	N/A
Xiao 2015	No	N/A	N/A	N/A
Youngwanichsetha 2013	30/07/2014	No	N/A	N/A
Zhang 2008	30/07/2014	No	N/A	N/A
N/A: not applicable; SD: standard deviation, BMI: body mass index

What's new

Date	Event	Description
24 July 2018	Amended	This protocol was withdrawn by the Editorial Office of the Cochrane Metabolic and Endocrine Disorders Group because finishing the project within adequate deadlines could not be achieved. The review is considered low priority and therefore this review is currently not open for new authors.

Contributions of authors

Junshan Zhou (JZ): search strategy development, data analysis, data interpretation, protocol and review drafting, and future review updates.

Hong Zhang (HZ): trial selection, data extraction, data analysis and review drafting.

Guomei Shi (GS): trial selection, data extraction, data analysis and review drafting.

Lizhen Zhang (LZ): search strategy development, acquiring trial reports, and protocol drafting.

Hao Liu (HL): protocol and review drafting, trial selection, data extraction, and data analysis.

Yewen Qin(YQ): review drafting and English writing.

Jie Yang (JY): search strategy development, data analysis, data interpretation, review and protocol drafting, and future review updates.

Sources of support

Internal sources

• No sources of support supplied

External sources

• Grant (201208008) from Nanjing Science and Technology Committee, China.

  The grant supported this meta‐analysis for staff training and statistical consulting.

Declarations of interest

JZ: none known.

HZ: none known.

GS: none known.

LZ: none known.

HL: none known.

YQ: none known.

JY: none known.

Notes

This protocol was withdrawn by the Editorial Office of the Cochrane Metabolic and Endocrine Disorders Group because finishing the project within adequate deadlines could not be achieved. The review is considered low priority and therefore this review is currently not open for new authors.

Withdrawn from publication for reasons stated in the review