Effects of traditional Chinese exercise on sleep quality: A systematic review and meta-analysis of randomized controlled trials

Abstract

Background:

The efficacy of traditional Chinese exercise (TCE)-based interventions in the improvement of sleep quality is controversial. In this systematic review, we aimed to summarize randomized controlled trials (RCTs) that outline the effects of TCE on sleep quality.

Methods:

Five databases (Web of Science, Embase, PubMed, Medline, and China National Knowledge Infrastructure) were searched for literature published before July 2022. RCTs examining TCE interventions were included. The treatment effects were estimated using a random-effect meta-analysis model with mean differences (MDs). There were 2 outcome scales for sleep quality; however, because they were extremely contrastive to be analyzed by standard MD, the scales were analyzed separately to ensure the accuracy of the results. This review was registered in the International Prospective Register of Systematic Reviews (identifier CRD42023421314).

Results:

Twenty studies were included for analysis at last. The outcome was calculated using the Verran and Snyder-Halpern Sleep Scale (MD: 344.17, 95% confidence interval: 316.95 to 371.39, P < .00001) and Pittsburgh Sleep Quality Index to measure sleep quality (MD: −2.24, 95% confidence interval: −3.05 to −1.43, P < .00001), both showed improvement effect. In subgroup analysis, for patients with fibromyalgia, normal older adults, and non-Hodgkin lymphoma, chronic fatigue syndrome-like illness, knee osteoarthritis, nasopharyngeal carcinoma, pausimenia, insomnia, TCE could improve sleep quality. However, there was no significant improvement in stroke patients, breast cancer patients, normal college students, and episodic migraine patients. Tai Chi had greater effects in improving sleep quality than Qigong. In addition, the participants practice site, duration, and age did not influence the effects of TCE.

Conclusion:

TCE can improve sleep quality in specific populations in specific populations clinical applications. Tai chi should be considered first to improve sleep quality. However, further extensive trials and rigorous study designs should be conducted to strengthen the findings of this study. In addition, considering the large heterogeneity, the findings of our study should be interpreted cautiously.

1. Introduction

Sleep plays a vital role in promoting physical and mental well-being, and sleep quality is crucial for the overall health. Poor sleep quality has been linked to a variety of health problems, including obesity, diabetes, cardiovascular disease, and depression.^[1–3] Sleep disturbances have become increasingly more common and the prevalence of sleep disturbance is 17.65% to 81% in the general population.^[4] Reportedly, 10.1% reported more than 9 hours per night and 35.3% reported < 7 hours per night.^[5]

Specifically, approximately 35% to 40% of adults struggle with issues such as difficulty falling asleep, maintaining sleep, and early morning awakenings.^[6] Previous studies have shown that sleep disturbances can negatively impact mental health, thus increasing the risk of depression and suicidal ideation and diminishing the overall quality of life.^[7–9]

Some of the most recognized treatment approaches for addressing sleep disturbances include pharmacological therapy and psychotherapy. Pharmacotherapy, which involves using medications such as nonbenzodiazepines, benzodiazepines, melatonin agonists, and antidepressants, has been the primary treatment modality for sleep-related problems for a considerable time. However, there are growing concerns over the safety and efficacy of prescription sleep medications, including an increased emphasis on patient autonomy and active involvement in healthcare decision-making.^[10] Consequently, there has been an increasing interest in alternatives to pharmacological therapy for treating sleep-related issues.

In addition to costs associated with disease management, there are critical concerns regarding the clinical value of conventional first-line interventions. Pharmacological therapy may yield adverse events,^[11] whereas psychotherapy involves substantial time investment from healthcare providers.^[12] Medication use can result in cognitive decline, falls, and fractures, among other adverse effects.^[13–15] In contrast, traditional Chinese exercises (TCEs), such as Tai Chi and Qigong, have demonstrated potential health benefits and represent a viable treatment alternative. Tai Chi emphasizes slow movements and deep breathing,^[16] while Qigong promotes movement and breathing regulation to balance the body’s vital energy.^[17,18] Studies have supported the use of Health Qigong as a complementary treatment option for physical and psychological conditions.^[19–21] Unlike many Western exercise modalities that emphasize intense movements, TCEs prioritize relaxation and mindfulness, thus making them suitable for individuals of all ages and fitness levels.^[22]

Many studies have investigated the impact of TCE on sleep quality; however, the results remained inconclusive. While some studies indicated the effectiveness of TCE on sleep outcomes,^[23,24] others indicated no such effects.^[25,26] To the best of our knowledge, no meta-analysis has specifically examined the effects of TCE, such as Tai Chi and Qigong, on sleep quality, despite the necessity. In addition to the effects of TCE on sleep quality, we evaluated factors that could potentially influence its effectiveness, including intervention type, physical condition, practice site, practice duration, and age.

2. Materials and Methods

2.1. Data source

We conducted a comprehensive literature search using 5 electronic databases, including Web of Science, Embase, PubMed, Medline, and China National Knowledge Infrastructure. We searched for articles published from the inception of each database to July 2023. Two review team members (HL and SL) developed exhaustive search strategies for each electronic database. The search in English-language databases contained the following relevant terms: “sleep or insomnia*”, “traditional Chinese exercise”, “Taichi”, “Tai ji”, “Taiji”, “Taijiquan”, “Tai chi”, “qigong”, “Baduanjin”, “Liuzijue”, “Yijinjing”, “Wuqinxi”, “RCT*”, “randomi*”, and “clinical trial*”. We used equivalent search terms in the Chinese-language databases. Search tactics are shown in Appendix 1, Supplemental Digital Content, http://links.lww.com/MD/K470.

2.2. Inclusion and exclusion criteria

This meta-analysis has been reported in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines^[27] and has been registered at the International Prospective Register of Systematic Reviews (identifier CRD42023421314).

The inclusion criteria were as follows: Randomized controlled trials (RCTs); Participants aged > 18 years; and The intervention group received TCE, including Tai Chi and Qigong (such as Baduanjin, Yijinjing, or Wuqinxi). The control group interventions were limited to no exercises, usual care, low-impact exercise, or sham Qigong.

The exclusion criteria were as follows: Participants with organic mental disorders, neurological disorders, or motor impairments. This is because these participants usually have difficulty completing basic movements of traditional Chinese sports in a complete or standard manner, resulting in significant intervention effects with errors; The intervention group received either combined intervention; The control group received moderate- to high- intensity exercise intervention, computerized balance training, or cognitive behavioral therapy; Non-RCT studies.

2.3. Study selection and data extraction

Two reviewers (HL and SL) independently assessed the eligibility of articles by examining their title, abstract, and full text based on the predefined inclusion and exclusion criteria. Any discrepancies between the 2 reviewers were resolved through discussion and consultation with a third reviewer (BL). Relevant data such as author name, publication year, mean age or age range, sample size, outcome measures, interventions employed in the experimental and control groups, duration, and assessment methods were extracted from the selected articles. The primary outcome measures included Pittsburgh sleep quality index (PSQI) and Verran and Snyder-Halpern sleep scale (VSHSS).

2.4. Quality assessment

Two researchers (HL and SL) independently assessed the quality of the selected studies according to the Cochrane Collaboration’s tool for RCTs.^[28] Items were evaluated in 3 categories: low risk of bias, unclear bias, and high risk of bias. The following characteristics were evaluated: random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), incomplete outcome data (attrition bias), selective reporting (reporting bias), and other biases resulting from these questions were graphed and assessed using Review Manager 5.4 (Cochrane Collaboration). Any discrepancies were resolved either through discussion among the authors or with the involvement of a third member (LX) as needed.

2.5. Statistical analyses

We conducted a meta-analysis using Revman 5.4 software developed by the Cochrane Collaboration. We used mean difference (MD) as the effect size index for continuous variables with the same measurements, and standard MD for continuous variables with different measurements, accompanied by 95% confidence intervals (CIs) plotted in forest plots. Sensitivity and subgroup analyses were performed to determine the potential sources of heterogeneity. For the subgroup analysis, the studies were categorized into different subgroups based on the physical condition, practice site, intervention method, intervention duration, and participants’ age to identify the sources of heterogeneity. The random-effects model was used for the meta-analysis.

For the sensitivity analysis, the sources of heterogeneity were identified by reestimating the combined effects using the 1-by-1 elimination method. Funnel plots were generated to assess potential publication bias qualitatively when at least 10 articles were included. Funnel plots were visually inspected for asymmetry, and Egger test was used to perform sensitivity analysis and statistically assess publication bias. All statistical analyses were performed using STATA 11.0 (STATA Corp, College Station, TX).

3. Results

3.1. Search results

Our initial search yielded 901 articles in total, of which 436 were removed because of duplication. After screening the titles and abstracts, an additional 437 items were removed. Overall, 35 articles were reviewed; however, only 20 articles were included in this meta-analysis. No further study was identified by manual search. The flow diagram of the study selection is shown in Figure 1.

Figure 1.

Screening process for trials included in the meta-analysis.

3.2. Study characteristics

Twenty studies with 1832 participants were included in the meta-analysis, and the sample size of the studies ranged from 20 to 206. Two studies used the VSHSS, whereas 18 studies used the PSQI to measure sleep quality. Table 1 presents the characteristics of the 20 RCTs. Types of Health Qigong were Liuzijue,^[41] Chaoyi Fanhuan Qigong,^[29,43] Goulin New Qigong,^[26] Tai chi,^{[24,30,32–34,39,44,45]} Chan-Chuang Qigong,^[42] and Baduanjin.^{[23,25,31,35–38]} Among the 20 studies, 10 studies were conducted in China, 5 in the United States, 1 in Germany, 1 in Italy, 1 in Spain, 1 in Australia, and 1 in Canada. The disease conditions were breast cancer, pausimenia, stroke, chronic fatigue syndrome-like illness, colorectal cancer, episodic migraine, fibromyalgia, insomnia, knee osteoarthritis, nasopharyngeal carcinoma, and non-Hodgkin lymphoma. Among the 20 RCTs, 2 had an active control group receiving active interventions, including sham Qigong and low-impact exercise. The treatment dosages for the Tai Chi, Baduanjin Qigong, Chaoyi Fanhuan Qigong, Goulin New Qigong, Six Healing Sounds Qigong, and Chan-Chuang Qigong intervention groups were 8 weeks to 6 months, 9 to 24 weeks, 8 weeks, 5 to 6 weeks, 10 weeks, and 3 weeks, respectively. Among the 20 studies, 10 reported home practice as part of the intervention. The main characteristics of the 19 articles are summarized in Table 1.

Table 1.

Characteristics of the 20 studies included in the meta-analysis.

Study	Case type	Age (mean ± sd or range, yr)	Simple size (IG/CG)	Home practice (yes/no)	Intervention Group	Control Group	Duration time	Measurements
Mary[29] 2012,Canada	Fibromyalgia	IG: 52.81 ± 8.91 CG: 52.13 ± 8.56	100 (53/47)	Yes	Chaoyi Fanhuan qigong,45-60 minutes/d, review/group practice session,60 minutes/wk	Usual care	8 wk	PSQI
Manh[30] 2022,Germany	Healthy older people	IG: 69.23 ± 5.30 CG: 68.73 ± 4.95	73 (39/34)	Yes	Tai Chi, 60 min/time, 2 times/wk	Daily life	6 mo	PSQI
Jingwen[23] 2022,Italy	Breast cancer	IG: 53.12 ± 7.02 CG: 54.63 ± 8.44	68 (33/35)	No	Baduanjin qigong, 70 min/time, 2 times/wk	Usual care	12 wk	PSQI
Yun[31] 2018,China	Colorectal Cancer	IG: 55.60 ± 11.23 CG: 54.63 ± 11.88	87 (43/44)	No	Baduanjin qigong, 20–40 min/session, ≥5 times/wk,	Usual care	24 wk	PSQI
Ruth[32] 2014,USA	Stroke	IG: 71.5 ± 10.3 CG: 68.2 ± 10.3	101 (53/48)	Yes	Tai chi, 60 min/time, 3 times/wk	Usual care	12 wk	PSQI
Shuting[33] 2022,China	Episodic Migraine	18 to 65 years	73 (40/33)	Yes	Tai chi, 45 min/time, 3 times/wk	Daily life	12 wk	PSQI
Parco[34] 2020,China	Insomnia	IG: 66.5 ± 6.4 CG: 68.0 ± 8.2	115 (105/110)	No	Tai chi, 35 min/time, 3 times/wk	Low-impact exercise	12 wk	PSQI
Liying[35] 2023,China	Nasopharyngeal carcinoma	IG: 45.55 ± 8.99 CG: 47.07 ± 9.43	88 (44/44)	No	Baduanjin qigong, 25 min/time, 5 sessions/wk	Usual care	12 wk	PSQI
Jessie[36] 2014,China	Chronic Fatigue Syndrome-Like Illness	IG: 39.1 ± 7.8 CG: 38.9 ± 8.1	150 (75/75)	Yes	Baduanjin qigong, more than 30 min/time, sixteen sessions	Daily life	9 wk	PSQI
Maria[37] 2022,Spain	Pausimenia	IG: 69.70 ± 6.15 CG: 69.75 ± 6.76	117 (57/60)	Yes	Baduanjin qigong, 60 min/time, 2 times/wk	Daily life	12 wk	PSQI
Zhen[26] 2012,USA	Breast Cancer	IG: 45.3 ± 6.3 CG: 44.7 ± 9.7	96 (49/47)	No	Goulin New qigong, 21–24 min/time, 5 times/wk	Usual care	5 to 6 wk	PSQI
Mei-Chuan[38] 2012,China	Healthy older people	IG: 70.48 ± 7.90 CG: 72.96 ± 8.30	55 (27/28)	Yes	Baduanjin qigong, 30 min/time, 3 times/wk	Daily life	12 wk	PSQI
Li-Qun[39] 2022,Australia	Breast cancer	IG: 45.3 ± 8.5 CG: 48.6 ± 7.8	72 (36/36)	No	Tai chi, 25–30 min/time, 2 times/wk	Usual care	8 wk	PSQI
Moyi Li[25] 2015,China	Health	IG: 20.63 ± 1.03 CG: 20.92 ± 1.15	206 (101/105)	Yes	Baduanjin qigong, 60 min/time, 5 times/wk	Daily life	12 weeks	PSQI
Jiulong[24] 2022,China	Knee Osteoarthritis	IG: 64.15 ± 8.56 CG: 64.15 ± 8.56	40 (20/20)	No	Tai chi, 60 min/time, 3 times/wk	Daily life	12 wk	PSQI
Srujitha[40] 2022,USA	Stroke	IG: 70.8 ± 9.9 CG: 68.3 ± 10.3	145 (97/48)	Yes	Tai chi, 60 min/time, 3 times/wk	Usual care	12 wk	PSQI
Caio[41] 2020,USA	Fibromyalgia	IG: 42.6 ± 10.7 CG: 56.1 ± 12.3	20 (10/10)	Yes	Six healing sounds qigong, 25 min/d	Sham qigong	10 wk	PSQI
Tzu-Yun[42] 2017,China	Non-Hodgkin lymphoma	IG: 55.85 ± 16.78 CG: 64.54 ± 15.51	96 (48/48)	No	Chan-Chuang qigong, 15 min/time,2–3 times/d	Usual care	21 d	VSHSS
Mei-Ling[43] 2016,China	Non-Hodgkin lymphoma	59.79 ± 16.54 years, 23 to 90 years	102 (51/51)	No	Chaoyi Fanhuan qigong; 45–60 min/d	Usual care	8 wk	VSHSS
Ruth[44] 2011, USA	Stroke	IG: 72.8 ± 10.1 CG: 64.5 ± 10.9	28 (16/12)	No	Tai chi, 60 min/time, 3 times/wk	Usual care	12 wk	PSQI

CG = control group, e.g. = experimental group, PSQI = Pittsburgh sleep quality index, SD = standard deviation, VSHSS = Verran and Snyder-Halpern sleep scale.

3.3. Quality of the evidence

Two authors (HL and SL) evaluated the methodological aspects of each study, and their assessments were presented in Figures 2 and 3. Twenty trials had adequate random sequence generation, and 19 trials had sufficient allocation concealment to minimize election bias. However, allocation concealment was unclear in 1 trial. Due to the nature of TCEs, blinding of participants and personnel was not possible; therefore, blinding of participants and personnel was judged high in 16 trials. Eleven trials had sufficient blinding of the outcome assessment to prevent detection bias. All trials had an adequate assessment of incomplete outcome data. No selective reporting of outcomes was observed, and the risk of other biases was classified as low.

Figure 2.

Risk of bias summary.

Figure 3.

Risk of bias graph.

3.4. Effects of health qigong on overall sleep quality

Two studies used the VSHSS, whereas 18 studies used the PSQI to measure sleep quality. There were significant differences in the evaluation methods of the 2 scales; therefore, we conducted a meta-analysis of the studies using the 2 scales separately to obtain accurate statistical results.

The combined effect size of the 2 studies that used the VSHSS was 344.17 (95% CI: 316.95–371.39, P < .00001). The forest plot is presented in Figure 4. There was no significant heterogeneity (I² = 0, P = .45). The 2 studies that used the VSHSS to assess sleep quality employed Chan-Chuang Qigong and Chaoyi Fanhuan Qigong, respectively, and both demonstrated a significant improvement in sleep quality through Qigong practice. Given the limited number of studies that utilized the VSHSS, sensitivity analysis and publication bias tests were deemed inappropriate.

Figure 4.

Forest plots for TCE on sleep quality with VSHSS. TCE = traditional Chinese exercise, VSHSS = Verran and Snyder-Halpern sleep scale.

For studies that used the PSQI to measure sleep quality, the overall effect size was −2.24 (95% CI: −3.05 to −1.43, P < .00001, Fig. 5). A homogeneity test revealed high heterogeneity (P < .00001, I² = 86%). Sensitivity analysis demonstrated the robustness of the results (Fig. 6).

Figure 5.

Forest plots for TCE on sleep quality with PSQI. PSQI = Pittsburgh sleep quality index. TCE = traditional Chinese exercise.

Figure 6.

Sensitivity analysis of TCE for sleep quality with PSQI. PSQI = Pittsburgh sleep quality index, TCE = traditional Chinese exercise.

3.5. Subgroup analysis

Regarding physical condition (Fig. 7), the effect size of TCE on fibromyalgia patients sleep quality was −3.29 (95% CI: −6.49 to −0.10, P = .04; I² = 58%, P = .12); the effect size on breast cancer patients sleep quality was −2.24 (95% CI: −5.15 to 0.67, P = .13; I² = 90%, P < .0001); the effect size of TCE on the sleep quality of older adults with unknown medical conditions was −4.33 (95% CI: −5.33 to −3.34, P < .00001; I² = 0%, P = .92); the effect size of TCE on the sleep quality of stroke patients sleep quality was −0.18 (95% CI: −1.13 to 0.76, P = .70; I² = 0%, P = .87); the effect size of TCE on sleep quality of patients with non-Hodgkin lymphoma was 344.17 (95% CI: 316.95–371.39, P < .00001; I² = 0%, P = .45).

Figure 7.

Subgroup analysis of subjects’ physical condition for TCE on sleep quality. TCE = traditional Chinese exercise.

For patients with chronic fatigue syndrome-like illness (MD: −1.30, 95% CI: −2.46 to −0.14, P = .03), knee osteoarthritis (MD: −3.72, 95% CI: −5.44 to −2.00, P < .0001), nasopharyngeal carcinoma (MD: −2.75, 95% CI: −4.46 to −1.04, P = .002), pausimenia (MD: −3.14, 95% CI: −4.43 to −1.85, P < .00001), and insomnia (MD: −3.00, 95% CI: −3.98 to −2.02, P < .00001), TCE could improve sleep quality. Only 1 study has been conducted in each of these populations.

For normal college students (MD: −0.35, 95% CI: −0.82 to 0.12, P = .14) and episodic migraine (MD: −0.70, 95% CI: −2.41 to 1.01, P = .42), TCE could not improve sleep quality. Only 1 study has been conducted in each of these populations.

Regarding practice location (Fig. 8), the effect size for practicing at home was −2.35 (95% CI: −3.23 to −1.47, P < .00001; I² = 75%, P < .0001), and the effect size for practicing in other places was −2.08 (95% CI: −3.48 to −0.69, P = .003; I² = 89%, P < .00001). There was no significant difference in the effect of practice between sites (P = .75 > 0.05).

Figure 8.

Subgroup analysis of practice location for TCE on sleep quality. TCE = traditional Chinese exercise.

Regarding intervention methods (Fig. 9), the effect size of Tai Chi on sleep quality was −2.27 (95% CI: −2.79 to −1.76, P < .00001; I² = 82%, P < .00001). The effect size of Qigong on sleep quality was −1.55 (95% CI: −1.87 to −1.24, P < .00001; I² = 88%, P < .00001). There was a significant difference between the effects of practicing Tai Chi and Qigong (P = .02 < 0.05).

Figure 9.

Subgroup analysis of intervention methods for TCE on sleep quality. TCE = traditional Chinese exercise.

Regarding intervention duration (Fig. 10), the effect size of TCE intervention < 12 weeks was −2.43 (95% CI: −4.11 to −0.74, P = .005; I² = 84%, P < .0001), and the effect size for intervention ≥ 12 weeks was −2.17 (95% CI: −3.14 to −1.20, P < .0001; I² = 88%, P < .00001), and there was no significant difference in the intervention effect between the 2 intervention durations (P = .80 > 0.05).

Figure 10.

Subgroup analysis of intervention duration for TCE on sleep quality. TCE = traditional Chinese exercise.

Regarding participants age (Fig. 11), the effect size of TCE on participants sleep quality with an average age of ≤ 61 years was −1.99 (95% CI: −3.04 to −0.94, P = .0002; I² = 86%, P < .00001). The effect size of TCE for participants aged > 61 years was −2.53 (95% CI: −3.75 to −1.30, P < .0001; I² = 83%, P < .00001). There was no significant difference in the intervention effect between the 2 age groups (P = .51 > 0.05).

Figure 11.

Subgroup analysis of subjects’ age for TCE on sleep quality. TCE = traditional Chinese exercise.

3.6. Evaluation of publication bias

The publication bias of outcomes was evaluated using funnel plots based on the 17 studies using PSQI. The funnel plot was symmetrical, indicating that no significant publication bias existed (Fig. 12). Regarding publication bias, Egger test results showed P = .071 > 0.05, indicating that there was no significant publication bias in the above results.

Figure 12.

Funnel plots of publication bias.

4. Discussion

This meta-analysis and systematic review provided a quantitative estimate of whether TCE was a significant and effective strategy for sleep quality improvement. Our meta-analysis showed that TCE intervention was effective in improving sleep quality except for breast cancer patients, stroke patients, normal college students, and episodic migraine patients. Besides, Tai Chi proving more effective than Qigong.

Tai chi did not improve sleep quality in stroke patients, consistent with Lyu findings.^[40] The subjects in the 3 RCTS included in this study were older adults. Stroke patients often experience fragmented sleep, increased wakefulness, and reduced slow wave sleep.^[46] Stroke patients may also have comorbid conditions such as sleep-disordered breathing, which further exacerbates sleep disturbances. Furthermore, the presence of sleep disorders such as insomnia and sleep apnea in elderly stroke patients can also contribute to the lack of improvement in sleep quality with exercise.^[47]

The findings indicated that although Tai Chi and Qigong were effective in improving sleep quality, Tai Chi proved more effective than Qigong, contrary to the conclusion of previous studies.^[48,49] However, because few randomized studies had been performed in individuals with each physical condition, our conclusions needed to be interpreted with caution.

However, it is noteworthy to mention that previous systematic reviews and meta-analyses had broader requirements for the interventions employed in the control group. Previous studies have shown that Health Qigong (Baduanjin, Wuqinxi, and Liuzijue) was more effective than Tai Chi in improving sleep quality, contrary to our results. This may be attributed to the following 4 reasons. First, the control group of the studies we included was unable to participate in aerobic exercise or psychotherapy, and we ensured the feasibility of comparing the Tai Chi study with the Health Qigong study. Second, a previous study had included participants with drug addiction, different from the subjects in our study and we excluded it. Third, Tai Chi involves more physical movements and, therefore, provides greater physical activity, which may help to improve sleep quality. Qigong is a practice that involves a series of movements, breathing techniques, and meditation, which helps to balance and regulate the functional activities of the meridians and internal organs and promote the smooth and harmonious flow of vital energy (or “chi”) throughout the body.^[50–52] Qigong exercises necessitate the unification of the body, breath, and mind (spirit) in all forms – dynamic or static – and the conscious regulation of the flow of “qi” inside the body is the paramount and fundamental aspect of the practice.^[53,54] The diseases of the patients in the included studies may have probably exceeded the therapeutic capacity of Qigong, and because Tai Chi has more dynamic movements, it is more suitable for patients with severe conditions. The slow, controlled movements of Tai Chi may be more effective in promoting physical relaxation and reducing muscle tension,^[55] which can further contribute to improved sleep quality.

In addition to the intervention method, we conducted subgroup analyses on the physical condition, practice location, intervention duration, and age of the participants. The findings of our study suggested that the 4 aspects mentioned above did not significantly affect the effectiveness of TCE on sleep quality. However, there was significant heterogeneity in each subgroup analysis, including physical condition, practice location, intervention methods, intervention duration, and participants’ age. Heterogeneity existed across the 19 trials; however, the sensitivity analysis confirmed the robustness of the results. Nonetheless, further high-quality RCTs should be conducted in the future.

5. Implications for future studies

First, more high-quality RCTs should be conducted to strengthen the current evidence on TCE for sleep quality. Trials including participants with different health statuses, different types of frequencies, and adopting specific TCE-based interventions are warranted to determine the best sleep quality improvement program. Intervention outcomes could include not only postintervention outcomes but also medium and long-term outcomes to provide more comprehensive information. In addition, the intervention plan should be more detailed during the study, particularly the information on coaches and management of participants life. Finally, attention should also be paid to the intervention design of the control group and the use of sham Tai Chi and sham Qigong or other alternative interventions.

6. Strengths and limitations

One of the strengths of this meta-analysis was that we controlled the intervention for the participants in the intervention and control groups included in our study to explore the effects of TCE on sleep quality as precisely as possible. In addition, we conducted a subgroup analysis based on the participants physical condition, practice site, duration, and age. Nonetheless, this study has some limitations. First, as there were few experimental studies of other Qigong types, the effect sizes obtained from the experiments of Qigong as an intervention were biased toward Baduanjin Qigong. The comparison of the effects of Tai Chi and Qigong on sleep quality might not be precise enough. Second, to observe the clinical benefits of Health Qigong, regular practice is necessary; however, because the follow-up durations varied in the included studies, it prevented us from determining the long-term effects of Health Qigong on sleep quality. Third, the subjects included in this study had various types of physical conditions, but there were only 1 or 2 articles on subjects for each physical condition, and the number of subjects was small, which limited the accuracy of the statistical results. More randomized controlled trials targeting subjects with the same physical condition are needed in the future. Finally, some of the included studies were of low quality, and only 2 studies employed an active control group to ensure participant blinding, which could lead to nonspecific effects such as expectancy bias.

7. Conclusion

Our meta-analysis showed that TCE intervention was effective in improving sleep quality in specific populations’ clinical applications. Besides, Tai Chi proving more effective than Qigong. However, further extensive trials and rigorous study designs are warranted to strengthen the clinical significance of our findings.

Author contributions

Data curation: Haoyu Liu, Siling Liu.

Formal analysis: Haoyu Liu, Lu Xiong.

Funding acquisition: Haoyu Liu.

Investigation: Haoyu Liu, Siling Liu.

Methodology: Haoyu Liu, Siling Liu.

Project administration: Lu Xiong.

Resources: Haoyu Liu.

Software: Haoyu Liu.

Writing – original draft: Haoyu Liu, Bingquan Luo.

Writing – review & editing: Haoyu Liu, Siling Liu.

Supplementary Material

medi-102-e35767-s001.doc ^{(17KB, doc)}