Next Article in Journal
Impact of a Training Intervention on the Prevention of Aggressions in Nursing Students: A Pre–Post Study
Previous Article in Journal
Home Fetal Heart Rate Monitoring in Pregnancy: Patient Experience and Acceptance in the Era of Digital Prenatal Care
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Healthcare
Volume 14
Issue 12
10.3390/healthcare14121703
healthcare-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Systematic Review

Yoga and High-Intensity Interval Training Show Comparable Effects on HbA1c in Type 2 Diabetes: A Systematic Review and Preliminary Pilot Network Meta-Analysis in Adult Populations

by
Saw Ye Win Thu
,
Sneha Patnaik
and
Yin-Hwa Shih
*
Department of Healthcare Administration, Asia University, Taichung 41354, Taiwan
*
Author to whom correspondence should be addressed.
Healthcare 2026, 14(12), 1703; https://doi.org/10.3390/healthcare14121703 (registering DOI)
Submission received: 13 May 2026 / Revised: 10 June 2026 / Accepted: 11 June 2026 / Published: 15 June 2026
Browse Figures
Review Reports Versions Notes

Abstract

Background/Objectives: Exercise is pivotal for glycemic control in type 2 diabetes mellitus (T2DM), yet the relative efficacy of various exercise modalities remains inconclusive. This network meta-analysis aimed to evaluate and provide a preliminary ranking of exercise interventions on HbA1c levels in adults with type 2 diabetes mellitus, to facilitate clinically relevant network comparisons and to generate evidence for future large-scale comparative trials. Methods: A systematic review and network meta-analysis were conducted in accordance with PRISMA guidelines. Electronic databases (PubMed, MEDLINE, Cochrane Library, CINAHL, and ProQuest) were searched from inception to Dec 2024. Randomized controlled trials evaluating exercise interventions in adults with T2DM were included. Risk of bias was assessed independently by two reviewers using the JBI critical appraisal tool. The primary outcome was the change in HbA1c level. Results: Six randomized controlled trials involving a total of 511 participants (256 in the treatment group and 255 in the control group) were included in the final analysis. Both high-intensity interval training (MD = −0.322; 95% CI: −0.559 to −0.084; p = 0.008) and yoga (MD = −0.366; 95% CI: −0.534 to −0.198; p < 0.001) significantly reduced HbA1c compared with the active control. Although the preliminary ranking analysis suggested a higher probability of effectiveness for yoga (SUCRA 1) than for HIIT (SUCRA 0.5), the indirect comparison revealed no statistically significant difference in HbA1c reduction between the two interventions (MD = −0.044; 95% CI: −0.335 to 0.247; p = 0.766). Conclusions: These findings provide preliminary, evidence-generating; however, given the sparse network and absence of head-to-head trials, the treatment hierarchy should be interpreted with extreme caution and selected based on patients’ preferences and tolerance. Registration: PROSPERO [CRD42025650162].

1. Background

Type 2 diabetes mellitus (T2DM) is a chronic metabolic disorder characterized by persistent hyperglycemia due to insulin resistance and impaired insulin secretion [1]. It continues to pose a substantial global health burden and is associated with an increased risk of cardiovascular disease. In 2024, an estimated 588.7 million adults aged 20–79 years were living with diabetes (11.1% of the global population), accounting for 579.5 million cases (98.4%). It is projected to rise to 852.5 million cases by 2050, reflecting a 17% increase driven by population aging and increasing urbanization worldwide [2]. Rapid urbanization, sedentary behavior, and dietary transitions have accelerated the growing burden of disease [3]. Alongside pharmacological interventions, lifestyle modification is essential for effective T2DM management [4].
Physical activity is one of the most effective non-pharmacological strategies for prediabetes and T2DM, enhancing insulin sensitivity and thereby improving glycemic regulation [5]. Evidence indicated that aerobic exercise (AE) [6,7], resistance training (RT), and combined training (CT) all improved glycemic control, with CT demonstrating the greatest reduction in fasting blood glucose (FBG) [8,9]. However, exercise dosage was critically associated with a reduction in HbA1c [10].
In addition to traditional exercise modalities, mind–body practices such as yoga and tai chi have gained increasing recognition for their beneficial effects on glycemic regulation, particularly for individuals who find high-intensity exercise challenging, undesirable, or excessively demanding. Yoga, originating in India, has been shown to improve the regulation of blood glucose [11]. Tai chi, a traditional Chinese mind–body exercise emphasizing slow, mindful movements and controlled breathing, demonstrated beneficial glycemic control, particularly beneficial for older adults with T2DM who were at increased risk of falls [12]. Moreover, tai chi has been reported to be superior to brisk walking in reducing blood glucose and glycated hemoglobin across studies [13].
Given the lack of head-to-head trials and the heterogeneity in exercise modalities, intensities, and outcomes reported across existing studies, the comparative effectiveness of different exercise interventions for glycemic control in individuals with T2DM remains unclear. To support clinically relevant network comparisons, we conducted a preliminary pilot network meta-analysis of a range of exercise interventions, including high-intensity interval training (HIIT) and mind–body practices. Our objective was to evaluate and provide a preliminary ranking of exercise interventions on HbA1c levels in adults with type 2 diabetes mellitus, to facilitate clinically relevant network comparisons and to generate evidence for future large-scale comparative trials.

2. Methodology

2.1. Methodological Standards and Registration

The study was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO; registration number: CRD42025650162).

2.2. Search Strategy

A systematic literature search was conducted across five electronic databases: CINAHL, the Cochrane Library, MEDLINE, PubMed, and ProQuest. The search strategy was constructed based on the PICO framework, using controlled vocabulary and free-text terms for type 2 diabetes mellitus (T2DM), hyperglycemia, physical exercise, aerobic exercise, resistance training, walking, running, high-intensity interval training (HIIT), yoga, tai chi, HbA1c, glycemic control, postprandial glucose (PPG), fasting blood glucose (FBG), insulin sensitivity, body mass index (BMI), body weight, blood pressure, waist circumference (WC), and blood glucose. Detailed search terms and strategy were provided in Supplementary Data S1.

2.3. Study Selection

Following database searches, all retrieved articles were exported from each database into EndNote X9, and duplicates were removed. Titles and abstracts were independently screened by two reviewers, after which full-text articles were assessed for eligibility in accordance with the PRISMA guidelines. Studies were excluded if they met any of the following criteria: (1) lacked a control group; (2) did not specifically include individuals with type 2 diabetes mellitus (T2DM); (3) were published in languages other than English; or (4) were review articles, editorials, expert opinions, or meta-analyses. Any discrepancies between reviewers (SYWT and SN) were resolved through discussion or adjudication by a third reviewer (YHS).

2.4. Quality Assessment and Data Extraction

Two authors independently assessed the methodological quality and risk of bias of the included studies using the Joanna Briggs Institute (JBI) Critical Appraisal Tool for randomized controlled trials (RCTs) [14]. This tool evaluates specific methodological features of clinical trials, including randomization procedures, allocation concealment, and the blinding of participants, personnel, and outcome assessors. The quality thresholds applied (<49% high risk, 50–69% moderate risk, and >70% low risk) were adopted from established literature [15] to provide a standardized and objective categorization of the risk of bias across the included RCTs, ensuring consistency with established methodological standards. Overall, 2 studies (6.9%) were rated as having a high risk of bias, 10 studies (34.5%) as moderate risk, and the remaining 17 studies (58.6%) as low risk of bias, suggesting that 93.1% have moderate or low risk of bias (Supplementary Table S1). Although the relatively small proportion of high-risk studies may suggest a generally acceptable level of methodological quality, the potential influence of study limitations should be cautious when interpreting the pooled estimates and treatment rankings, particularly given the analysis’s preliminary nature.
Data extraction was performed independently using a standardized form and included (1) study characteristics: author, year of publication, country, study design, population characteristics, mean age, sample size, intervention groups, type of exercise intervention, and intensity monitoring; (2) outcomes, mean values, and standard deviations of HbA1c. All available full-text and Supplementary Materials were carefully examined for the mean and standard deviation of HbA1c change. Studies with missing or non-extractable data, inconsistent outcome reporting, or insufficient network connectivity were included in the qualitative (narrative) synthesis summarizing all 29 studies but were excluded from the quantitative network meta-analysis.

2.5. Network Meta-Analysis

Initial pairwise meta-analyses examining the effects of exercise interventions on HbA1c were conducted using Comprehensive Meta-Analysis (CMA) software version 2 (Biostat, Englewood, NJ, USA). Heterogeneity was assessed using the Q test and I2 statistic. Interventions demonstrating significant reductions in HbA1c were subsequently included in a network meta-analysis (NMA) performed using Stata 18 (StataCorp, College Station, TX, USA).
Because no direct head-to-head trials comparing high-intensity interval training (HIIT) and yoga were identified, NMA was used to facilitate indirect comparisons through a common comparator (active control). Arm-level data (study, treatment, mean change, standard deviation of change, and sample size) were entered using the network setup command (Supplementary Table S4). Analyses were based on change-from-baseline HbA1c values derived from reported pre- and post-intervention summary statistics. No imputation methods or assumed correlation coefficients were applied.
Treatment effects were synthesized using a random-effects model with the restricted maximum-likelihood (REML) estimator to account for between-study heterogeneity. Pooled mean differences (MDs) and corresponding 95% confidence intervals (CIs) were estimated relative to the common reference group.
The transitivity assumption was assessed by examining potential effect modifiers across studies, including baseline HbA1c levels, intervention duration, exercise intensity, and control group conditions. Although some variation existed, these characteristics were considered reasonably comparable across treatment comparisons, supporting the plausibility of indirect comparisons. Nevertheless, residual clinical heterogeneity cannot be excluded and should be considered when interpreting the findings.
The treatment network was sparse, contained no closed loops, and included no multi-arm trials. Therefore, indirect estimates were derived from independent studies connected through the common comparator, and covariance between treatment effect estimates was assumed to be zero. Formal inconsistency testing was not possible; consequently, the results should be interpreted as exploratory and hypothesis-generating.
Potential publication bias and small-study effects were explored using a comparison-adjusted funnel plot (Supplementary Table S3). Given the sparse network structure, the funnel plot was interpreted descriptively.
Network meta-analysis results were presented using network plots, forest plots, league tables, and Surface Under the Cumulative Ranking Curve (SUCRA) values. SUCRA values range from 0% to 100%, with higher values indicating a greater probability of ranking among the most effective interventions. Because treatment rankings in sparse networks may be unstable, SUCRA results were interpreted cautiously and considered exploratory.

3. Results

3.1. Screening Result

A total of 7152 records were identified through searches of five electronic databases, and an additional 15 records were identified through manual searching. After removing 290 duplicate records, 6862 articles remained for title and abstract screening. Of these, 6048 articles were excluded, leaving 814 articles for full-text review. Following full-text assessment, 799 articles were excluded because they did not meet the study objectives, reported mixed effects of exercise and medication, or failed to provide outcome data in the form of means and standard deviations. Ultimately, 29 studies met the inclusion criteria for the systematic review. Of these, 23 studies were excluded from the network meta-analysis due to missing or non-extractable HbA1c data, inconsistent reporting of outcomes, or insufficient connections within the treatment network (Figure 1).

3.2. Characteristics of the Included Study

All included studies were randomized controlled trials (RCTs). The sample size of the intervention groups ranged from 10 participants [16,17] to 163 participants [18], while the control size ranged from 9 [16] to 219 participants [19]. A total of 29 studies were included, comprising two studies from Egypt [20,21], one from the United Arab Emirates [19], one from Saudi Arabia [22], six from India [23,24,25,26,27,28], one from the United Kingdom [29], eight from China [13,18,30,31,32,33,34,35], one from Turkey [36,37,38], two from Japan [37,38], one from the United States [39], two from Iran [16,17], one from Portugal [40], one from Sri Lanka [41], one from Australia [42] and one from Indonesia [43].
Eleven studies investigated high-intensity interval training or moderate-intensity continuous training (HIIT/MICT) [16,20,21,22,29,31,36,39,40,42,43]; four examined tai chi [13,30,32,33]; six evaluated yoga interventions [24,25,26,27,28,37]; and eight focused on structured exercise programs combined with resistance training [17,18,19,23,34,35,38,41]. The intervention duration across studies ranged from 10 to 48 weeks. Outcomes were assessed at baseline and at the end of the intervention period, with measurement protocols varying among studies.
Exercise intensity was monitored using ergometers, Swedish Monark cycle ergometers, electronic treadmills, and electromagnetic bicycles. Resistance training protocols employed small equipment, such as dumbbells, resistance bands, and kettlebells, for exercises including seated rows, chest presses, leg presses, and planks, as well as weight-training machines for bench press, lat pulldown, biceps curl, and rowing. Aerobic exercise interventions included stretching, walking, stepping, and cycling. Yoga interventions primarily emphasized breathing techniques, meditation, and relaxation. Tai chi interventions included Chen-style tai chi, focusing on breathing, flexibility, stress reduction, and balance, as well as the 24-form Yang-style tai chi, which incorporated coordinated physical movements and breathing exercises. Articles included in the network meta-analysis were presented in Table 1, whereas studies included in the qualitative analysis were summarized in Table S2.

3.3. Evaluation of the HbA1c Reduction Effect Using Preliminary Meta-Analysis

Firstly, we categorized studies based on the PICO criteria, the primary outcome, and the similarity of control groups. Intervention classifications were based on the descriptions provided in the original articles and grouped according to their primary exercise characteristics.
Based on the consistency of outcomes, availability of comparable data, and methodological quality, 13 studies with comparable characteristics and at least three eligible trials were selected, grouped, and initially analyzed using meta-analysis. (Figure 2). The HIIT compared to usual care (SMD = −1.897, 95%: −3.313 to −0.480, p = 0.009) or active control (SMD = −0.684, 95%: −1.336 to −0.032, p = 0.009), and yoga compared to active control (SMD = −0.38, 95%: −0.562 to −0.166, p < 0.001) showed a significant effect on HbA1c reduction. The RT pool effect showed no significant difference compared with usual care (SMD = −0.205, 95% CI: −0.568 to 0.159, p = 0.27) (Figure 2).

3.4. Comparison of the HbA1c Reduction Between HIIT and Yoga in Network Meta-Analysis

Subsequently, studies evaluating HIIT and yoga compared with active control were included in the network meta-analysis [25,28,31,36,37,42]. A total of six randomized controlled trials were included, comprising three interventions: HIIT, yoga, and an active control. The network plot demonstrated a connected three-node structure with active control serving as the reference intervention. Most studies provided direct evidence for comparisons between active control and HIIT, or between active control and yoga, whereas the contrast between HIIT and Yoga was primarily based on indirect evidence (Figure 3).

3.5. Relative Treatment Effects on HbA1c Reduction

A multivariate random-effects model was fitted to evaluate network consistency (Table 2). The between-study variance was negligible (τ = 2.91 × 10−12), suggesting minimal heterogeneity across studies and supporting the assumption of model consistency. HIIT (MD = −0.322, 95% CI: −0.559, −0.084, p = 0.008) and yoga (MD = −0.366, 95% CI: −0.534, −0.198, p < 0.001) were significantly more effective than active control in improving HbA1c level. The indirect comparison between HIIT and yoga showed no statistically significant difference (MD = −0.044, 95% CI: −0.335 to 0.247, p = 0.766) (Table 2 and Figure 4). Visual inspection of the comparison-adjusted funnel plot did not reveal substantial asymmetry (Supplementary Table S3). Because the network included only a limited number of studies and lacked closed loops, the interpretation of publication bias and the formal assessment of inconsistency were limited.

3.6. Relative Treatment Effects and Ranking Probabilities

Compared with active control, both HIIT and yoga significantly reduced HbA1c levels. The mean difference (MD) for HIIT versus active control was 0.322 (95% CI: 0.084 to 0.559), while the MD for yoga versus active control was 0.366 (95% CI: 0.198 to 0.534). An indirect comparison between HIIT and yoga revealed no significant difference in HbA1c reduction, with an MD of −0.044 (95% CI: −0.335 to 0.247). Although yoga showed a numerically greater reduction in HbA1c than HIIT, the difference between the two interventions was not statistically significant (Table 3).
Ranking probability analysis suggested that yoga had a higher probability of being the most effective intervention for HbA1c reduction, with a probability of 1.00 for being ranked first and a SUCRA value of 1.00. HIIT was consistently ranked second, with a probability of 1.00 for being the second-best intervention and a SUCRA value of 0.50. Active control was ranked as the least effective intervention, with a probability of 1.00 for being ranked third and a SUCRA value of 0.00 (Table 4). However, it is essential to note that SUCRA rankings are probabilistic estimates and do not imply definitive clinical superiority, particularly as the indirect comparison between yoga and HIIT showed no statistically significant difference (p = 0.766).

4. Discussion

This study addressed an important gap in the literature where direct head-to-head comparisons between HIIT and yoga are lacking. Using a network meta-analysis framework, both HIIT and yoga were associated with significant reductions in HbA1c compared with an active control. Although yoga had the highest probability and SUCRA values, the indirect comparison between yoga and HIIT was not statistically significant. Therefore, the ranking results should be interpreted as probabilistic and exploratory rather than definitive evidence of superiority. Because all included studies involved adults with type 2 diabetes mellitus and were connected through a common comparator, the transitivity assumption was considered reasonably plausible. Nevertheless, residual heterogeneity may have influenced the indirect comparison, and the findings should be interpreted with caution.
The beneficial effects of HIIT on glycemic control are believed to result primarily from improved insulin sensitivity and enhanced skeletal muscle glucose uptake [21,29,44,45]. In the included studies, exercise intensity was prescribed using objective physiological measures, such as heart rate or VO2max, which may have improved the consistency of intervention delivery. In addition, the shorter time commitment of HIIT may make it more practical for some individuals. Yoga may improve glycemic control through multiple mechanisms [46,47,48], including enhanced insulin sensitivity, increased glucose uptake, and attenuation of hypothalamic–pituitary–adrenal axis activation, thereby reducing cortisol levels and improving metabolic regulation [49,50]. Previous studies have reported glycemic benefits of yoga comparable to those achieved with conventional exercise programs [51,52], suggesting that yoga may represent a feasible alternative exercise modality for selected patients.
Considerable heterogeneity existed across studies, including differences in exercise intensity, intervention duration, participant characteristics, and intervention protocols. Yoga interventions ranged from physical postures alone to integrated mind–body practices incorporating breathing exercises and meditation, while exercise programs varied substantially in duration and training prescription. These factors may have contributed to variability in treatment effects. Furthermore, only 13 studies were eligible for the preliminary meta-analysis, and only six studies could be incorporated into the final network because of the requirement for a connected treatment network. Consequently, more advanced analyses, such as subgroup analysis and meta-regression, were not feasible. Therefore, the present findings should be viewed as preliminary evidence intended to inform future research rather than provide definitive conclusions regarding the comparative effectiveness of exercise modalities for glycemic control in adults with type 2 diabetes mellitus.

5. Limitations of the Study

Several limitations of this study should be acknowledged. First, although 29 studies were included in the systematic review, only six trials were eligible for the final network meta-analysis. The sparse treatment network, limited number of interventions, and absence of closed-loop structures reduced statistical power and prevented formal inconsistency testing, node-splitting analyses, and robust assessment of publication bias.
Second, substantial clinical and methodological heterogeneity existed across studies, including variations in exercise intensity, intervention duration (10–52 weeks), participant characteristics, and intervention protocols. In addition, baseline physical activity levels, dietary intake, age at T2DM onset, and disease duration were not consistently reported, limiting the ability to perform subgroup analyses and fully evaluate potential effect modifiers.
Third, the comparison between HIIT and yoga was based entirely on indirect evidence through a common comparator. Therefore, the estimated treatment effects, SUCRA rankings, and ranking probabilities should be interpreted cautiously as exploratory findings rather than definitive evidence of superiority, particularly because the indirect comparison was not statistically significant (p = 0.766).
Fourth, the participant eligibility criterion was modified from the original PROSPERO-registered protocol. The initial protocol specified sedentary individuals with T2DM; however, this restriction was removed because sedentary status was inconsistently defined and reported across studies. Although this modification improved the comprehensiveness of the evidence base and did not alter the primary research question, it represents a protocol deviation and may have introduced selection bias.
Finally, the exclusion of non-English publications may have introduced language bias. This restriction was necessary because of limited access to professional translation resources required for accurate data extraction and synthesis.
Therefore, larger, well-designed head-to-head randomized controlled trials with standardized reporting of participant characteristics and intervention protocols are needed to confirm the comparative effectiveness of different exercise modalities for glycemic control in T2DM.

6. Conclusions

In conclusion, this pilot network meta-analysis suggests that both high-intensity interval training and yoga may improve glycemic control in patients with type 2 diabetes mellitus compared with active control. However, the treatment network was sparse, and the comparison between yoga and HIIT relied entirely on indirect evidence. Although yoga had the highest probability ranking, no statistically significant difference was observed between the two interventions. Therefore, the findings should be considered exploratory and hypothesis-generating rather than definitive.
Further well-designed head-to-head randomized controlled trials are needed to establish the comparative effectiveness of different exercise modalities for glycemic management in patients with type 2 diabetes mellitus.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/healthcare14121703/s1. Supplementary File S1: Date S1: Search term and search string; Table S1: JBI quality assessment of qualitative articles, Table S2: Characteristics of all the included articles in qualitative synthesis; Table S3: Study effect and comparison-adjusted funnel plot; Table S4: Arm-level data of studies included in NMA. Supplementary File S2: PRISMA checklist. Ref. [53] is cited in File S2.

Author Contributions

Conceptualization, Y.-H.S. and S.Y.W.T.; methodology, S.Y.W.T. and S.P.; software, Y.-H.S.; validation, S.Y.W.T. and S.P.; formal analysis, S.Y.W.T. and Y.-H.S.; investigation, S.Y.W.T. and Y.-H.S.; resources, Y.-H.S.; data curation, Y.-H.S.; writing—original draft preparation, S.Y.W.T.; writing—review and editing, Y.-H.S.; project administration, Y.-H.S. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

No new data were created or analyzed in this study.

Acknowledgments

The authors gratefully acknowledge Tzong-Ming Shieh for his valuable contributions to resources, data curation, and manuscript review and editing during the initial phase of the project.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Holt, R.I.G.; DeVries, J.H.; Hess-Fischl, A.; Hirsch, I.B.; Kirkman, M.S.; Klupa, T.; Ludwig, B.; Norgaard, K.; Pettus, J.; Renard, E.; et al. The Management of Type 1 Diabetes in Adults. A Consensus Report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care 2021, 44, 2589–2625. [Google Scholar] [CrossRef]
  2. International Diabetes Federation. Diabetes Around the World in 2025. Available online: https://diabetesatlas.org/ (accessed on 13 October 2025).
  3. Sun, H.; Saeedi, P.; Karuranga, S.; Pinkepank, M.; Ogurtsova, K.; Duncan, B.B.; Stein, C.; Basit, A.; Chan, J.C.N.; Mbanya, J.C.; et al. IDF Diabetes Atlas: Global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res. Clin. Pract. 2022, 183, 109119. [Google Scholar] [CrossRef]
  4. Colberg, S.R.; Albright, A.L.; Blissmer, B.J.; Braun, B.; Chasan-Taber, L.; Fernhall, B.; Regensteiner, J.G.; Rubin, R.R.; Sigal, R.J.; American College of Sports, M.; et al. Exercise and type 2 diabetes: American College of Sports Medicine and the American Diabetes Association: Joint position statement. Exercise and type 2 diabetes. Med. Sci. Sports Exerc. 2010, 42, 2282–2303. [Google Scholar] [CrossRef]
  5. Bird, S.R.; Hawley, J.A. Update on the effects of physical activity on insulin sensitivity in humans. BMJ Open Sport Exerc. Med. 2016, 2, e000143. [Google Scholar] [CrossRef]
  6. Zhang, L.; Cheng, X.; Yang, Y.; Li, X.; Yuan, Y. Optimal dosage and modality of exercise on glycemic control in people with prediabetes: A systematic review and network meta-analysis. Front. Endocrinol. 2025, 16, 1560676. [Google Scholar] [CrossRef]
  7. de Mello, M.B.; Righi, N.C.; Schuch, F.B.; Signori, L.U.; da Silva, A.M.V. Effect of high-intensity interval training protocols on VO(2)max and HbA1c level in people with type 2 diabetes: A systematic review and meta-analysis. Ann. Phys. Rehabil. Med. 2022, 65, 101586. [Google Scholar] [CrossRef]
  8. Colberg, S.R.; Sigal, R.J.; Yardley, J.E.; Riddell, M.C.; Dunstan, D.W.; Dempsey, P.C.; Horton, E.S.; Castorino, K.; Tate, D.F. Physical Activity/Exercise and Diabetes: A Position Statement of the American Diabetes Association. Diabetes Care 2016, 39, 2065–2079. [Google Scholar] [CrossRef]
  9. Pan, B.; Ge, L.; Xun, Y.Q.; Chen, Y.J.; Gao, C.Y.; Han, X.; Zuo, L.Q.; Shan, H.Q.; Yang, K.H.; Ding, G.W.; et al. Exercise training modalities in patients with type 2 diabetes mellitus: A systematic review and network meta-analysis. Int. J. Behav. Nutr. Phys. Act. 2018, 15, 72. [Google Scholar] [CrossRef]
  10. Jayedi, A.; Emadi, A.; Shab-Bidar, S. Dose-Dependent Effect of Supervised Aerobic Exercise on HbA(1c) in Patients with Type 2 Diabetes: A Meta-analysis of Randomized Controlled Trials. Sports Med. 2022, 52, 1919–1938. [Google Scholar] [CrossRef]
  11. Cui, J.; Yan, J.H.; Yan, L.M.; Pan, L.; Le, J.J.; Guo, Y.Z. Effects of yoga in adults with type 2 diabetes mellitus: A meta-analysis. J. Diabetes Investig. 2017, 8, 201–209. [Google Scholar] [CrossRef]
  12. Wang, Y.; Yan, J.; Zhang, P.; Yang, P.; Zhang, W.; Lu, M. Tai Chi Program to Improve Glucose Control and Quality of Life for the Elderly with Type 2 Diabetes: A Meta-analysis. Inquiry 2022, 59, 469580211067934. [Google Scholar] [CrossRef]
  13. Chan, A.W.K.; Chair, S.Y.; Lee, D.T.F.; Leung, D.Y.P.; Sit, J.W.H.; Cheng, H.Y.; Taylor-Piliae, R.E. Tai Chi exercise is more effective than brisk walking in reducing cardiovascular disease risk factors among adults with hypertension: A randomised controlled trial. Int. J. Nurs. Stud. 2018, 88, 44–52. [Google Scholar] [CrossRef]
  14. Joanna Briggs Institute. Critical Appraisal Tools. Available online: https://jbi.global/critical-appraisal-tools (accessed on 15 May 2024).
  15. Goplen, C.M.; Verbeek, W.; Kang, S.H.; Jones, C.A.; Voaklander, D.C.; Churchill, T.A.; Beaupre, L.A. Preoperative Opioid Use is Associated with Worse Patient Outcomes After Total Joint Arthroplasty: A Systematic Review and Meta-analysis. BMC Musculoskelet. Disord. 2019, 20, 234. [Google Scholar] [CrossRef]
  16. Mir, E.; Moazzami, M.; Bijeh, N.; Hakak Dokht, E.; Rahimi, N. Changes in SFRP5, WNT5A, HbA1c, BMI, PBF, and insulin resistance in men with type 2 diabetes after 12 weeks of combined exercise (HIIT and resistance). Int. J. Diabetes Dev. Ctries. 2020, 40, 248–254. [Google Scholar] [CrossRef]
  17. Nazari, M.; Shabani, R.; Darsara, S.G. Effect of circuit resistance and jogging exercise training on the markers of metabolic syndrome in women with type 2 diabetes. J. Phys. Act. Horm. 2023, 5, 76. [Google Scholar]
  18. Zheng, H.P.; Qiao, X.N.; Zhang, Q.; Li, Q.C.; Yi, N.; Ji, L.J.; Liu, S.Y.; Zhang, S.; Zhou, X.M.; Liu, X.X.; et al. The impact on glycemic control through progressive resistance training with bioDensity(TM) in Chinese elderly patients with type 2 diabetes: The PReTTy2 (Progressive Resistance Training in Type 2 Diabetes) Trial. Diabetes Res. Clin. Pract. 2019, 150, 64–71. [Google Scholar] [CrossRef]
  19. Ali, H.I.; Al Ketbi, L.B.; Platat, C.; Abdl El Baki, H.; Elmi, F.; Ibrahim, W.; Zoubeidi, T.; Al Dhaheri, A.S.; Cheikh Ismail, L.; Tariq, M.N.M.; et al. Impact of Skills for Change Program on metabolic control, diet and physical activity levels in adults with type 2 diabetes: A cluster randomized trial. PLoS ONE 2024, 19, e0304639. [Google Scholar] [CrossRef]
  20. Abdelbasset, W.K.; Tantawy, S.A.; Kamel, D.M.; Alqahtani, B.A.; Elnegamy, T.E.; Soliman, G.S.; Ibrahim, A.A. Effects of high-intensity interval and moderate-intensity continuous aerobic exercise on diabetic obese patients with nonalcoholic fatty liver disease: A comparative randomized controlled trial. Medicine 2020, 99, e19471. [Google Scholar] [CrossRef]
  21. Ahmad, A.M.; Mahmoud, A.M.; Serry, Z.H.; Mohamed, M.M.; Abd Elghaffar, H.A. Effects of low-versus high-volume high-intensity interval training on glycemic control and quality of life in obese women with type 2 diabetes. A randomized controlled trial. J. Exerc. Sci. Fit. 2023, 21, 395–404. [Google Scholar] [CrossRef]
  22. Al-Rawaf, H.A.; Gabr, S.A.; Iqbal, A.; Alghadir, A.H. High-Intensity Interval Training Improves Glycemic Control, Cellular Apoptosis, and Oxidative Stress of Type 2 Diabetic Patients. Medicina 2023, 59, 1320. [Google Scholar] [CrossRef]
  23. Amaravadi, S.K.; Maiya, G.A.; Vaishali, K.; Shastry, B.A. Effectiveness of structured exercise program on insulin resistance and quality of life in type 2 diabetes mellitus-A randomized controlled trial. PLoS ONE 2024, 19, e0302831. [Google Scholar] [CrossRef]
  24. Chattopadhyay, K.; Mishra, P.; Singh, K.; Singh, K.; Harris, T.; Hamer, M.; Greenfield, S.M.; Manjunath, N.K.; Nair, R.; Mukherjee, S.; et al. Yoga Programme for Type 2 Diabetes Prevention (YOGA-DP) Among High-Risk People in India: A Multicenter Feasibility Randomized Controlled Trial. Diabetes Ther. 2023, 14, 1137–1154. [Google Scholar] [CrossRef]
  25. Gupta, U.; Gupta, Y.; Jose, D.; Mani, K.; Jyotsna, V.P.; Sharma, G.; Tandon, N. Effectiveness of Yoga-based Exercise Program Compared to Usual Care, in Improving HbA1c in Individuals with Type 2 Diabetes: A Randomized Control Trial. Int. J. Yoga 2020, 13, 233–238. [Google Scholar] [CrossRef]
  26. Ranga, S.; Agrawal, S.; Rukadikar, C.; Rukadikar, A. Study of effect of yoga on blood pressure, blood glucose, and glycosylated hemoglobin levels of patients with Type 2 diabetes mellitus. Natl. J. Physiol. Pharm. Pharmacol. 2021, 11, 449. [Google Scholar] [CrossRef]
  27. Sivapuram, M.S.; Srivastava, V.; Kaur, N.; Anand, A.; Nagarathna, R.; Patil, S.; Biman, S.; Chander, I.; Jyoti, S.; Nagendra, H.R. Ayurveda Body-Mind Constitutional Types and Role of Yoga Intervention Among Type 2 Diabetes Mellitus Population of Chandigarh and Panchkula Regions. Ann. Neurosci. 2020, 27, 214–223. [Google Scholar] [CrossRef]
  28. Viswanathan, V.; Sivakumar, S.; Sai Prathiba, A.; Devarajan, A.; George, L.; Kumpatla, S. Effect of yoga intervention on biochemical, oxidative stress markers, inflammatory markers and sleep quality among subjects with type 2 diabetes in South India: Results from the SATYAM project. Diabetes Res. Clin. Pract. 2021, 172, 108644. [Google Scholar] [CrossRef]
  29. Cassidy, S.; Vaidya, V.; Houghton, D.; Zalewski, P.; Seferovic, J.P.; Hallsworth, K.; MacGowan, G.A.; Trenell, M.I.; Jakovljevic, D.G. Unsupervised high-intensity interval training improves glycaemic control but not cardiovascular autonomic function in type 2 diabetes patients: A randomised controlled trial. Diab Vasc. Dis. Res. 2019, 16, 69–76. [Google Scholar] [CrossRef]
  30. Leung, L.Y.; Chan, A.W.; Sit, J.W.; Liu, T.; Taylor-Piliae, R.E. Tai Chi in Chinese adults with metabolic syndrome: A pilot randomized controlled trial. Complement Ther. Med. 2019, 46, 54–61. [Google Scholar] [CrossRef]
  31. Li, J.; Cheng, W.; Ma, H. A Comparative Study of Health Efficacy Indicators in Subjects with T2DM Applying Power Cycling to 12 Weeks of Low-Volume High-Intensity Interval Training and Moderate-Intensity Continuous Training. J. Diabetes Res. 2022, 2022, 9273830. [Google Scholar] [CrossRef]
  32. Li, X.; Si, H.; Chen, Y.; Li, S.; Yin, N.; Wang, Z. Effects of fitness qigong and tai chi on middle-aged and elderly patients with type 2 diabetes mellitus. PLoS ONE 2020, 15, e0243989. [Google Scholar] [CrossRef]
  33. Liu, X.; Zhu, H.; Peng, Y.; Liu, Y.; Shi, X. Twenty-Four week Taichi training improves pulmonary diffusion capacity and glycemic control in patients with Type 2 diabetes mellitus. PLoS ONE 2024, 19, e0299495. [Google Scholar] [CrossRef]
  34. Ma, X.; Ai, Y.; Lei, F.; Tang, X.; Li, Q.; Huang, Y.; Zhan, Y.; Mao, Q.; Wang, L.; Lei, F.; et al. Effect of blood flow-restrictive resistance training on metabolic disorder and body composition in older adults with type 2 diabetes: A randomized controlled study. Front. Endocrinol. 2024, 15, 1409267. [Google Scholar] [CrossRef]
  35. Qian, Z.; Ping, L.; Xu, D.M.; Zhang, X.L. Slow-velocity eccentric-only resistance training improves symptoms of type 2 diabetic mellitus patients by regulating plasma MMP-2 and -9. Medicine 2024, 103, e38855. [Google Scholar] [CrossRef]
  36. Findikoglu, G.; Altinkapak, A.; Yaylali, G.F. Is isoenergetic high-intensity interval exercise superior to moderate-intensity continuous exercise for cardiometabolic risk factors in individuals with type 2 diabetes mellitus? A single-blinded randomized controlled study. Eur. J. Sport Sci. 2023, 23, 2086–2097. [Google Scholar] [CrossRef]
  37. Hirosaki, M.; Ohira, T.; Wu, Y.; Eguchi, E.; Shirai, K.; Imano, H.; Funakubo, N.; Nishizawa, H.; Katakami, N.; Shimomura, I.; et al. Laughter yoga as an enjoyable therapeutic approach for glycemic control in individuals with type 2 diabetes: A randomized controlled trial. Front. Endocrinol. 2023, 14, 1148468. [Google Scholar] [CrossRef]
  38. Yamamoto, Y.; Nagai, Y.; Kawanabe, S.; Hishida, Y.; Hiraki, K.; Sone, M.; Tanaka, Y. Effects of resistance training using elastic bands on muscle strength with or without a leucine supplement for 48 weeks in elderly patients with type 2 diabetes. Endocr. J. 2021, 68, 291–298. [Google Scholar] [CrossRef]
  39. Hwang, C.L.; Lim, J.; Yoo, J.K.; Kim, H.K.; Hwang, M.H.; Handberg, E.M.; Petersen, J.W.; Holmer, B.J.; Leey Casella, J.A.; Cusi, K.; et al. Effect of all-extremity high-intensity interval training vs. moderate-intensity continuous training on aerobic fitness in middle-aged and older adults with type 2 diabetes: A randomized controlled trial. Exp. Gerontol. 2019, 116, 46–53. [Google Scholar] [CrossRef]
  40. Magalhaes, J.P.; Judice, P.B.; Ribeiro, R.; Andrade, R.; Raposo, J.; Dores, H.; Bicho, M.; Sardinha, L.B. Effectiveness of high-intensity interval training combined with resistance training versus continuous moderate-intensity training combined with resistance training in patients with type 2 diabetes: A one-year randomized controlled trial. Diabetes Obes. Metab. 2019, 21, 550–559. [Google Scholar] [CrossRef]
  41. Ranasinghe, C.; Devage, S.; Constantine, G.R.; Katulanda, P.; Hills, A.P.; King, N.A. Glycemic and cardiometabolic effects of exercise in South Asian Sri Lankans with type 2 diabetes mellitus: A randomized controlled trial Sri Lanka diabetes aerobic and resistance training study (SL-DARTS). Diabetes Metab. Syndr. 2021, 15, 77–85. [Google Scholar] [CrossRef]
  42. Way, K.L.; Sabag, A.; Sultana, R.N.; Baker, M.K.; Keating, S.E.; Lanting, S.; Gerofi, J.; Chuter, V.H.; Caterson, I.D.; Twigg, S.M.; et al. The effect of low-volume high-intensity interval training on cardiovascular health outcomes in type 2 diabetes: A randomised controlled trial. Int. J. Cardiol. 2020, 320, 148–154. [Google Scholar] [CrossRef]
  43. Sudarsono, N.C.; Tulaar, A.B.M.; Jusman, S.W.A.; Soewondo, P.; Sudaryo, M.D.K.; Siagian, M.; Karhiwikarta, W. The Effects of Combined High-Intensity Interval and Resistance Training on Glycemic Control and Oxidative Stress in T2DM. Asian J. Sports Med. 2019, 10, e91841. [Google Scholar] [CrossRef]
  44. Liu, J.X.; Zhu, L.; Li, P.J.; Li, N.; Xu, Y.B. Effectiveness of high-intensity interval training on glycemic control and cardiorespiratory fitness in patients with type 2 diabetes: A systematic review and meta-analysis. Aging Clin. Exp. Res. 2019, 31, 575–593. [Google Scholar] [CrossRef]
  45. Poon, E.T.; Li, H.Y.; Kong, A.P.S.; Little, J.P. Efficacy of high-intensity interval training in individuals with type 2 diabetes mellitus: An umbrella review of systematic reviews and meta-analyses. Diabetes Obes. Metab. 2025, 27, 1719–1734. [Google Scholar] [CrossRef] [PubMed]
  46. Balaji, P.A.; Varne, S.R.; Ali, S.S. Physiological effects of yogic practices and transcendental meditation in health and disease. N. Am. J. Med. Sci. 2012, 4, 442–448. [Google Scholar] [CrossRef]
  47. Raveendran, A.V.; Deshpandae, A.; Joshi, S.R. Therapeutic Role of Yoga in Type 2 Diabetes. Endocrinol. Metab. 2018, 33, 307–317. [Google Scholar] [CrossRef] [PubMed]
  48. Misra, P.; Sharma, G.; Tandon, N.; Kant, S.; Sangral, M.; Rai, S.K.; Yadav, K.; Vishnubhatla, S.; Mandal, S.; Kardam, P.; et al. Effect of Community-Based Structured Yoga Program on Hba1c Level among Type 2 Diabetes Mellitus Patients: An Interventional Study. Int. J. Yoga 2021, 14, 222–228. [Google Scholar] [CrossRef] [PubMed]
  49. Chiarpenello, C.; Brodmann, K. What can the psychoneuroimmunology of yoga teach us about depression’s psychopathology? Brain Behav. Immun.—Health 2024, 42, 100877. [Google Scholar] [CrossRef]
  50. Marshall, M.; McClanahan, M.; McArthur Warren, S.; Rogers, R.; Ballmann, C. A Comparison of the Acute Effects of Different Forms of Yoga on Physiological and Psychological Stress: A Pilot Study. Int. J. Environ. Res. Public Health 2020, 17, 6090. [Google Scholar] [CrossRef] [PubMed]
  51. Li, X.; Gao, M.; Hua, J. Comparative efficacy of various mind-body exercise types on cardiometabolic health in patients with type 2 diabetes: A network meta-analysis of randomized controlled trials. BMC Cardiovasc Disord. 2025, 25, 291. [Google Scholar] [CrossRef]
  52. Singh, V.P.; Khandelwal, B. Effect of Yoga and Exercise on Glycemic Control and Psychosocial Parameters in Type 2 Diabetes Mellitus: A Randomized Controlled Study. Int. J. Yoga 2020, 13, 144–151. [Google Scholar] [CrossRef]
  53. Page, M.J.; McKenzie, J.E.; Bossuyt, P.M.; Boutron, I.; Hoffmann, T.C.; Mulrow, C.D.; Shamseer, L.; Tetzlaff, J.M.; Tetzlaff, E.A.; Brennan, S.E.; et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ 2021, 372, n71. [Google Scholar] [CrossRef] [PubMed]
Healthcare 14 01703 g001
Figure 1. PRISMA 2020 flow diagram of study identification, screening, eligibility, and inclusion.
Figure 1. PRISMA 2020 flow diagram of study identification, screening, eligibility, and inclusion.
Healthcare 14 01703 g001
Healthcare 14 01703 g002
Figure 2. Forest plot of the effects of different exercise modalities on HbA1c improvement (preliminary meta-analysis). (A) High-intensity interval training compared with usual care [20,21,29,39]. (B) High-intensity interval training compared to active control [31,36,42]. (C) Yoga compared to active control [25,28,37]. (D) Resistance training compared to usual care [18,34,38].
Figure 2. Forest plot of the effects of different exercise modalities on HbA1c improvement (preliminary meta-analysis). (A) High-intensity interval training compared with usual care [20,21,29,39]. (B) High-intensity interval training compared to active control [31,36,42]. (C) Yoga compared to active control [25,28,37]. (D) Resistance training compared to usual care [18,34,38].
Healthcare 14 01703 g002
Healthcare 14 01703 g003
Figure 3. Network plot of exercise interventions included in the network meta-analysis. Each node represents an intervention (HIIT, yoga, and active control) [25,28,31,36,37,42], with the node size proportional to the number of participants receiving that treatment. The connecting lines indicate direct comparisons between interventions, and their thickness corresponds to the number of studies comparing the respective treatments.
Figure 3. Network plot of exercise interventions included in the network meta-analysis. Each node represents an intervention (HIIT, yoga, and active control) [25,28,31,36,37,42], with the node size proportional to the number of participants receiving that treatment. The connecting lines indicate direct comparisons between interventions, and their thickness corresponds to the number of studies comparing the respective treatments.
Healthcare 14 01703 g003
Healthcare 14 01703 g004
Figure 4. Prediction interval of the effects of exercise modalities on HbA1c reduction.
Figure 4. Prediction interval of the effects of exercise modalities on HbA1c reduction.
Healthcare 14 01703 g004
Table 1. Characteristics of included articles in the preliminary meta-analysis.
Table 1. Characteristics of included articles in the preliminary meta-analysis.
No.Author/Year/CountryDesign/Study PopulationIntervention Group/Mean AgeExercise Dose (Duration and Frequency)IntensityMonitoring of Exercise IntensitySample SizeOutcome (HbA1c) ChangesMain Findings
1Abdelbasset et al. (2020)/Egypt [20]RCT/T2DMHIIT (Cycling Ergometer)
Age (54.4 ± 5.8) 		8 weeks: 3 times/week
40 min/session		 80–85% VO2max (intense effort)
50% VO2max (light cycling)		VO2max16BaselineAfter 8 weeksBoth the high-intensity (HII) and moderate-intensity (MIC) exercise groups showed reduced liver fat and belly fat.
There was no clear difference between HII and MIC in their effects.
6.6 ± 0.46.2 ± 0.3
Control
Age (55.2 ± 4.3) 		Receive medical treatment without exercise intervention 166.7 ± 0.66.5 ± 0.5
2Ahmad et al. (2023), Egypt [21]Randomized Control Parallel/Women with T2DMlow-volume HIIT (Treadmill exercise)
Age (42.96 ± 5.87)		12 weeks: 3 sessions/week, 19 min/sessionWarm-up: 65–70% HRpeak
Interval: 85–90% HRpeak
Recovery: 65–75% Hrpeak		HR24BaselineAfter 12 weeksBoth low- and high-volume HIIT showed improvements in TC, HDL, SBP, DBP, BMI, WC, and waist-to-hip ratio (p < 0.05).
The high-volume HIIT group showed more significant improvements in HbA1c, FBG, 2 h PPBG, TG, LDL,
(p < 0.05)
8.15 ± 0.527.12 ± 0.49
High-volume HIIT (treadmill exercise)
Age (43.29 ± 6.20)		12 weeks: 3 sessions/week, 16 min/sessionWarm-up: 65–70% HRpeak
Interval: 85–90% HRpeak
Recovery: 65–75% Hrpeak		248.15 ± 0.566.65 ± 0.17
Usual care (non-exercising control group
Age (42.46 ± 5.57)		 248.14 ± 0.558.19 ± 0.50
3Cassidy et al. (2019)/UK [29]RCT/T2DMHIIT (Cycling Ergometer)
Age (60 ± 3)		12 weeks: 3 sessions/week, 30–40 min/sessionRPE 16–17 during intervalsBorg RPE scale11Baseline12 weeksHIIT group improved HbA1c (from 7.13% to 6.87%), while the control group worsened (from 7.18% to 7.36%), p = 0.03.
7.13 ± 0.316.87 ± 0.29
Control
Age (59 ± 3)		Usual careMaintain their normal routine and do not change their medication, physical
activity, diet or body weight.		 117.18 ± 0.171.36 ± 0.21
4* Findikoglu et al. (2023)/Turkey [36]a single-blinded, 3-arm, randomized, controlled prospective study/Individuals with T2DM for less than 10 years but more
than 1 year		HIIT (Electromagnetic bicycle)
Age (57.5 ± 7.82)		12 weeks: 3 sessions/week, 24 min in week (1–4) with 8 cycles, 36 min in week (5–8) with 12 cycles, and 48 min in week (9–12) with 16 cycles.high-intensity (90% VO2peak for 60 s)
low-intensity (30% VO2peak for 120 s)		VO2max20Baselineafter 12 weeksBoth HIIT and MICT improved VO2peak and HbA1c after
12 weeks of training. Only MICT caused additional
improvements in cardiovascular responses, anthropometric measures, and abdominal fat
compared to baseline (p < 0.05).
6.9 ± 0.686.59 ± 0.49
Control
Age (55.75 ± 8.56)		Doing simple static
stretches of major muscle groups 		206.99 ± 0.666.76 ± 0.66
5* Gupta et al. (2020)/India [25]RCT/Individuals with T2DMYoga (asanas, kriyas, pranayama, and meditation)
Age (50.6 ± 8.5)		16 weeks: Training phases: 3 sessions/week (weeks 1–2);
Supervision phase: 2 sessions/week (weeks 3–4); Maintenance phase: 1 session/month (months 2–4).
One Session/45 min.		 40BaselineAfter 16 weeksAn HbA1c drop of ≥0.5% was observed in 44.7% of YBEP participants, while 37.5% in usual care. ≥75% of attending YBEP showed a 0.3% drop, compared to 0.1% for lower attendance.
8.53 ± 0.718.31 ± 1.32
Control
Age (50.6 ± 8.5)		usual care (Dietary counseling plus 30 min walking (5–6 km/h), ≥5 days/week.) 418.39 ± 0.658.38 ± 1.37
6Zheng et al. (2019)/China [18]RCT/T2DM high-intensity Progressive Resistance
Training (using non-conventional equipment)
Age (65.66 ± 8.58)		24 weeks: 1 session/week, 5–10 min/session
Four isometric exercises (chest press, leg press, core pull, vertical lift) are supervised by a qualified trainer		Isometric contractions performed against resistance scaled to body weight, with progressive load increases.Multiples of body weight (MOBs) generated during the isometric contractions165BaselineAfter 24 weeksThere were no significant changes in HbA1c between the control and PRT groups overall.
The intervention group had significant improvements in HDL and LDL
6.83 ± 1.316.75 ± 0.93
Control
Age (66.72 ± 6.68)		Participants maintained their usual medical care, dietary habits, and lifestyle. 1006.92 ± 1.266.85 ± 1.17
7* Hirosaki et al. (2023)/Japan [37]RCT/People with T2DMLaughter Yoga Program (warm-up, deep breathing, laughter exercises, and calming activities)
Age (71.8 ± 6.4))		12 weeks: Mini-lecture (30 min) + laughter yoga (60 min); 1 session/week (weeks 1–4), then 1 session/2 weeks (weeks 5–12). 21BaselineAfter 12 weeksLaughter yoga group had a significant HbA1c reduction (−0.31%, 95% CI: −0.54 to −0.09) with increased positive affect scores (+0.62, 95% CI: 0.003 to 1.23).
7.07 ± 0.76.82 ± 0.6
Control
Age (70.6 ± 8.2)		Oral hypoglycemic medications, physician advice, and physical activity guidance under Japan’s Treatment Guide for Diabetes 217.19 ± 0.77.26 ± 0.7
8Hwang et al. (2019)/USA [39]RCT/T2DMHIIT (Cycling Ergometer)
Age (65 ± 2)		8 weeks: 4 sessions/week, 40 min/session90% HRpeak (4 × 4 min intervals)
70% HRpeak (3 × 3 min recovery)		HR23BaselineAfter 8 weeksNo improvement has been found in glycemic control, lipid profile, or blood pressure. However, both intervention groups improved in aerobic fitness.
7.1 ± 0.36.8 ± 0.2
Control
Age (61 ± 2)		Followed regular physical activity, dieting, and meditation 167.4 ± 0.47.5 ± 0.4
9* Li et al. (2022)/China [31] parallel randomized
controlled clinical trial/T2DM patients		HIIT (Swedish Monark power bike)
Age (38 ± 6)		12 weeks: 5 sessions/week, 30 min/session80–95% HRmax/VO2peakHR13BaselineAfter 12 weeksThe MICT group showed significant weight loss (difference = 3.52, p < 0.01) and improved FBG (p < 0.05). BMI significantly decreased within groups (p < 0.01), though not between groups. HbA1c levels improved significantly over time (p < 0.01), but not between groups. SBP improved significantly in both HIIT and MICT groups compared to the control (p < 0.05).
7.18 ± 0.506.79 ± 0.41
Control
Age (40 ± 7)		Received standard counseling on conventional T2DM exercise guidelines 127.06 ± 0.387.09 ± 0.33
10Ma et al. (2024)/China [34]Single-blinded RCT/People diagnosed with T2DMRT (using dumbbells, resistance bands, and kettlebells)
Age (66.65 ± 4.94)		24 weeks: 3 times/week, 50 min/time40–50% of 1-RM during initial training, progressively increasing by 5–10% up to 60–70% of 1RM.% of 1-RM31Baseline24 weeksFasting Plasma Glucose (FPG), HbA1c, blood lipids, diastolic blood pressure, body composition, and muscle performance significantly improved in both exercise groups compared to the control group and their own baseline (p < 0.05).
7.80 ± 0.937.11 ± 0.75
blood flow restrictive resistance exercise group (using KAATSU Air Bands)
Age (66.41 ± 4.97)		24 weeks: 3 times/week, 50 min/time20–30% of one-repetition maximum (1RM) combined with blood flow restriction pressure sufficient to partially restrict arterial inflow.% of 1-RM for load resistance and limb occlusion pressure (LOP) for BFR347.75 ± 0.977.24 ± 0.85
Control
Age (65.55 ± 4.41)		Usual care (followed a diabetes management program but received no formal training and continued their usual self-selected diet and exercise habits.) 337.98 ± 0.967.93 ± 0.75
11* Viswanathan et al. (2021)/India [28]RCT/T2DMYoga (loosening exercises, asanas, pranayama, and relaxation techniques, along with diabetes education)
Age (50.8 ± 8.3)		12 weeks: 5 days/week, 50 min/day 150BaselineAfter 12 weeksYoga group lowers BMI, blood sugar, HbA1c, cholesterol, and inflammation markers.
7.5 ± 0.57.2 ± 0.9
Non-Yoga
Age (52.8 ± 7.0)		Simple physical exercises for 50 min per session, 5 days per week. 1507.5 ± 0.67.6 ± 1.1
12* Way et al. (2020)/Australia [42]RCT/T2DMHIIT (Cycling Ergometer)
Age (56.9 ± 2.1)		12 weeks: 3 sessions/week, 19 min/sessionEach session with 4 min of high-intensity cycling at 90% VO2peak% of VO2peak12BaselineAfter 12 weeksAfter the intervention of HIIT and MICT, significant improvements were found in VO2peak (p < 0.01), HbA1c (p = 0.03), systolic blood pressure (p< 0.01), and waist size (p = 0.03).
7.1 ± 1.56.8 ± 0.9
Control
Age (51.9 ± 1.4)		Sham Exercise Placebo (Stretching and core exercises every 2 weeks; ≤30 min/session with 5 min light cycling (20 W) warm-up and cool-down.) 117.6 ± 0.58.0 ± 0.5
13Yamamoto et al. (2021)/Japan [38]Prospective RCT/T2DMResistance exercise (using an elastic band)
Age (73.2 ± 2.6)		48 weeks of daily exercise: one time/15 minBodyweight exercises combined with progressive elastic band resistance (TBB-1 to TBB levels) representing increasing resistance levels.Daily exercise logs (adherence tracking).18BaselineAfter 48 weeksNo changes in physical function, muscle mass, or cognitive function in any group. The leucine supplement brought no extra benefits for muscle strength or mass.
7.4 ± 0.97.3 ± 0.8
Control
Age (73.3 ± 2.5)		Maintain daily activities 177.0 ± 0.76.8 ± 0.6
* articles included in NMA.
Table 2. Results of the random-effects network meta-analysis consistency model for HbA1c Change.
Table 2. Results of the random-effects network meta-analysis consistency model for HbA1c Change.
ComparisonMean Difference (MD)95% CIp-Value
HIIT vs. Active control−0.322−0.559, −0.0840.008
Yoga vs. Active control−0.366−0.534 to −0.198<0.001
HIIT vs. Yoga (indirect)−0.044−0.335 to 0.2470.766
Heterogeneity parameters: between-study standard deviation (τ): 2.91 × 10−12; assumed between-study correlation: 0.50.
Table 3. League table of network meta-analysis for HbA1c change.
Table 3. League table of network meta-analysis for HbA1c change.
TreatmentActive ControlHIITYoga
Active control—0.322 (0.084 to 0.559)0.366 (0.198 to 0.534)
HIIT−0.322 (−0.559 to −0.084)—−0.044 (−0.335 to 0.247)
Yoga−0.366 (−0.534 to −0.198)0.044 (−0.247 to 0.335)—
Values are presented as mean differences (MDs) with 95% confidence intervals. Negative values indicate greater reductions in HbA1c.
Table 4. Ranking probabilities and SUCRA values for interventions.
Table 4. Ranking probabilities and SUCRA values for interventions.
RankInterventionProbability of
Being Best		Probability of
Being Second		Probability of
Being Third		SUCRA
1Yoga1.00.00.01.0
2HIIT0.01.00.00.5
3Active control0.00.01.00.0
SUCRA: surface under the cumulative ranking curve.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Thu, S.Y.W.; Patnaik, S.; Shih, Y.-H. Yoga and High-Intensity Interval Training Show Comparable Effects on HbA1c in Type 2 Diabetes: A Systematic Review and Preliminary Pilot Network Meta-Analysis in Adult Populations. Healthcare 2026, 14, 1703. https://doi.org/10.3390/healthcare14121703

AMA Style

Thu SYW, Patnaik S, Shih Y-H. Yoga and High-Intensity Interval Training Show Comparable Effects on HbA1c in Type 2 Diabetes: A Systematic Review and Preliminary Pilot Network Meta-Analysis in Adult Populations. Healthcare. 2026; 14(12):1703. https://doi.org/10.3390/healthcare14121703

Chicago/Turabian Style

Thu, Saw Ye Win, Sneha Patnaik, and Yin-Hwa Shih. 2026. "Yoga and High-Intensity Interval Training Show Comparable Effects on HbA1c in Type 2 Diabetes: A Systematic Review and Preliminary Pilot Network Meta-Analysis in Adult Populations" Healthcare 14, no. 12: 1703. https://doi.org/10.3390/healthcare14121703

APA Style

Thu, S. Y. W., Patnaik, S., & Shih, Y.-H. (2026). Yoga and High-Intensity Interval Training Show Comparable Effects on HbA1c in Type 2 Diabetes: A Systematic Review and Preliminary Pilot Network Meta-Analysis in Adult Populations. Healthcare, 14(12), 1703. https://doi.org/10.3390/healthcare14121703

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop