Active mind‐body movement therapies as an adjunct to or in comparison with pulmonary rehabilitation for people with chronic obstructive pulmonary disease

Louis McCusky Gendron; Andre Nyberg; Didier Saey; François Maltais; Yves Lacasse

doi:10.1002/14651858.CD012290.pub2

. 2018 Oct 10;2018(10):CD012290. doi: 10.1002/14651858.CD012290.pub2

Active mind‐body movement therapies as an adjunct to or in comparison with pulmonary rehabilitation for people with chronic obstructive pulmonary disease

Louis McCusky Gendron ¹, Andre Nyberg ¹, Didier Saey ¹, François Maltais ¹, Yves Lacasse ^1,^✉

Editor: Cochrane Airways Group

PMCID: PMC6517162 PMID: 30306545

Abstract

Background

Active mind‐body movement therapies (AMBMTs), including but not limited to yoga, tai chi, and qigong, have been applied as exercise modalities for people with chronic obstructive pulmonary disease (COPD). AMBMT strategies have been found to be more effective than usual care; however, whether AMBMT is inferior, equivalent, or superior to pulmonary rehabilitation (PR) in people with COPD remains to be determined.

Objectives

To assess the effects of AMBMTs compared with, or in addition to, PR in the management of COPD.

Search methods

We searched the Cochrane Airways Group Specialised Register of trials and major Chinese databases, as well as trial registries from inception to July 2017. In addition, we searched references of primary studies and review articles. We updated this search in July 2018 but have not yet incorporated these results.

Selection criteria

We included (1) randomised controlled trials (RCTs) comparing AMBMT (i.e. controlled breathing and/or focused meditation/attention interventions for which patients must actively move their joints and muscles for at least four weeks with no minimum intervention frequency) versus PR (any inpatient or outpatient, community‐based or home‐based rehabilitation programme lasting at least four weeks, with no minimum intervention frequency, that included conventional exercise training with or without education or psychological support) and (2) RCTs comparing AMBMT + PR versus PR alone in people with COPD. Two independent review authors screened and selected studies for inclusion.

Data collection and analysis

Two review authors independently selected trials for inclusion, extracted outcome data, and assessed risk of bias. We contacted study authors if necessary to ask them to provide missing data. We calculated mean differences (MDs) using a random‐effects model.

Main results

We included in the meta‐analysis 10 studies with 762 participants across one or more comparisons. The sample size of included studies ranged from 11 to 206 participants. Nine out of 10 studies involving all levels of COPD severity were conducted in China with adults from 55 to 88 years of age, a higher proportion of whom were male (78%). Nine out of 10 studies provided tai chi and/or qigong programmes as AMBMT, and one study provided yoga. Overall, the term 'PR' has been uncritically applied in the vast majority of studies, which limits comparison of AMBMT and PR. For example, eight out of 10 studies considered walking training as equal to PR and used this as conventional exercise training within PR. Overall study quality for main comparisons was moderate to very low mainly owing to imprecision, indirectness (exercise component inconsistent with recommendations), and risk of bias issues. The primary outcomes for our review were quality of life, dyspnoea, and serious adverse events.

When researchers compared AMBMT versus PR alone (mainly unstructured walking training), statistically significant improvements in disease‐specific quality of life (QoL) (St. George's Respiratory Questionnaire (SGRQ) total score) favoured AMBMT: mean difference (MD) ‐5.83, 95% confidence interval (CI) ‐8.75 to ‐2.92; three trials; 249 participants; low‐quality evidence. The common effect size, but not the 95% CI around the pooled treatment effect, exceeded the minimal clinically important difference (MCID) of minus four. The COPD Assessment Test (CAT) also revealed statistically significant improvements favouring AMBMT over PR, with scores exceeding the MCID of three, with an MD of 6.58 units (95% CI ‐9.16 to – 4.00 units; one trial; 74 participants; low‐quality evidence). Results show no between‐group differences with regard to dyspnoea measured by the modified Medical Research Council Scale (MD 0.00 units, 95% CI ‐0.37 to 0.37; two trials; 127 participants; low‐quality evidence), the Borg Scale (MD 0.44 units, 95% CI ‐0.88 to 0.00; one trial; 139 participants; low‐quality evidence), or the Chronic Respiratory Questionnaire (CRQ) Dyspnoea Scale (MD ‐0.21, 95% CI ‐2.81 to 2.38; one trial; 11 participants; low‐quality evidence). Comparisons of AMBMT versus PR alone did not include assessments of generic quality of life, adverse events, limb muscle function, exacerbations, or adherence.

Comparisons of AMBMT added to PR versus PR alone (mainly unstructured walking training) revealed significant improvements in generic QoL as measured by Short Form (SF)‐36 for both the SF‐36 general health summary score (MD 5.42, 95% CI 3.82 to 7.02; one trial; 80 participants; very low‐quality evidence) and the SF‐36 mental health summary score (MD 3.29, 95% CI 1.45 to 4.95; one trial; 80 participants; very low‐quality evidence). With regard to disease‐specific QoL, investigators noted no significant improvement with addition of AMBMT to PR versus PR alone (SGRQ total score: MD ‐2.57, 95% CI ‐7.76 to 2.62 units; one trial; 192 participants; moderate‐quality evidence; CRQ Dyspnoea Scale score: MD 0.04, 95% CI ‐2.18 to 2.26 units; one trial; 80 participants; very low‐quality evidence). Comparisons of AMBMT + PR versus PR alone did not include assessments of dyspnoea, adverse events, limb muscle function, exacerbations, or adherence.

Authors' conclusions

Given the quality of available evidence, the effects of AMBMT versus PR or of AMBMT added to PR versus PR alone in people with stable COPD remain inconclusive. Evidence of low quality suggests better disease‐specific QoL with AMBMT versus PR in people with stable COPD, and evidence of very low quality suggests no differences in dyspnoea between AMBMT and PR. Evidence of moderate quality shows that AMBMT added to PR does not result in improved disease‐specific QoL, and evidence of very low quality suggests that AMBMT added to PR may lead to better generic QoL versus PR alone. Future studies with adequate descriptions of conventional exercise training (i.e. information on duration, intensity, and progression) delivered by trained professionals with a comprehensive understanding of respiratory physiology, exercise science, and the pathology of COPD are needed before definitive conclusions can be drawn regarding treatment outcomes with AMBMT versus PR or AMBMT added to PR versus PR alone for patients with COPD.

Plain language summary

Active mind‐body movement therapies for chronic obstructive pulmonary disease

Background

Chronic obstructive pulmonary disease (COPD) is a major cause of ill health and is currently the fourth leading cause of death worldwide. COPD describes a chronic lung condition in which shortness of breath, tiredness, and exercise intolerance are typical symptoms. Active mind‐body movement therapies (AMBMTs) consist of mind‐body therapies such as controlled breathing and/or focused meditation/attention interventions whereby participants must actively move their joints and muscles (e.g. yoga, tai chi, qigong). Although different forms of AMBMT may differ in origin, they usually share similar principles: movement/posture, controlled breathing, and focused attention/meditation. AMBMT strategies applied to people with COPD have been found to be more effective than usual care. However the effect of AMBMT as an adjunct to or in direct comparison with pulmonary rehabilitation (PR), the cornerstone of COPD management, remains to be determined; this is the objective of the current review.

Study characteristics

We included 10 studies involving 762 participants who were randomly assigned to receive AMBMT alone or in combination with PR or PR alone (mainly unstructured walking training). The quality of included studies was generally poor.

Key results

Given the quality of available evidence, effects of AMBMT in comparison with PR or of AMBMT added to PR in comparison with PR alone remain inconclusive. One key reason for this is that PR programmes used as comparators had major design flaws, for example, the term 'PR' was uncritically used in the vast majority of studies, and PR was often considered equal to unstructured walking training. This, together with the poor quality of evidence, limits our confidence in the observed effects. Available evidence suggests that when AMBMT was compared to PR alone, larger improvements in disease‐specific quality of life were observed with AMBMT, although AMBMT was not superior to PR with regard to dyspnoea (breathlessness). AMBMT added to PR resulted in large improvement in generic quality of life when compared with PR alone, although the addition of AMBMT to PR did not lead to further improvements in disease‐specific quality of life. However, before definitive conclusions can be drawn, future research studies comparing AMBMT to PR are needed, and these should follow current PR guidelines for designing comparator interventions, preferably delivered by properly trained professionals with a comprehensive understanding of respiratory physiology, exercise science, and the pathology of COPD.

Summary of findings

Background

Description of the condition

Chronic obstructive pulmonary disease (COPD) is characterised by chronic airflow limitation. It is considered a systemic and multi‐component disease with cough, sputum production, and chronic, progressive dyspnoea (breathlessness) and exercise intolerance as typical symptoms (Butcher 2012; GOLD 2018). The number of people living with COPD is steadily increasing, and COPD is a major cause of morbidity and mortality; it is projected to become the third leading cause of death worldwide in 2030 (WHO 2013). The international prevalence of Global Initiative on Obstructive Lung Disease (GOLD) stage 2 and higher COPD has been reported to be 10.1% and is believed to be increasing, which can be explained to some extent by continued exposure to risk factors in combination with ageing of the world's population (Buist 2007;GOLD 2018).

COPD is usually caused by long‐term exposure to lung irritants that cause damage to the lungs and airways. The most prevalent irritant known to cause COPD is cigarette smoke, but others such as second‐hand smoke, air pollution, chemicals, and dust particles from workplace environments have been involved.

Individuals who present with symptoms of dyspnoea, chronic cough, and/or sputum production should be considered for a diagnosis of COPD. To diagnose the disease, a spirometry is mandatory to confirm the presence of airflow limitation. According to GOLD, a ratio of forced expiratory volume in one second (FEV₁)/forced vital capacity (FVC) < 0.70 indicates airflow limitation. If consistent clinical features are noted, persistent airflow limitation confirms the diagnosis of COPD. COPD assessment should determine (1) the severity of airflow limitation, (2) the impact of the disease on the patient's health status, and (3) the risk of future exacerbations, to guide therapy. Coexistent diseases should be actively looked for and treated appropriately if present (GOLD 2018).

Exercise intolerance is considered one of the key features of COPD (Butcher 2012; Houchen 2009). Ventilatory limitation, hyperinflation, gas exchange abnormalities, and cardiac and limb muscle dysfunction, alone or in combination, have been proposed as factors contributing to exercise limitation in COPD (Spruit 2013). As the disease progresses, patients experience recurrent exacerbations, increased dyspnoea, and more frequent hospital visits, collectively leading to poor quality of life (GOLD 2018; Miravitlles 2007).

Management of COPD includes identification and reduction of exposure to risk factors, as well as pharmacological and non‐pharmacological treatments. With regard to the latter, pulmonary rehabilitation is recognised as a core component of COPD management (Vogelmeier 2017).

Description of the intervention

Active mind‐body movement therapies (AMBMTs)

Mind‐body practices, as defined by the National Center for Complementary and Integrative Health, from the National Institutes of Health, are classified amongst the most popular complementary health approaches used by adults in the general population (Harris 2012). These practices focus on the interaction between mind and body by using internal awareness, anatomical alignment, and deep breathing to improve individual wellness. Mind‐body practices can be categorised into three different subgroups (adapted from Lee 2014). First, the intervention may be motionless (e.g. meditation, hypnosis, relaxation therapy). Second, the intervention may involve passive movement. In this case, motion is transmitted to parts of the body by an external force (e.g. massage therapy, spinal manipulation). Third, the intervention may include active mind‐body movement therapies (AMBMTs), whereby participants must move their joints and muscles (e.g. yoga, tai chi, qigong, Pilates). Although different forms of AMBMT may differ in origin, they usually share similar principles: movement/posture, controlled breathing, and focused attention/meditation (Table 3). The intensity of these interventions has rarely been reported. However, studies have shown that some AMBMTs are exercises of light to moderate intensity (Hagins 2007; Lan 2008). Participants may adjust the difficulty of such exercises to their own desired level of conditioning. A fundamental component of AMBMTs, which sets them apart from conventional exercise training, is mindful awareness. Participants learn to recognise tension or strain in different areas of the body, or aspects of breathing that feel arduous or effortless, while performing a posture or movement. This mindful awareness may complement and improve other components such as posture and relaxation. Preliminary evidence suggests that tai chi and qigong may improve exercise capacity, dyspnoea, quality of life, and lung function compared to conventional exercise or usual care in patients with COPD (Ng 2014a; Yan 2013).

1. Description of different AMBMTs.

Intervention	Description	Reference
Tai chi	Originally developed as a martial art in China around the 16th century (origins unclear), training emphasises focusing attention, co‐ordinating breathing with movements, and aligning proper posture in a rhythmical sequence and at a constant rate	Ng 2014a
Qigong	Originating from Traditional Chinese Medicine (TCM) and refined for over 5000 years, qigong exercises consist of postures, breathing techniques, and meditation, all designed to enhance Qi function (concept of force or energy in TCM) through attainment of deeply focused and relaxed states	Jahnke 2010
Yoga	Practised in India as early as 3000 before common era and can be simplified as a combination of postures (Asanas), breathing techniques (Pranayamas), and sustained concentration (Dhyana)	Taneja 2014
Pilates	Created in the United States in the early 20th century and matured as a combination of structured physical exercises focusing on concentration, control, centreing, flowing movement, precision, and breathing	Bullo 2015

Active mind‐body movement therapies (AMBMTs) vs pulmonary rehabilitation (PR) for chronic obstructive pulmonary disease
Patient or population: patients with chronic obstructive pulmonary disease Settings: any setting Intervention: AMBMT Control: PR
Outcomes	*Illustrative comparative effects (95% CI)**		No. of participants  (studies)	Quality of the evidence  (GRADE)
	Response on control	Treatment effect
	PR	AMBMT vs PR
QoL ‐ Change in SGRQ total score Scale 0 to 100. Lower is better and 4 units is a clinically important difference	Mean change = ‐1.92 units	Mean change in SGRQ total score in the AMBMT group was 5.83 units lower (95% CI 8.75 to 2.92 to lower)	249  (3 RCTs)	⊕⊕⊝⊝  low^a
QoL ‐ Change in CAT score Scale 0 to 40. Lower is better and 3 units is a clinically important difference	Mean change = 2.74 units	Mean change in CAT score in the AMBMT group was 6.58 units lower (95% CI 9.16 to 4.00 lower)	74  (1 RCT)	⊕⊕⊝⊝  low^b
Dyspnoea ‐ change in CRQ Dyspnoea CRQ Dyspnoea. Scale 1 to 7. Higher is better and 0.5 unit is a clinically important difference	Mean change = 0.24 units	Mean change in CRQ score in the AMBMT group was 0.21 units lower (95% CI 2.81 lower to 2.38 higher)	11 (1 RCT)	⊕⊝⊝⊝  very low^c
Dyspnoea ‐ change in mMRC Scale 0 to 4. Lower is better and 1 unit is a clinically important difference	Mean change = ‐0.75 units	Mean change in mMRC score in the AMBMT group was 0.00 units lower (95% CI 0.37 lower to 0.37 higher)	127 (2 RCTs)	⊕⊝⊝⊝  very low^d
Dyspnoea ‐ change in Borg Scale Borg CR‐10. Scale 0 to 10. Lower is better and 1 unit is a clinically important difference	Mean change = 0.32 units	Mean change in Borg Scale score in the AMBMT group was 0.44 units lower (95% CI 0.88 to 0.00 lower)	139 (1 RCT)	⊕⊝⊝⊝  very low^e
Adverse events^f	‐	‐	490 (8 RCTs)	‐
*The basis for the response on control is the mean control group response across studies  AMBMT: active mind‐body movement therapy; CAT: COPD Assessment Test; CI: confidence interval; CRQ: Chronic Respiratory Questionnaire; MD: mean difference; mMRC: modified Medical Research Council Scale; PR: pulmonary rehabilitation; QoL: quality of life; RCT: randomised controlled trial; SGRQ: St. George's Respiratory Questionnaire
GRADE Working Group grades of evidence  High quality: further research is very unlikely to change our confidence in the estimate of effect  Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate  Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate  Very low quality: we are very uncertain about the estimate

Database	Frequency of search
CENTRAL (the Cochrane Library)	Monthly
MEDLINE (Ovid)	Weekly
Embase (Ovid)	Weekly
PsycINFO (Ovid)	Monthly
CINAHL (EBSCO)	Monthly
AMED (EBSCO)	Monthly

Conference	Years searched
American Academy of Allergy, Asthma and Immunology (AAAAI)	2001 onwards
American Thoracic Society (ATS)	2001 onwards
Asia Pacific Society of Respirology (APSR)	2004 onwards
British Thoracic Society Winter Meeting (BTS)	2000 onwards
Chest Meeting	2003 onwards
European Respiratory Society (ERS)	1992, 1994, 2000 onwards
International Primary Care Respiratory Group Congress (IPCRG)	2002 onwards
Thoracic Society of Australia and New Zealand (TSANZ)	1999 onwards

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 SGRQ	3		Mean Difference (IV, Random, 95% CI)	Subtotals only
1.1 Total	3	249	Mean Difference (IV, Random, 95% CI)	‐5.83 [‐8.75, ‐2.92]
1.2 Activity	2	213	Mean Difference (IV, Random, 95% CI)	‐8.96 [‐13.04, ‐4.89]
1.3 Impact	2	213	Mean Difference (IV, Random, 95% CI)	‐3.75 [‐6.78, ‐0.73]
1.4 Symptoms	2	213	Mean Difference (IV, Random, 95% CI)	‐2.22 [‐5.66, 1.23]
2 CAT	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
3 CRQ	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
3.1 Dyspnoea	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
3.2 Fatigue	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
3.3 Mastery	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
3.4 Emotion	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
4 Zhongshan COPD Questionnaire	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
4.1 ADL	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
4.2 Social participation	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
4.3 Depression	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
4.4 Anxiety	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
5 Borg CR10 ‐ Dyspnoea	1		Mean Difference (IV, Fixed, 95% CI)	Totals not selected
6 mMRC	2	127	Mean Difference (IV, Random, 95% CI)	0.00 [‐0.37, 0.37]
7 6MWD	7	436	Mean Difference (IV, Random, 95% CI)	3.05 [‐9.97, 16.07]
7.1 Tai chi	3	253	Mean Difference (IV, Random, 95% CI)	19.22 [‐1.86, 40.30]
7.2 Qigong	3	172	Mean Difference (IV, Random, 95% CI)	‐0.16 [‐10.11, 9.80]
7.3 Yoga	1	11	Mean Difference (IV, Random, 95% CI)	‐69.3 [‐117.73, ‐20.87]
8 ICE	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
9 CWR	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
10 FEV₁	6	318	Mean Difference (IV, Random, 95% CI)	0.07 [‐0.01, 0.15]
10.1 Tai chi	2	179	Mean Difference (IV, Random, 95% CI)	0.10 [‐0.04, 0.24]
10.2 Qigong	3	128	Mean Difference (IV, Random, 95% CI)	0.07 [‐0.08, 0.23]
10.3 Yoga	1	11	Mean Difference (IV, Random, 95% CI)	‐0.01 [‐0.22, 0.19]
11 FVC	2	150	Mean Difference (IV, Random, 95% CI)	‐0.02 [‐0.21, 0.17]
12 FEV₁/FVC	5	260	Mean Difference (IV, Random, 95% CI)	3.84 [0.68, 7.01]
12.1 Tai chi	1	74	Mean Difference (IV, Random, 95% CI)	6.02 [3.72, 8.32]
12.2 Qigong	3	175	Mean Difference (IV, Random, 95% CI)	3.51 [‐1.24, 8.26]
12.3 Yoga	1	11	Mean Difference (IV, Random, 95% CI)	1.90 [‐3.41, 7.21]
13 FEV₁ predicted	6	325	Mean Difference (IV, Random, 95% CI)	3.95 [1.77, 6.13]
13.1 Tai chi	2	114	Mean Difference (IV, Random, 95% CI)	4.97 [2.16, 7.79]
13.2 Qigong	4	211	Mean Difference (IV, Random, 95% CI)	2.42 [‐1.07, 5.91]
14 AECOPD	1		Mean Difference (IV, Random, 95% CI)	Totals not selected

Methods	Design Single‐blind RCT; COPD patients were randomly assigned to 3 groups: tai chi/qigong group, exercise group, and control group Three‐month intervention programme and 6‐month follow‐up
Participants	Setting Participants were recruited from 5 general outpatient clinics in Hong Kong, China Participants 206 participants were randomly assigned to groups: tai chi/qigong (TCQ) group (n = 70), exercise group (n = 69), and control group (n = 67) Ages ranged from 55 to 88, with a mean age of 73 187 participants (91%) were male; 19 (9%) were female Mean duration of COPD was 11 years; 43% were at a severe stage of COPD, 42% at a moderate stage, and 16% at a mild stage Inclusion criteria Clinical diagnosis of COPD as defined by the AmericanThoracic Society Ability to walk independently Exclusion criteria Inability to walk independently Severe sensory or cognitive impairment Symptomatic ischaemic heart disease Practice in TCQ within a year before commencement of the study Baseline imbalances Gender and exacerbation rates
Interventions	AMBMT Participants completed 60‐minute TCQ practice sessions twice a week for 3 months. The TCQ intervention routine consisted of 13 movements of Breathing Regulating TCQ (BRTCQ), which were selected and modified from the 18 movements of Taiji Qigong produced by the Department of Health. The 13 modified forms of BRTCQ were designed for easy learning and for mastery in a shorter period. The 13 BRTCQ forms were referred to 2 TCQ experts to ensure their validity and feasibility for use with COPD clients. The TCQ class was led by a qualified TCQ master. Participants were required to co‐ordinate their breathing with prescribed movements. They were also advised to practice TCQ exercises for an hour every day, apart from the 2 TCQ sessions. A DVD and TCQ pictures were given to each participant to facilitate daily self‐practice. A diary was given to participants, so they could record the frequency of their self‐practice. PR Participants were taught pursed‐lip breathing and diaphragmatic breathing, co‐ordinated with self‐paced walking as a physical exercise. Expiration time was twice the inspiration time to slow down the breathing rate. Participants were instructed to inhale through the nose, push the stomach out, then exhale through pursed lips slowly, while pulling the stomach in. Return demonstrations of breathing techniques were performed by participants to ensure proper practice. Participants were advised to perform breathing and walking exercises for 1 hour, which could be split into 2 or 3 sessions to prevent fatigue. Leaflets with pictures and instructions were given to participants to facilitate daily self‐practice. A diary was also given to each participant for recording the frequency of self‐ practice sessions. The breathing techniques of participants were re‐assessed at the 6‐week and 3‐month time points to ensure that proper skills were maintained. The breathing component in this group was different from that in the TCQ group, as the TCQ group did not require pursed‐lip breathing and placed no restriction on time for breathing in and out. Other group (not of interest for this review) Participants were advised to maintain routine activities Time points measured Baseline, week 6, 3 months, 6‐month follow‐up
Outcomes	Hong Kong Chinese version of the SGRQ, unit Multi‐dimensional Scale of Perceived Social Support ‐ Chinese version (MSPSS‐C), unit Pulmonary function: spirometry (FEV₁, L; FVC, L) 6MWD, m Self‐rated dyspnoea and fatigue scale (modified Borg Scale (0 to 10), unit) Oxygen saturation level (SaO₂), % Compliance rate, % TQG attendance rate, % TQG skill performance (1 to 3, unit) TQG programme satisfaction (1 to 5, unit)
Notes	Funding source Health and Health Services Research Fund (HHSRF:06070201) Lee Hysan Foundation Research Grant and Endowment Fund Research Grant (CA11125)
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random allocation done via a randomiser software
Allocation concealment (selection bias)	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Participants not blinded to the interventions
Blinding of outcome assessment (detection bias)   All outcomes	High risk	Research assistants for data collection were blinded to the study minimise researcher bias; however, quality of life was assessed by participants who were not blinded to the interventions
Incomplete outcome data (attrition bias)   All outcomes	High risk	To preserve the value of randomisation, an intention‐to‐treat analysis was applied. Data from last observation were carried forward for withdrawals; attrition rate was 23%
Selective reporting (reporting bias)	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Other bias	High risk	Confounding bias: unbalanced exacerbation rates at baseline

Methods	Design Patients with COPD were randomly divided into 3 groups: tai chi group, exercise group, and control group 12‐Week intervention programme
Participants	Setting Participants were recruited in Qiaodong District of Zhangjiakou, Hebei, China Participants 112 participants were randomly assigned to groups: Taijiquan (TJQ) group (n = 36), exercise group (n = 38), and control group (n = 38) Age ranged from 60 to 79 years 70 participants (63%) were male; 42 (37%) were female 71% were at a moderate stage of COPD; 29% were at a mild stage Inclusion criteria Clinical diagnosis of COPD as defined by GOLD Lack of engagement in tai chi practice for 5 years before study commencement Exclusion criteria Mental disorders Severe perceptual disorders Inability to walk independently Symptomatic ischaemic heart disease Poor compliance Dropout due to increased disease severity during the intervention period or for another reason
Interventions	AMBMT Participants followed supervised 60‐minute tai chi sessions twice a week plus daily self‐practice (60 minutes with instructional tai chi CD and log book) for a total of 3 months. The tai chi intervention consisted of a 24‐form Yang style routine with breathing co‐ordination. PR Participants were instructed to perform daily 30 minutes of self‐paced walking sessions and 60 minutes of pursed‐lip breathing and diaphragmatic breathing. Participants could split their training into 2 or 3 sessions to prevent fatigue. Leaflets with instructions were given to participants to facilitate daily self‐practice. A diary was given to each participant for recording self‐ practice sessions and was monitored at 6‐week and 3‐month time periods. Other group (not of interest for this review) Participants were advised to maintain routine activities Time points measured Baseline, week 6, 3 months
Outcomes	BODE Index, unit Pulmonary function: spirometry (% pred FEV₁, %; ratio FEV₁/FVC, %; MVV, L) COPD Assessment Test (CAT), unit Oxygen saturation level (SaO₂), % Blood gas analysis (pH, unit; PaCO₂, mmHg; PaO₂, mmHg) SGRQ, unit mMRC, unit BMI, kg/m² 6MWD, m Self‐rated dyspnoea and fatigue scale (modified Borg Scale (0 to 10)), unit
Notes	Funding source Not mentioned
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random number table
Allocation concealment (selection bias)	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Participants not blinded to interventions
Blinding of outcome assessment (detection bias)   All outcomes	High risk	Quality of life assessed by participants who were not blinded to the interventions
Incomplete outcome data (attrition bias)   All outcomes	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Selective reporting (reporting bias)	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Other bias	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'

Methods	Design Single‐blind, RCT design Patients with COPD were randomly allocated to Health Qigong Group (HQG), conventional PR group, or medical treatment group
Participants	Setting Patients were recruited from settings under the care of respiratory specialists of Jiangsu Province Hospitals, China, from October 2008 to October 2010 Participants A total of 132 patients with confirmed diagnosis of COPD were randomly assigned to groups: HQG group (n = 60), PR group (n = 36), and control group (n = 36) Average age of all participants was 62 years 77% were male 48% were GOLD stage 1, and 52% were GOLD stage 2 Inclusion criteria Severity level at GOLD stages 1 and 2 Exclusion criteria Serious comorbidities (e.g. pulmonary tuberculosis, emphysema, congestive heart failure)
Interventions	AMBMT The HQG group received 1 week of HQG training under the supervision of professional HQG coaches, and then were encouraged to participate in a peer‐led weekly practice group thrice a week, lasting 1 hour each time, for 6 months. The adopted HQG protocol was developed by Professor Hong‐Zhu Jin and his team at the Jiangsu Province Hospitals, and development was guided by the following 2 principles. First, it should be based on routines selected from the 4 national standardised HQG protocols: Tendon Changing Classic or Yijinjing (易筋經), Frolics of Five Animals or Wuqinxi (五禽戲), The Art of Expiration in Producing Six Different Sounds or Liuzijue (六字訣), and Eight Excellent Movements or Baduanjin (八段錦). Second, selected routines should have been studied for their effect in fostering lung health according to TCM principles. PR The conventional PR group also received a similar amount of professional input on coaching pursed‐lip breathing and aerobic exercise (e.g. walking, cycling on static bicycle). Participants were then encouraged to participate in peer‐led weekly walking and ball game activities thrice a week, 1 hour each time, for 6 months. Other group (not of interest for this review) The medical treatment group received only health education and was advised to continue exercising by themselves before the 6‐month follow‐up Time points measured Baseline data were taken before randomisation, and outcome data were ascertained at 6‐month follow‐up
Outcomes	Pulmonary function: spirometry (% pred FEV₁, %; ratio FEV₁/FVC, % Six‐minute walk test: distance (6MWD, m) Zhongshan COPD Questionnaire for QoL, unit Immune cell factors (TNF‐α, ng/mL; IL‐8, ng/mL; IL‐6, ng/mL) Acute exacerbation of COPD (AECOPD), unit
Notes	Funding source Not mentioned
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Participant allocation list was drawn based on the random order of a block
Allocation concealment (selection bias)	Unclear risk	All outcome assessors were blinded to each participant's allocated group as well as to objectives of the study to minimise bias. However, no information on how allocation was kept concealed
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Participants were not blinded to the interventions
Blinding of outcome assessment (detection bias)   All outcomes	High risk	Research assistants for data collection were blind to the study to minimise researcher bias; however, quality of life was assessed by participants who were not blinded to the interventions
Incomplete outcome data (attrition bias)   All outcomes	Low risk	To preserve the value of randomisation, an intention‐to‐treat analysis was applied in calculating missing values at other time points. In the case of withdrawals, data were carried forward; attrition rate was 11%
Selective reporting (reporting bias)	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Other bias	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'

Methods	Design 80 patients with COPD receiving conventional PR programme were randomised to the Health Qigong (HQG) intervention group (n = 40) and the control exercise group (n = 40)
Participants	Setting The study was undertaken in a respiratory care hospital that has provided ambulatory rehabilitation programmes for patients with COPD since the early 1990s, in Hong Kong, China Participants A total of 80 patients with COPD were randomly assigned to groups: HQG + PR group (n = 40) and PR group (n = 40) Average age of all participants was 73 years 87% were male Inclusion criteria Diagnosis of COPD confirmed by both medical history and air flow limitation (i.e. ratio of forced expiratory volume in the first second (FEV₁) to forced vital capacity was less than 70%), as reflected by spirometric readings measured after post‐bronchodilator therapy and according to the American Thoracic Society standard Medically stable as interpreted by no hospital admissions for chest problems in the month before PR Cessation of smoking for at least 6 months No other disabling diseases (e.g. stroke, Parkinson's) Exclusion criteria Prior history of practising qigong
Interventions	PR + AMBMT Each patient received four 45‐minute training sessions on HQG led by a trained therapist at the ninth, tenth, eleventh, and twelfth sessions of the PR programme, together with a home learning package in the form of audiovisual materials. The specific form of HQG chosen was Baduanjin. It consisted of 8 distinct movement routines, with each movement routine repeated 6 times. The whole protocol usually took 12 to 15 minutes to complete at a usual pace. To address the special needs of patients with COPD, an expert panel review was conducted to assess potential clinical effectiveness. Before the RCT was begun, a field test was conducted to study safety in its application to patients with COPD. To maximise the potential benefits of training, the minimal dosage of the training protocol that experts advised was practice at least 1 time per day at least 4 times a week until 6‐month follow‐up. To keep a record of their own practice, patients were issued a daily log. PR Each patient received the same number and duration of additional training sessions, reinforcing the breathing (including pursed‐lip and co‐ordinated breathing) and walking exercises of the conventional programme to make it comparable to the treatment group in terms of additional staff attention given to participants. In addition, another set of audiovisual training material on corresponding aspects was issued to all control participants, who were also advised to keep walking daily for not less than 30 minutes until 6‐month follow‐up. Time points measured Baseline and 6‐month follow‐up
Outcomes	SF‐36 general health subscale, unit SF‐36 mental health subscale, unit CCRQ, unit 6MWD, m MFTE, unit Compliance rate, % Attrition rate, %
Notes	Funding source Work was supported by the Area of Excellence Grant (A102), Department of Rehabilitation Sciences of The Hong Kong Polytechnic University
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random numbers
Allocation concealment (selection bias)	High risk	The occupational therapist opened the list file and assigned participants to treatment or control group accordingly
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Participants were not blinded to the interventions
Blinding of outcome assessment (detection bias)   All outcomes	High risk	Outcomes assessments were administered by blinded assessors who were ignorant of the treatment arm to which each participant was assigned; however, quality of life was assessed by participants who were not blinded to the interventions
Incomplete outcome data (attrition bias)   All outcomes	High risk	Analysis was first conducted via the ITT approach, which included all randomised participants classified according to their randomisation and irrespective of their completion of treatment and follow‐up assessments. Missing values at discharge, 3‐month follow‐up, and 6‐month follow‐up were imputed via the "last observation carried forward" method; however, attrition rate was 35%
Selective reporting (reporting bias)	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Other bias	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'

Methods	Design Single‐blind randomised controlled study conducted to compare self‐efficacy and quality of life parameters of patients with COPD who underwent pulmonary rehabilitation with or without incorporation of tai chi elements in the exercise component
Participants	Setting Study conducted from March 2011 to May 2012, with a total of 192 participants recruited from 4 general outpatient clinics in Hong Kong, China Participants A total of 192 patients with clinical diagnosis of COPD were randomly assigned to groups: PR + AMBMT group (n = 94) and PR group (n = 98) Average age of all participants was 74 years 91% were male 20% were GOLD stage 1, 41% GOLD stage 2, 30% GOLD stage 3, and 9% GOLD stage 4 Inclusion criteria Clinical diagnosis of COPD as defined by American Thoracic Society (ATS) and Global Initiative for Chronic Obstructive Lung Disease (2010 GOLD) Medical Research Council (MRC) dyspnoea score ≥ 2 on the 1 to 5 scale version Decreased forced expiratory ratios (FEV₁/FVC) < 70% poorly reversible with bronchodilators Exclusion criteria Poor mobility Severe sensory or cognitive impairment Severe comorbidities including acute myocardial infarction in 6 months before the programme Tai chi practised within a year before study commencement
Interventions	PR + AMBMT Both groups received rehabilitation consisting of 2 sessions per week for 6 weeks with totally identical content, except that tai chi exercises were added for tai chi group. Data collection was performed at baseline and at 2 and 6 months post intervention. Each session would last for 1 hour and 20 minutes, with 6 to 10 participants per session. Upon entering the intervention protocol, all participants in both groups were required to participate in a 30‐minute educational session conducted by a physiotherapist to improve their knowledge and skills in COPD management. A COPD booklet and home diary were provided as supplementary materials. In the TC group, the formal pulmonary rehabilitation programme was identical to that in the PR group, except that 15 minutes of tai chi exercises was substituted for 15 minutes of relaxation exercise. Five forms of Sun Style tai chi were taught by a tai chi accredited physiotherapist for ease of mastery. The TC group was also instructed to perform 1 hour of daily unsupervised home exercises identical to those performed by the PR group, except for 15 minutes of tai chi replacing the relaxation exercises for 5 to 7 days per week. Both groups were instructed to document their home exercises in their compliance diary. PR Each session would last for 1 hour and 20 minutes with 6 to 10 participants per session. Upon entering the intervention protocol, all participants in both groups were required to participate in the 30‐minute educational session conducted by the physiotherapist to improve their knowledge and skills in COPD management. A COPD booklet and home diary were provided as supplementary materials. The formal pulmonary rehabilitation programme in the PR group would consist of warm‐up and cool‐down exercises and aerobic exercises. Patients performed 5 minutes of warm‐up exercises, then 2 aerobic activities including treadmill and lower limb ergometry exercises, each lasting 20 minutes. A 15‐minute rest period was given between exercises. After aerobic exercises, a 5‐minute cool‐down period was followed by 15 minutes of relaxation exercise conducted before participants completed their session. The PR group was instructed to perform 1 hour of daily unsupervised home exercises consisting of a 5‐minute warm‐up, 5 minutes of Thera‐Band exercises, 30 minutes of aerobic exercises, 5 minutes of cool‐down, and 15 minutes of relaxation exercises for 5 to 7 days a week. Thera‐Band exercises, which utilised a rubber‐based resistance band known to optimise muscle strengthening, were also taught as home exercises for both groups. Both groups were instructed to document home exercises in their compliance diary.
Outcomes	Hong Kong Chinese version of the SGRQ, unit Pulmonary function: spirometry (FEV₁, L; FVC, L; % pred FEV₁, %) 6MWD, m Self‐rated dyspnoea scale (modified Borg Scale (0 to 10)), unit SaO₂, % COPD self‐efficacy scale (COPD‐CSES), unit Self‐efficacy for managing shortness of breath (SEMSOB), unit Compliance rate, % Attendance rate, % Time points measured Outcome measures were collected at baseline (T1) and at 2 months (T2) and 6 months (T3) post intervention
Notes	Funding source Health Services Research Fund
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random number sequences
Allocation concealment (selection bias)	Low risk	The same research assistant would conceal the allocation sequence using sequentially numbered envelopes. Each participant would receive 1 envelope that would be opened for allocation to assigned grouping by a different research assistant upon recruitment
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Participants were not blinded to the interventions
Blinding of outcome assessment (detection bias)   All outcomes	High risk	To minimise researcher bias, research assistants for data collection were blinded to the study; however, quality of life was assessed by participants, who were not blinded to the interventions
Incomplete outcome data (attrition bias)   All outcomes	High risk	Intention‐to‐treat approach was used for data analysis. For participants lost to follow‐up, the last known data were carried forward to replace missing values; however, attrition rate was 28%
Selective reporting (reporting bias)	Low risk	No difference between results of registered study (NCT01259245) and outcomes reported in the manuscript
Other bias	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'

Methods	Design Single‐blind randomised controlled trial Participants were assigned randomly to 2 groups: the control group and the tai chi group, in a 1:1 allocation ratio
Participants	Setting Patients with COPD were recruited from January 2011 to June 2011, from outpatient clinics of the Department of Respiratory Medicine, Xiangya Hospital, Central South University, China Participants A total of 40 patients with clinical diagnosis of COPD were randomly assigned to groups: tai chi group (n = 20) and control group (n = 20) Average age of all participants was 60 years 91% were male Inclusion criteria Clinical diagnosis of moderate to severe COPD FEV₁ < 65% of predicted, and ratio of FEV₁ to forced vital capacity < 0.70 45 years of age or older Exclusion criteria COPD exacerbation that required systemic steroids, antibiotics, a visit to the ER, or hospitalisation during the month before the programme Any planned major pulmonary intervention (e.g. lung volume reduction surgery) in the coming 6 months of the programme Severe peripheral vascular disease and claudication or other physical condition that would preclude a 6MWT Severe cognitive dysfunction (Mini Mental State Examination score ≤ 24)
Interventions	AMBMT Participants in the tai chi group underwent 4 sessions of a supervised tai chi programme and 3 sessions of a home‐based tai chi programme per week, 1 session per day. The entire programme lasted 6 months. Each session consisted of a 10‐minute pre‐exercise warm‐up, followed by 30 minutes of tai chi and 10 minutes of post‐exercise relaxation. The intensity was then adjusted for each patient with COPD according to her/his toleration of the programme. Participants in the tai chi group were trained by a physiotherapist who was an accredited tai chi trainer with extensive experience in chronic lung disease therapy. A tai chi training booklet and DVD developed by an accredited trainer were provided to each participant in the tai chi group for use in home sessions. A workout diary was supplied to all participants to record their practice time for each home session and was routinely checked. Programme adherence for both supervised and at‐home sessions was calculated separately as an average percentage of sessions completed. PR Participants were instructed to walk for 30 minutes daily and were subjected to routine medical care for 6 months after baseline measurement. Routine medical care included, in addition to daily walking, tiotropium 10 mg inhalation and lip breathing for 10 to 15 minutes.
Outcomes	Lung function (FEV₁, L; FEV₁ % of pred, %) Exercise tolerance (6MWD, m) Time points measured Outcome measures were collected at baseline and at 6 months post intervention
Notes	Funding source Self‐funded study with no external support
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	A computer‐generated randomisation algorithm was adopted to generate treatment assignments
Allocation concealment (selection bias)	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Participants were not blinded to the interventions
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Outcomes were assessed by researchers who were unaware of the study design
Incomplete outcome data (attrition bias)   All outcomes	Unclear risk	Attrition rate of 2%
Selective reporting (reporting bias)	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Other bias	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'

Methods	Design The aim of this RCT was to evaluate the effects and feasibility of hatha yoga (HY) vs a conventional training programme (CTP) on functional capacity, lung function, and quality of life in patients with obstructive pulmonary disease
Participants	Setting The study was performed at the Karolinska University Hospital, Stockholm, Sweden, among outpatients Participants Forty patients (24 women, median age 64, age range 40 to 84 years) were randomised to HY or CTP. Out of 36 participants who completed the programme, 25 were given a diagnosis of asthma and 11 COPD (GOLD stages 1 to 3). Of 11 patients with a diagnosis of COPD, 8 were randomised to the HY group and 3 to the CTP group Inclusion criteria Age 35 to 85 years with diagnosis of COPD (mild to severe obstruction) Diagnosis of asthma with FEV₁ % predicted between or equal to 30% and 90% Exclusion criteria Severe neurological, orthopaedic, or rheumatologic injury or disease Inability to walk less than 200 metres Decreased mobility and chronic disease that can affect performance Surgery within 6 months Severe mental disease diagnosis Heart infarction within the last 12 months Change in medications during the last 6 weeks before the programme
Interventions	AMBMT Yoga exercises were based on hatha yoga. Two sessions a week, with sessions extending over 60 to 70 minutes, for 12 weeks, with a total of 24 sessions. Started on chair or meditation chair, in a quiet room without music. Props included yogic blocks and blankets. Exempting questions for the teacher, no talking was permitted during the yoga class. Most participants were novice to yoga. Asanas (eyes open): important instruction throughout class to rest when needed by switching into child pose (balasana), or alternatively to lean forward on the chair with legs apart, and to work at the participant's own capacity. General postural instructions and centreing were given at the beginning of classes. General breathing instructions: extended exhalations, all breathing through nostrils (if possible). Complete deep yogic breathing using the diaphragm extensively during all asanas. Uddiyana bandha (pulling stomach in using pelvic floor muscles first, and then the navel, with light pull in towards the spine during all exhalations) was introduced after approx. 6 weeks. Used complete yogic breathing ‐ 3 levels on inhalation and 3 levels on exhalation. Instructions were given to move hands along with the breathing movement, with inhalation starting at the lower abdominal navel area, then at the upper waist/middle chest and at the upper chest below the clavicles. Exhalation starting at upper chest, then at the middle chest, and finally ending as lower abdominal movement. Vinyasa with back bending tadasana on inhalation and deep utkatasana on exhalation, instruction without arms and then with arms extended above the head, alternatively keeping the hands resting on the shoulders. Utkatasana progressed by sitting all the way down to the chair, and later by squatting all the way down to the floor while letting the heels come up. Ardha Chandrasana (standing side stretch, 1 arm up with the other arm down, the other arm supported on the thigh). Dynamic, then progressed to static holding (approx. 30 seconds on each side) alternatively while sitting on chair if breathing restriction occurred. Trikonasana with chair, foot on floor under chair's seat, and hand on chair's seat or on the back‐rest of the chair. Rotation of head during the final pose. Small sun salutation, dog down and dog up (adho mukha svanasana, urdhva mukha svanasana) dynamic poses with breathing, 4 to 5 times with hands on the seat of the chair. Dog pose with the wall (hands on the wall at shoulder height) was also suggested if no weight on the upper body was possible (approx. 2 minutes). Parsvakonasana with chair with foot under chair's seat and hand on the seat (after 4 to 5 weeks) with emphasis on turning the chest and head. Bharadvajasana, using both sitting and standing variations, with both legs folded to the left and twisted right first. Standing variation (utthita marichyasana), with right side to the wall with right leg up on a chair and with hands used to the wall and twisted to the right (alternative given with block under heel of standing foot), then changing sides. Tiger breathing flow, cat (marjaryasana) and cow, on all fours using extended exhalations. Strong abdominal draw‐in while exhaling. Variation used by some; sitting vertically on a chair using simple back flexion and extension (no weight on arms). Another variation in cat pose consisted of flexion of the wrists on the floor (palms down, thumbs facing out, and palms up, thumbs facing towards each other). Sphinx pose (salamba bhujangasana), walked on underarms to the right, and then to the left. Bhujangasana, resting on abdomen, then arms wide with fingertips facing outwards extending up on inhalation. Setu bhandasana (bridge pose) with 1 leg up, progressed to feet on a chair to achieve inversion. Hip resting on the block with both legs up (viparita dandasana variation), then lowered 1 leg at a time to the floor (stretching hip flexors), then in setu bandha with block. Universal pose: supine twist to the right first with strong abdominal activity and straight spine. Rotation of head opposite the legs to improve neck mobility. Fish pose (matsyasana) supported back bent with yoga block (wood) on thoracic spine behind the heart, to increase mobility in the chest (chest opener), with the option of using a rolled mat or blanket under the chest. Alternative was given with weights on elbows, and alternatively baddha konasana. Yogic breathing exercises (pranayama), sitting on a chair or meditation chair, straight spine, chin slightly lowered (eyes closed, approx. 30 minutes). Asthma mudra was used during some exercises. Metronome was used during the last 5 to 6 weeks to get the timing and awareness of breathing during breathing exercises. Kapalabhati (breath of fire), starting with 10 times worked up to 30 times. Hands on lower abdominals to increase awareness of the abdominals drawing in during exhalation. Slowly first, then trying to speed up the pace. At the end of courses, this involved 20 x 4. Focused on exhalation, with no attention to inhalation, 2 to 3 minutes. Bhastrika, hands on lower abdomen, 20 times. Alternated nostril breathing (nadi shodhana), using nashiki mudra, started exhalation through left nostril, inhaled left, exhaled right, inhaled right, exhaled left. First equal inhaled and exhaled (5 seconds in and 5 seconds out suggested), then tried varied lengths of breathing ratios; 2‐04‐0, 6‐0‐12‐0 with no pauses, with important prolonged exhalation. No pauses after inhalation. Focus on extended exhalations and, if possible, on short pausing after each exhalation, 5 to 10 minutes. Sitkari (inhaled through teeth, exhaled through nose and closed mouth) to be used only during the first 4 to 5 weeks. Bharmari (mmmmm sound on exhalation), humming bee, extended outbreaths ‐ 10 minutes, powerful outbreaths with sound (sometimes with sanmukhi mudra), moved sound to upper back palate and towards third eye. Viloma (Dirghal part breath). Divided inhalations and exhalations into 3 parts, inhalation started at abdomen, same inhalation at middle ribs and last, then at collarbones. Exhalation started at collarbones, middle ribs, abdomen for 2 to 3 minutes (sometimes with prana mudra). Hands moving to the active region of the chest. Sitting at beginning of courses while supined on back at the end. Sukhasana with forward bend, hands on block or chair. Shavasana (body scanning, focus on synchronisation of equal relaxation between left and right sides of the body and letting the weight drop to the floor), sometimes elevated legs with feet or calves on the meditation chair, once during the intervention with weights on thighs. Feet closer than arms, keeping back of shoulders on the floor. Approx. 5 to 8 minutes of home training with this programme was provided on a DVD (in Swedish). The time in each yoga pose was gradually increased, and each pose was held from 5 to 40 seconds; each breathing exercise was performed for a longer duration with fewer pauses towards the end of the intervention. Different variations of poses were gradually introduced (using walls, chairs, and floor). Synchronisation of breathing during exercises was emphasised. Strength was not measured in the HY group. PR Two sessions per week, with sessions extending over 60 to 70 minutes for 12 weeks, including a total of 24 sessions. Working with strength and conditioning equipment (each 2 to 4 sets with 10 to 20 repetitions) and cycling 10 to 15 minutes (intensity ratings of 12 to 14 on Borg 20 Scale, 50 to 60 rpm per minute approx.). Conventional training was performed at a gym (with gym equipment and stationary exercise bikes, adjacent to the yoga room) with background radio/music while 2 physiotherapists coached participants. Participants were allowed to talk to each other and to the physiotherapists, and sometimes to the other participants in the room. Exercises included leg extensions (sitting position), standing arm pull‐backs with thera‐band, sitting leg press, shoulder press, squats to chair with crossed arms in front of chest, seated rowing arm pull‐backs in machine, heel lifts holding back of chair, triceps press in machine, torso rotation in machine, standing biceps curls with free weights, wide sitting on chair and torso twists with stick behind chest, hands resting shoulder height on stick, seated side stretches with extended arm (lower arm resting on hip), shoulder shrugs, seated big upward swimming arm movements with arms above the head. Home programme for pulmonary rehabilitation on a DVD was distributed with the programme on paper. Thera‐bands were supplied for home training. Strength increased during the 12‐week period in the conventional training programme: leg extension by 49%, straight arm pull‐backs by 44.6%, and during leg press by 29.4%, with starting kg/kg increased during the intervention.
Outcomes	*Mixed populations; only data from participants with COPD were used and were obtained by contacting study authors.* Pulmonary function: spirometry (FEV₁, L; FVC, L; FEV₁/FVC,%) 6MWD, m CRQ, unit EQ‐5D, unit Time points measured Outcome measures were collected at baseline and at 12 weeks post intervention
Notes	Funding source None mentioned
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation was performed by an external person. Blank papers were scattered on a table, with each participant's identification code facing down. Then 1 paper was randomly categorised into the HY group, while the next was categorised into the CTP group, and so on
Allocation concealment (selection bias)	Low risk	Randomisation was performed by an external person. Blank papers were scattered on a table, with each participant's identification code facing down. Then 1 paper was randomly categorised into the HY group, while the next was categorised into the CTP group, and so on
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Participants were not blinded to the interventions
Blinding of outcome assessment (detection bias)   All outcomes	High risk	Quality of life was assessed by participants who were not blinded to the interventions
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Intention‐to‐treat approach was used for data analysis. For participants lost to follow‐up, the last known data were carried forward to replace missing values; attrition rate was 10%
Selective reporting (reporting bias)	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Other bias	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'

Methods	Design The aim of this study was to determine the effects of TCM on patients with COPD by randomly dividing patients into 3 groups: TCM group, Western group, and control group
Participants	Setting From April 2008 to March 2009, participants were recruited from Guangzhou's Chinese Medicine Hospital Respiratory Clinic Participants A total of 57 participants were randomly divided into 3 groups: TCM group (n = 19), Western group (n = 19), and control group (n = 19) Average age of all participants was 67 years 76% were male No significant differences between the 3 groups at baseline Inclusion criteria Meeting Western criteria Age 40 or older Stable COPD for 4 or more weeks Meeting Chinese medicine criteria of deficiency of vital energy and phlegm stasis Constitution fitting deficiency of vital energy type, damp (phlegmy) type, or blood stasis type Exclusion criteria History of bronchitis Inability to take part in leg exercises Serious cardiovascular disease: unstable angina pectoris, heart condition not controlled by medication, recent infarction, frequent heart extrasystole Severe high blood pressure History of fainting after exercise Serious liver or kidney function problems Inability to co‐operate because of cognitive or other mental issues
Interventions	AMBMT Baduanjin (8 brocades) light exercises to warm up, mid‐level abdominal breathing and pursed‐lip breathing, then heavier exercise to increase heart rate. Specifically within the 8 brocades: push up to support the sky, left and right to pull the bow, turn the waist to push back, left and right swinging, press the knees, and solidify the waist, horse stance with fists. These exercises used large movements and relatively increased strength, such as push up the sky, which required that the heels leave the ground, with the waist turning to push back, required a full turn of the torso; pull the bow, swinging, solidify the waist and fists all used the horse stance. Speed was adjusted appropriately with the degree of movement and height of the horse stance to achieve a heart rate of 60% to 80% of max. The group first underwent a detailed teaching session and received corrections to their movements, along with instructions on how to regulate their heart rate. They were then given a recorded 8‐brocade video to take home, to help them with their performance. They practised at home 4 times a week for 30 minutes per time According to the constitutional type of the patient and basic diet therapy methods, a diet therapy plan for each patient was determined for (1) deficiency of vital energy type: ginseng black chicken soup, yam mutton porridge, hetao milk drink; (2) Damp (phlegmy) type: lotus seed duck soup, white fulling fungus porridge, yam winter melon drink; and (3) blood stasis type: pseudo‐ginseng stewed chicken soup, peach kernel non‐glutinous rice porridge, lingzhi fungus pseudo‐ginseng hawthorn drink. Note: participants took the 3 potions according to their symptoms, picking as they wished, at least 3 times a week A packet of Chinese medicine (secret recipe) from Peili Pharmaceutical Company, Nanning, was added to 300 mL boiled water, and was drunk when it was fully dissolved, once per day after breakfast PR (Western group) Education on rehabilitation: smoking cessation, respiratory system (anatomy and physiology), symptoms of COPD, breathing training, education in nutrition, standard physiotherapy methods (use of medicine, side effects, oxygen treatment, inhalation treatment, self‐care), how to recognise and take care of exacerbations, how to reduce breathing difficulties, the meaning of pulmonary rehabilitation treatments Home training: stair climbing or walking outside as training, 4 times per week for 30 minutes per time (with 5 minutes warm‐up and cool‐down), to reach the desired heart rate of 60% to 80% of max. Participants were taught how to take their heart rate. They received a phone call once a week to check up
Outcomes	Pulmonary function: spirometry (FEV₁, L; FEV₁/FVC, %; percentage of predicted FEV₁, %) Time points measured Outcome measures were collected at baseline and at 3 months
Notes	Funding source None mentioned
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation via software PEMS 3.1
Allocation concealment (selection bias)	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Participants were not blinded to the interventions
Blinding of outcome assessment (detection bias)   All outcomes	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Incomplete outcome data (attrition bias)   All outcomes	Unclear risk	Attrition rate of 5%
Selective reporting (reporting bias)	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Other bias	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'

Methods	Design Aim was to observe the efficacy of Health Qigong of Wuqinxi in early intervention for patients with stable COPD 78 cases were randomly divided into Wuqinxi group, walking group, and control group
Participants	Setting From July 2004 to April 2008, patients were recruited from the Chinese Medicine Hospital Respiratory Department, Jiangsu, China Participants A total of 78 patients with stable COPD were randomly divided into groups: Wuqinxi group (n = 26), walking group (n = 27), and control group (n = 21) Average age of all participants was 55 years 49% were male No significant differences between the 3 groups at baseline Inclusion criteria Meeting diagnostic criteria for Chinese and Western medicine for COPD Grade I to II in stable period Age 45 to 75 Consent to take part in the experiment Exclusion criteria Confirmed tuberculosis, fungi, cancer, silicosis, irritating gas, allergies, chronic cough, wheezing, and diagnosis of bronchial asthma, asthma, bronchiectasis, congestive heart failure, occlusive bronchioles, bronchitis, diffuse pan‐bronchiolitis Serious concurrent cardiopulmonary dysfunction These conditions combined with cardiovascular, urinary, digestive, hematopoietic, endocrine system, and other serious primary disease and mental illness Pregnancy or lactation
Interventions	AMBMT Patients underwent daily early morning collective practice, led by an instructor of the 2003 State Sports General Administration of China, New Health Qigong Wu Qinxiu (5 animal frolics). From the starting position to regulate breathing, they went through tiger play, deer play, bear play, ape play, and bird play, and then repeated bird play, and finally returned to original qi position. They repeated this sequence 2 to 3 times. Exercise time and intensity: First week of basic training consisted of learning movements and breathing essentials to master techniques. Practised once every day, each time for 45 minutes, so that the practitioner's heart rate reached the target heart rate, which was maintained for longer than 10 minutes. Practice was adjusted in accordance with the different problems of patients; appropriate adjustment of activity levels involved adjusting the height and strenuousness of movements. Continuous practice for 3 months. Target heart rate: 60 years old and above, heart rate = 170 minus age; younger than 60 years old, heart rate = 180 minus age (exercise intensity was based on the Jungmann formula). PR The walking group was treated with the same conventional therapy with the addition of walking exercises. Every morning, walking was done on a flat surface, speed was maintained at 80 to ˜ 120 steps per minute, with duration of at least 45 minutes, so that the practitioner's heart rate reached the target heart rate, which was maintained for longer than 10 minutes. A small number of patients, in the initial phase, rested a number of times, with gradual extension of each session duration, shortening the time and number of rest periods.
Outcomes	Pulmonary function: spirometry (FVC, FEV₁, L; FEV₁/FVC, %; percentage of predicted FEV₁, %) BMI Dyspnoea scale (mMRC) Exercise tolerance (6MWD) Number of hospitalisations due to acute exacerbations Time points measured Outcome measures were collected at baseline and at 3 months
Notes	Funding source None mentioned
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random number digital table
Allocation concealment (selection bias)	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Participants were not blinded to the interventions
Blinding of outcome assessment (detection bias)   All outcomes	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Incomplete outcome data (attrition bias)   All outcomes	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Selective reporting (reporting bias)	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'
Other bias	Unclear risk	Information insufficient to permit judgement of 'low risk' or 'high risk'

Study	Reason for exclusion
Chen 2015	No PR comparator
Cheng 2000	Not an RCT
Cheng 2015	No PR comparator
Deng 2016	No PR comparator
Donesky‐Cuenco 2009	No PR comparator
Dongxing 2011	No PR comparator
Fukuoka 2016	No proper AMBMT
Gao 2016	No PR comparator
Guo 2013	No PR comparator
Li 2012	No PR comparator
Li 2015	No PR comparator
Li 2016	No PR comparator
Liang 2016	No PR comparator
Liu 2013	No PR comparator
Ma 2000	No PR comparator
Ma 2009	No PR comparator
Ni 2017	No PR comparator
Shi 2011	No PR comparator
Tan 2016	No PR comparator
Tsang 2003	Not COPD
Wang 2015	No PR comparator
Wei 2015	No PR comparator
Xu 2010	No PR comparator
Xue 2015	No PR comparator
Yao 2004	No PR comparator
Yuan 2012	No PR comparator
Zhang 2006	No PR comparator
Zhang 2012	No PR comparator
Zhang 2014	No PR comparator
Zhang 2016c	No PR comparator
Zhang 2016d	Not an RCT (Review)
Zhou 2009	No PR comparator
Zhu 2012	No PR comparator
Zhu 2016	No PR comparator

Methods	Not yet assessed
Participants	Not yet assessed
Interventions	Not yet assessed
Outcomes	Not yet assessed
Notes	Not yet assessed

Trial name or title	Effects of Shaolin‐Exercise on Pulmonary Rehabilitation for Elderly Patients With Stable COPD
Methods	Randomised parallel controlled trial
Participants	Countries of recruitment and research settings Beijing Hospital of Traditional Chinese Medicine Affiliated to Capital Medical University Inclusion criteria • Male or female aged 60 to 75 years old • Stable stage of chronic obstructive pulmonary disease under regular treatment with some symptoms • Obstructive ventilation dysfunction of any severity • Willingness to participate in the programme actively Exclusion criteria • Refusal to sign the informed consent form • Communication barriers • Other pulmonary diseases, such as pneumonia, and AECOPD ruled out • Determination by a doctor that individual cannot complete the programme or 6MWT safely • Unstable factors that impeded implementation of rehabilitation, like instability of complications, serious life‐threatening primary disease and mental disease • Already doing pulmonary rehabilitation exercise
Interventions	AMBMT Shaolin exercise (SE) PR Western exercise
Outcomes	SGRQ 6MWT CRQ Lung function TCM symptom score scale CAT
Starting date	2017‐08‐21 to 2019‐12‐31
Contact information	Jinxiu Lin; mayjinxiu@outlook.com
Notes

Trial name or title	A Randomized Controlled Trial of Health Qigong for Chronic Obstructive Pulmonary Disease Rehabilitation
Methods	Random parallel controlled trial
Participants	Inclusion criteria • Stable chronic obstructive pulmonary disease • GOLD stage 1 to 4 Exclusion criteria • Acute myocardial ischaemia • Unstable angina • Severe stenosis or regurgitation • Congenital heart disease • Decompensated heart failure • Uncontrollable arrhythmia • Unstable hypertension or systolic blood pressure ≥ 180 mmHg • Diabetes with serious complications • Post stroke • Rheumatoid or rheumatoid arthritis or musculoskeletal disease or physical disability that seriously affects normal physical activity
Interventions	AMBMT Health qigong group PR Routine exercise rehabilitation
Outcomes	6MWT mMRC CAT SGRQ SF‐36 Lung function
Starting date	2014
Contact information	Chen Xianhai; Chenxianhai18@163.com
Notes

Trial name or title	Tai Chi After Pulmonary Rehabilitation in Patients With COPD: A Randomized Trial (LEAP)
Methods	The main purpose of this study is to determine the feasibility and effects of a 6‐month tai chi exercise programme as compared to a 6‐month group walking programme and standard care for patients with COPD who have recently completed a pulmonary rehabilitation programme
Participants	Inclusion criteria • COPD defined as FEV₁ (forced expiratory volume in 1 second)/FVC (forced vital capacity) < 0.70 • Chest CT evidence of emphysema • Age > 40 years • COPD of any severity as defined by GOLD (Global Obstructive Lung Disease) stage 1, 2, 3, or 4 • Completion of standard pulmonary rehabilitation of at least 8 weeks' duration within 24 weeks before study entry* Defined as attending 65% of programme sessions with a minimum of 10 sessions Exclusion criteria* • COPD exacerbation requiring steroids, antibiotics, ED visit, or hospitalisation within the past 2 weeks unless physician deems patient at baseline • Hypoxaemia on walk test (O₂ sat < 85% on oxygen) • Inability to ambulate due to vascular or other neuromuscular conditions that would preclude a walk test • Clinical signs of unstable cardiovascular disease (i.e. chest pain on walk test) • Severe cognitive dysfunction (documented Mini Mental Status Exam ≤ 24) • Non‐English‐speaking • Current regular practice of tai chi • Current diagnosis of lung cancer or treatment for lung cancer within the past 5 years • Unstable/untreated mental health issue that precludes informed consent or otherwise affects ability to participate in the intervention
Interventions	AMBMT Tai chi 24‐Week tai chi intervention designed for individuals with COPD PR Group walking exercise 24‐Week group walking intervention
Outcomes	Feasibility of tai chi intervention (time frame: 24 weeks) Willingness to participate, adherence, and safety CRQ 6MWT University of California, San Diego, Shortness of Breath Questionnaire Center for Epidemiologic Studies Depression Scale Perceived Stress Scale COPD Self‐Efficacy Scale Multi‐dimensional Scale of Perceived Social Support Pulmonary function Daily exercise activities, step counts
Starting date	August 2012 to June 2018
Contact information	Gloria Y. Yeh; gloria_yeh@hms.harvard.edu
Notes

Trial name or title	Chen‐style Tai Chi in Pulmonary Rehabilitation for Chronic Obstructive Pulmonary Disease
Methods	This study will evaluate the effects of Chen‐style tai chi compared to conventional exercise in pulmonary rehabilitation for patients with COPD. Half of participants will receive the Chen‐style tai chi intervention, and the other half will receive the conventional exercise intervention. Both groups will receive the same eduction and support during pulmonary rehabilitation
Participants	Inclusion criteria FEV₁ between 30% and 80% of predicted normal values FEV₁/FVC ratio < 70% Exclusion criteria Very severe COPD (GOLD stage 4) COPD exacerbation within 2 weeks before baseline assessment Significant hypoxaemia at rest or during exercise (SpO₂ < 85%) Already following a rehabilitation programme Physical limitations that compromise participation in a tai chi or conventional exercise programme
Interventions	AMBMT Tai chi 12‐Week tai chi intervention, 3 sessions per week, 90 minutes per session PR 12‐Week conventional exercise intervention, 3 sessions per week, 90 minutes per session
Outcomes	Lung function CRQ 6MWT Endurance cycle test Berg's Balance Test CAT Quads isometric force Quads isokinetic function
Starting date	March 2014
Contact information	Louis Gendron; louis.gendron@criucpq.ulaval.ca
Notes

PERMALINK

Active mind‐body movement therapies as an adjunct to or in comparison with pulmonary rehabilitation for people with chronic obstructive pulmonary disease

Louis McCusky Gendron

Andre Nyberg

Didier Saey

François Maltais

Yves Lacasse

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Summary of findings

Background

Description of the condition

Description of the intervention

Active mind‐body movement therapies (AMBMTs)

1. Description of different AMBMTs.

Pulmonary rehabilitation (PR)

How the intervention might work

Why it is important to do this review

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

AMBMTs

PR

Types of outcome measures

Primary outcomes

Secondary outcomes

Search methods for identification of studies

Electronic searches

Searching other resources

Data collection and analysis

Selection of studies

Data extraction and management

Assessment of risk of bias in included studies

Assesment of bias in conducting the systematic review

Measures of treatment effect

Continuous data

Dichotomous data

Unit of analysis issues

Dealing with missing data

Assessment of heterogeneity

Assessment of reporting biases

Data synthesis

'Summary of findings' table

Subgroup analysis and investigation of heterogeneity

Sensitivity analysis

Results

Description of studies

Results of the search

1.

Included studies

Design

Sample size

Setting

Participants

Interventions

Excluded studies

Studies awaiting classification

Ongoing studies

Risk of bias in included studies

2.

3.

Allocation

Blinding

Incomplete outcome data

Selective reporting

Other potential sources of bias

Effects of interventions

Summary of findings for the main comparison. Summary of findings for AMBMT vs PR.

Summary of findings 2. Summary of findings for AMBMT + PR vs PR.

AMBMT versus PR