Abstract
Objective:
To systematically investigate and evaluate the effects of different exercise (TT: treadmill training; ST: sensorimotor training; NW: Nordic walking; CT: cycling training; WT: walking training; DT: dance training; VRT: VR training; RT: resistance training; AQT: aquatic training; BOX: boxing training; Qigong and Yoga) modalities on both motor and non-motor abilities in patients with Parkinson’s disease.
Methods:
Randomized controlled trials (RCTs) concerning the impact of various exercise modes on motor and non-motor abilities in Parkinson’s patients were identified by searching the PubMed, Web of Science, Cochrane library, CINAHL and CNKI databases. The search period spanned from the inception of each database to December 2025. The methodological quality of the included studies was assessed using the Cochrane Risk of Bias tool.
Results:
A total of 67 RCTs involving 2642 patients were included. Network meta-analysis results indicated that WT most effectively improved UPDRS-III (SUCRA=92.5%); DT was most effective in improving TUG (SUCRA=89.9%) and 6MWT (SUCRA=73.3%); NW best improved BBS (SUCRA=91.1%) and MoCA (SUCRA=91.1%). AQT best improved PDQ-39 (SUCRA=88.5%).
Conclusion:
Current evidence suggests that exercise involving walking and dance, which stimulates neural regulation, can help improve motor function and cognitive abilities in people with Parkinson’s disease. Patients can choose exercises based on their current fitness level.
Systematic review registration:
1 Introduction
The proportion of the global elderly population continues to rise, accompanied by an increasing burden of age-related diseases. Notably, the incidence of Parkinson’s disease (PD) shows a significant positive correlation with age (Schootemeijer et al., 2020), and its steadily rising prevalence poses a major challenge to public health systems worldwide. PD occurs more frequently in middle-aged and elderly populations, with its typical clinical manifestations including progressive cognitive impairment and memory decline. From a pathological perspective, Parkinson’s disease is characterized by the progressive degeneration and death of dopaminergic neurons in the substantia nigra pars compacta of the midbrain, accompanied by the deposition of Lewy bodies composed of alpha-synuclein within these neurons (Schootemeijer et al., 2020). This leads to characteristic motor impairments, including bradykinesia, resting tremor, rigidity, and postural instability. As the disease progresses, pathological involvement extends to the cerebral cortex and other neurotransmitter systems, subsequently triggering a range of non-motor symptoms such as comprehensive impairment in cognitive function and executive function (Zhang et al., 2018). Based on these characteristics, assessments of cognitive function, emotional state, and motor performance have become crucial indicators for PD clinical interventions.
Since there is currently no cure for Parkinson’s disease, and drug treatments carry a range of side effects (Duncker and Bache, 2008), exercise interventions have gained increasing attention in recent years as a complementary approach to PD management. Numerous studies indicate that regular physical activity can reduce the risk of developing PD and positively impact both motor and non-motor symptoms in patients. These benefits include improved motor function, reduced risk of falls, enhanced cognitive abilities and increased overall quality of life (Ernst et al., 2023; Lei et al., 2025).
However, the improvement effects achieved by different exercise programs may vary. Most current meta-analyses focus on specific exercise types (Zhou et al., 2022) and fail to systematically evaluate all exercise modalities. Although the American Academy of Neurology (AAN) has highlighted the benefits of Tai Chi in enhancing patients’ balance (Gladfelter, 2011), and the Chinese Association of Rehabilitation Medicine recommends traditional exercises (such as Tai Chi and Baduanjin) as exercise interventions for Parkinson’s disease while increasing cultural adaptability, their comparative metrics primarily emphasize patients’ motor performance. (Pang et al., 2022) For Parkinson’s patients, quality of life and cognitive function are equally critical, as non-motor symptoms can directly erode psychological wellbeing and social engagement (Malkki, 2013). To address these gaps, this study employs a network meta-analysis to systematically compare the effects of different exercise modalities on motor performance, cognition and quality of life in PD patients. This review aims to provide evidence-based recommendations for personalized and comprehensive exercise intervention strategies.
2 Materials and methods
The study was conducted, and the manuscript was prepared in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Page et al., 2021). The study was registered with Open Science Framework, https://osf.io/7x5kc/.
2.1 Search strategy and study selection
We conducted a systematic literature search across multiple databases including PubMed, Web of Science, Cochrane library, CINAHL and CNKI. The search scope covered the period from each database’s inception to December 2025. The search strategy combined Medical Subject Headings (MeSH) terms with free-text keywords, finalized after multiple preliminary tests. Additionally, manual searches and review of reference lists from included studies were employed to identify further relevant research. Search terms included “Parkinson’s disease,” “exercise,” “movement,” “athletic ability,” “Balance,” “cognition,” “depression,” and “quality of life,” combined using Boolean operators. Two reviewers (JF and XH) independently screened titles and abstracts to identify potentially relevant articles, followed by full-text review of selected documents. Additional relevant articles were identified by consulting the reference lists of articles retrieved from the initial database search. Figure 1 illustrates the flowchart describing the literature selection process.
Figure 1
Search results were imported into Zotero software (version 6.0.30) to remove duplicates. Two researchers conducted an initial assessment of the titles and abstracts of the retrieved literature, then read the full texts for further screening. When disagreements arose during article screening, a third researcher made the final decision.
2.2 Inclusion and exclusion criteria
Inclusion criteria were established based on the PICO framework: (1) All participants had been diagnosed with Parkinson’s disease by a neurologist, based on the clinical diagnostic criteria established by the Movement Disorder Society (MDS) (Postuma et al., 2015); (2) The study was a randomized controlled trial (RCT); (3) The intervention was exercise-based; (4) Outcomes: Motor performance assessed via Unified Parkinson’s Disease Rating Scale (UPDRS III), Timed Up and Go Test (TUG), Berg Balance Scale (BBS), and 6-Minute Walk Test (6MWT); non-motor function evaluated using Parkinson’s Disease Quality of Life-39 (PDQ-39) and Montreal Cognitive Assessment (MoCA).
Exclusion criteria were as follows: (1) conference abstracts, case reports, commentaries, reviews, and animal studies; (2) outcome measures not meeting inclusion criteria; (3) incomplete data where mean and standard deviation could not be obtained from the literature or author communication.
2.3 Data extraction
Using seven standardized data extraction forms, record the following data from included studies under the following headings: (1) Authors, (2) Publication year, (3) Country, (4) Study period, (5) Sample size, (6) Mean age, (7) Exercise intervention details. (8) Outcomes: Motor performance assessed via Unified Parkinson’s Disease Rating Scale (UPDRS III), Timed Up and Go Test (TUG), Berg Balance Scale (BBS), and 6-Minute Walk Test (6MWT); non-motor function evaluated using Parkinson’s Disease Quality of Life-39 (PDQ-39) and Montreal Cognitive Assessment (MoCA).
2.4 Risk of bias of individual studies
The risk of bias in the included studies was assessed using the Cochrane risk of bias tool. Two researchers independently evaluated seven domains: random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), and selective reporting (reporting bias). Each study was categorized as “high risk,” “low risk,” or “uncertain risk.” Disagreements were resolved through discussion or consensus among a third researcher. Studies were classified into three groups: high risk (five or more risk factors), moderate risk (three to four risk factors), and low risk (two or fewer risk factors) (Higgins et al., 2011).
2.5 Data analysis
Using R software (RStudio V4.19, Boston, Massachusetts, USA), effect size pooling and comparisons were conducted within a Bayesian framework employing the Markov Chain Monte Carlo (MCMC) method, based on the PRISMA network meta-analysis guidelines (Shamseer et al., 2015). The random-effects NMA model was fitted using functions from the MCMCpack package. Consistency in direct and indirect comparisons was quantified and visualized using the node method, with consistency established when P-values > 0.05 (Salanti et al., 2011).
Network diagrams depicting different exercise interventions were generated and described using R software. Within the generated network diagrams, each node represented a distinct exercise intervention and control condition, while connecting lines indicated direct head-to-head comparisons between interventions. Node size and link width are proportional to the number of studies. Based on MCMC posterior samples, the probability of each intervention occupying different ranking positions (from best to worst) is calculated. The cumulative area under the ranking curve (SUCRA) value is then computed. SUCRA ranges from 0% to 100%, with higher values indicating a higher-ranked intervention (better efficacy). Although SUCRA values can be effectively converted into percentages representing the effectiveness or acceptability of exercise interventions, such scores should be interpreted cautiously unless clinically meaningful differences exist between interventions. To detect bias arising from small-scale studies, network funnel plots were generated and visually inspected according to symmetry criteria.
3 Results
See Table 1. This study included 67 RCT publications. A total of 2642 patients from 19 countries were enrolled. 59 articles reported patients’ Hoehn-Yahr staging. The vast majority of articles were published within the last decade, with 14 articles published within the last five years.
Table 1
| First author, year | country | Intervention-a | Intervention-b | Control | Sample size | years | Hoehn-Yahr | Outcome |
|---|---|---|---|---|---|---|---|---|
| (Altmann et al., 2016) | French | treadmill training 16 weeks Freq:three times a week Duration: 20-45min | BLT(balance training) | CON | 11 or 9/10 | 63.3 ± 7.3 | 1-3 | ① |
| (Sage and Almeida, 2009) | USA | treadmill training 12 weeks Freq:three times a week Duration: 20min | – | CON | 13/15 | 64.61±9.68 | 1-3 | ①③ |
| (Nadeau et al., 2014) | Canada | treadmill training 3 months Freq:three times a week Duration: 60min | – | CON | 30/34 | 64.0 ± 6.6 | 1-2 | ② |
| (Picelli et al., 2016) | Italy | treadmill training 4 weeks Freq:three times a week Duration: 45min | – | CON | 9/8 | 71.2 ± 9.2 | 3 | ②⑤ |
| (Carvalho et al., 2015) | USA | treadmill training 12 weeks Freq:two times a week Duration: 30min | resistance training 12 weeks Freq:two times a week Duration: 30min | BLT(balance training) | 5/8/9 | 64.4±11.7 | 1-3 | ① |
| (Canning et al., 2012) | Australia | treadmill training 6 weeks Freq:four times a week Duration: 30-40min | – | CON | 10/10 | 61.8±7.9 | 1-2 | ①②⑥ |
| (Arcolin et al., 2016) | Italy | treadmill training 3 weeks Freq:five times a week Duration: 60min | cycling training 3 weeks Freq:five times a week Duration: 60min | – | 18/13 | 68.32±8.38 | 1.5-3 | ①②③ |
| (Bello et al., 2013) | Spain | walking training 5 weeks Freq:three times a week 10m | treadmill training 5 weeks Freq:three times a week 10m | – | 11/11 | 58.75±10.17 | 1-3 | ①③ |
| (Mitarnun et al., 2022) | Thailand | walking training 12 weeks Freq:three times a week Duration: 30min | – | CON | 17/16 | 61.3 ± 9.1 | 1-3 | ①③ |
| (Cugusi et al., 2015) | Italy | Nordic Walking 12 weeks Freq:two times a week Duration: 60min | – | CON | 10/10 | 67.3 ± 7.8 | 1-3 | ①②③④ |
| (Granziera et al., 2020) | Italy | Nordic Walking 8 weeks Freq:two times a week Duration: 75min | walking training 8 weeks Freq:two times a week Duration: 75min | – | 18/18 | 68.5 ± 8.3 | 2-3 | ①③⑥ |
| (Bang and Shin, 2017) | Korea | Nordic Walking 5 weeks Freq:five times a week Duration: 60min | treadmill training 5 weeks Freq:five times a week Duration: 60min | – | 10/10 | 59.45±7.02 | 1-3 | ①③④ |
| (Sacheli et al., 2019) | Canada | cycling training 12 weeks Freq:three times a week Duration: 60min | – | CON | 20/15 | 66.76 ± 5.98 | 1-3 | ①③⑤ |
| (Harper et al., 2019) | America | cycling training 12 weeks Freq:three times a week Duration: 60min | – | CON | 20/15 | 65.05 ± 9.13 | – | ⑤ |
| (Qutubuddin et al., 2013) | USA | cycling training 8 weeks Freq:two times a week Duration: 30min | – | CON | 13/10 | – | – | ①④⑥ |
| (Ferraz et al., 2018) | Brazil | cycling training 10 activities Duration: 30min | – | CON | 25/22 | 63.2 ± 9.9 | 2-3 | ②⑥ |
| (Hashimoto et al., 2015) | Japan | Dance training 12 weeks Freq:one time a week Duration: 60min | – | CON | 15/14 | 67.9 ± 7.0 | 1-4 | ③④ |
| (Hackney and Earhart, 2009) | USA | Dance training-Tango 13 weeks Freq:two times a week Duration: 60min | – | CON | 14/17 | 67.27±2.4 | 1-4 | ②③④ |
| (Solla et al., 2019) | Italy | Dance training-Tango 12 weeks Duration: 90min | – | CON | 10/10 | 67.4 ± 6.1 | ≤3 | ①②③④ |
| (Rios Romenets et al., 2015) | Canada | Dance training-tango 12 weeks Freq:two times a week Duration: 60min | – | CON | 18/15 | 64.3 ± 8.1 | 1-3 | ①③⑤⑥ |
| (Michels et al., 2018) | USA | Dance training-Dance to the rhythm of music 10 weeks Duration: 60min | – | CON | 7/6 | 69.2 ± 8.7 | 2-2.5 | ①③④⑤⑥ |
| (Volpe et al., 2013) | Italy | Dance training-Irish set dancing 24 weeks Freq:one time a week Duration: 90min | – | CON | 12/12 | 63.3±5.11 | 1-2.5 | ①④⑥ |
| (Duncan and Earhart, 2012) | USA | Dance training-community-based dancing 12 weeks Freq:two times a week Duration: 60min | – | CON | 26/26 | 69.15±1.7 | 1-4 | ① |
| (Tollár et al., 2018) | Holland | VR training 3 weeks Freq:five times a week Duration:60 min | – | CON | 35/20 | 67.3 ± 3.4 | – | ①③⑥ |
| (Tollár et al., 2019) | Holland | VR training 5 weeks Freq:five times a week Duration:60 min | Cycling training 5 weeks Freq:five times a week Duration:60 min | CON | 25/24 | 67.5±4.28 | 2-3 | ②④⑥ |
| (Lee et al., 2015) | Korea | VR training 6 weeks Freq:two times a week Duration: 45min | – | CON | 10/10 | 69.25±3.15 | – | ④ |
| (Santos et al., 2019) | Brazil | VR training 8 weeks Freq:two times a week Duration: 50min | resistance training 8 weeks Freq:two times a week Duration: 50min | – | 13/14 | 63.15±8.64 | 1-3 | ③④⑥ |
| (van der Kolk et al., 2018) | Netherlands | VR training 6 months Freq:three times a week Duration: 30min | – | CON | 15/22 | 30-75 | 1-2 | ①③⑥ |
| (Song et al., 2018) | Australia | VR training 12 weeks Freq:three times a week Duration: Minimum 15 min | – | CON | 31/30 | 66.52±7.10 | 1-3 | ③⑤ |
| (Pazzaglia et al., 2020) | Italy | VR training 6 weeks Freq:three times a week Duration: 40min | – | CON | 25/26 | 72 ± 7 | 1-3 | ④ |
| (Yan et al., 2024) | China | resistance training 12 weeks Freq:three times a week Duration: 60min | – | CON | 23/23 | 70.5 ± 5.4 | ≤3 | ①⑤⑥ |
| (Amano et al., 2013) | USA | Qigong-Tai Chi 16 weeks Freq:two times a week Duration: 60min | – | CON | 15/9 | 66±9.52 | 1-3 | ① |
| (Choi, 2016) | Korea | Qigong-Tai Chi 12 weeks Freq:two times a week Duration: 60min | – | CON | 11/9 | 62.94±7.46 | 1-2 | ②③ |
| (Zhu et al., 2020) | China | Qigong-Tai Chi 12 weeks Freq:two times a week Duration: 40~50 minutes | – | CON | 19/22 | 68.53±1.90 | 2 ±2.2 | ①④⑤⑥ |
| (Xiao et al., 2016) | China | Qigong-baduanjin 6 months Freq:four times a week Duration: 60min | – | CON | 35/33 | 67.8±9.4 | 1-4 | ①②③④ |
| (Wang Xiangyu et al., 2021) | China | Qigong-baduanjin 6 weeks Freq:seven times a week Duration: 60min | – | BLT(balance training) | 27/24 | 65.52 ± 7.29 | 1-4 | ① |
| (Li et al., 2012) | USA | Qigong-Tai Chi 24 weeks Freq:two times a week Duration: 60min | resistance training 24 weeks Freq:two times a week Duration: 60min | BLT(balance training) | 65/65 | 68.11±8.85 | 1-4 | ①③ |
| (Khuzema et al., 2020) | India | Qigong-Tai Chi 8 weeks Freq:five times a week Duration: 30-40min | Yoga 8 weeks Freq:five times a week Duration: 30-40min | CON | 9/9 | 70.06±5.4 | 2.5-3 | ③④ |
| (Nocera et al., 2013) | USA | Qigong-Tai Chi 16 weeks Freq:three times a week Duration: 60min | – | CON | 15/6 | 66 ± 11 | 1-3 | ⑥ |
| (Vergara-Diaz et al., 2018) | USA | Qigong-Tai Chi 12 weeks Freq:two times a week Duration: 60min | – | CON | 16/16 | 63.85±6.32 | 1-2 | ①③⑥ |
| (Volpe et al., 2017) | Italy | Aquatic training 8 weeks Freq:five times a week Duration: 60min | – | BLT(balance training) | 15/15 | 70.3±7.67 | 1-3 | ①③④⑥ |
| (Kurt et al., 2018) | Turkey | Aquatic training 5 weeks Freq:five times a week Duration: 60min | – | BLT(balance training) | 20/20 | 63.02±6.91 | 2-3 | ①③④⑥ |
| (Volpe et al., 2014) | Italy | Aquatic training 8 weeks Freq:five times a week Duration: 60min | – | BLT(balance training) | 17/17 | 68±7.52 | 2.5–3 | ①③④⑥ |
| (vivas et al., 2011) | Spain | Aquatic training 4 weeks Freq:two times a week Duration: 45min | – | BLT(balance training) | 12/12 | 67±5.59 | 1-3 | ④ |
| (Carroll et al., 2017) | Italy | Aquatic training 6 weeks Freq:two times a week Duration: 45min | – | CON | 10/8 | 71.5±4.67 | 1-3 | ①⑥ |
| (Son and Choi, 2018) | Korea | resistance training 8 weeks Freq:one time a week Duration: 120 min | – | CON | 33/30 | 66.5 ± 5.7 | 1-3 | ⑤ |
| (Schlenstedt et al., 2015) | Germany | resistance training 7 weeks Freq:two times a week Duration: 60min | – | BLT(balance training) | 17/15 | 75.7 ± 5.5 | 1-3 | ①③ |
| (Santos et al., 2017) | spain | resistance training 8 weeks Freq:two times a week Duration: 60-70min | – | CON | 13/15 | 73.6±15.86 | 1-3 | ①⑥ |
| (Paul et al., 2014) | Australia | resistance training 12 weeks Freq:two times a week Duration: 45min | – | CON | 20/20 | 66.3±6.73 | – | ③ |
| (Ni et al., 2016b) | USA | resistance training 12 weeks Freq:three times a week Duration: 3 circuits of 10e12 repetitions on each machine | – | CON | 10/14 | 73.3±7.45 | 1-3 | ⑥ |
| (Goodwin et al., 2011) | UK | resistance training 10 weeks Freq:two times a week Duration: 50min | – | CON | 61/63 | 72.0 ± 8.6 | 1-4 | ③④ |
| (Ferreira et al., 2018) | Brazil | resistance training 6 month Freq:two times a week Duration:30-40min | – | CON | 18/17 | 65.9±7.95 | 1-3 | ①⑥ |
| (Dibble et al., 2009) | USA | resistance training 12 weeks Freq:three times a week Duration: 45-60min | – | CON | 10/10 | 65.7±9.9 | 2-3 | ①③⑥ |
| (Dibble et al., 2015) | USA | resistance training 12 weeks Freq:two times a week Duration: 50min | – | BLT(balance training) | 20/21 | 68.4±11.95 | – | ①② |
| (Demonceau et al., 2017) | Belgium | resistance training 12 weeks Freq:two times a week Duration: 60-90min | cycling training 12 weeks Freq:two times a week Duration: | CON | 20/17/15 | 66±9 | 2-3 | ③⑥ |
| (Corcos et al., 2013) | USA | resistance training 24 months Freq:two times a week | – | BLT(balance training) | 24/24 | 59.0 ± 4.6 | – | ①⑥ |
| (Vieira de Moraes Filho et al., 2020) | Brazil | resistance training 9 weeks Freq:two times a week Duration: 50-60min | – | CON | 25/15 | 64.6±5.5 | 1-3 | ①③ |
| (Ni et al., 2016a) | USA | Yoga 12 weeks Freq:two times a week Duration: 60min | – | CON | 13/10 | 71.2 ± 6.5 | 1-3 | ⑥ |
| (Kwok et al., 2019) | China | Yoga 8 weeks Freq:one time a week Duration: 90min | resistance training 8 weeks Freq:one time a week Duration: 90min | – | 71/67 | 63.6 ± 8.7 | 1-3 | ①③ |
| (Cheung et al., 2018) | USA | Yoga 12 weeks Freq:two times a week Duration: 60min | – | CON | 10/10 | 64.65±7.66 | 1-3 | ①⑤ |
| (Combs et al., 2013) | USA | boxing training group 12 weeks 24–36 sessions Duration: 60min | resistance training 12 weeks 24–36 sessions Duration: 60min | – | 14/17 | 68.0 ± 31.0 | 2.0 ± 3.0 | ②③④⑥ |
| (Schenkman et al., 2012) | USA | sensorimotor training 4 months Freq:three times a week Duration: 50min | treadmill training 4 months Freq:three times a week Duration: 50min | CON | 39/41/41 | 64.5 ± 10.0 | 1-3 | ①⑥ |
| (Tanaka et al., 2009) | Brazil | sensorimotor training 24 weeks Freq:three times a week Duration: 60min | – | CON | 10/10 | 65.4 ± 7.23 | 1-3 | ① |
| (Hubble et al., 2018) | Australia | sensorimotor training 12 weeks Freq:one time a week Duration: 90min | – | CON | 11/11 | 65.4 ± 5.7 | 1-3 | ①③⑥ |
| (Moratelli et al., 2025) | Denmark | sensorimotor training 12 weeks Freq:two times a week Duration: 60min | Yoga 12 weeks Freq:two times a week Duration: 60min | CON | 11 or 8/9 | 63.3 ± 8.9 | – | ⑤ |
| (Cholewa et al., 2013) | Poland | sensorimotor training 12 weeks Freq:two times a week Duration: 60min | – | CON | 40/30 | 70.2 ± 5.75 | 3 | ①⑥ |
| (Acarer et al., 2015) | Turkey | sensorimotor training 8 weeks Freq:two times a day Duration: 30-40min | – | CON | 29/11 | 65.08 ± 8.09 | 2-3 | ①③④⑥ |
Characteristics of the studies included in the meta-analysis.
Outcome measures include: ①UPDRS-III ②6MWT ③TUG test ④BBS ⑤MoCA ⑥depression scale ⑦PDQ-39; “—” indicates not reported.
3.1 Risk bias assessment
According to the study by Higgins and Green, two researchers assessed the risk of bias for each included study (Figure 2). A total of 67 articles were included in this study. All articles employed random allocation methods, provided detailed baseline characteristics of study participants, and clearly described intervention methods and assessment indicators. All studies reported information on allocation concealment.
Figure 2
3.1.1 Motor abilities
motor performance was assessed using four measures (UPDRS-III, TUG test, BBS, and 6MWT). Among the 44 included studies assessing motor performance via UPDRS-III, all direct and indirect comparisons underwent consistency and inconsistency testing. With 15 total comparisons, 1 comparison showed significant inconsistency (6.7% inconsistency rate), specifically between CON and TT. Although minor inconsistencies existed within the network, overall consistency was acceptable, indicating good network consistency.
Network meta-analysis results indicated that compared with conventional interventions in the control group, WT [MD = -8.46, 95% CI = (-13.62, -3.13)], DT [MD = -7.96, 95% CI = (-11.62, -4.26)], NW [MD = -6.41, 95% CI = (-11.62, -4.26)], and TT [MD = -3.31, 95% CI = (-6.14, -0.52)] were significantly lower than the control group in reducing UPDRS-III scores, as detailed in Supplementary Table 1. With regard to the UPDRS-III score, previous studies have shown that the MCID range for PD typically falls between 2 and 5 points (Mishra et al., 2024). In this study, all of the aforementioned exercise interventions achieved the established MCID threshold for the UPDRS-III, suggesting that these interventions not only demonstrated statistical significance but also resulted in clinically perceptible improvements in motor function for patients. Probability-based ranking of different exercise interventions showed WT ranked highest in SUCRA analysis (92.5%), followed by DT (91.2%), as illustrated in Figure 3.
Figure 3
The 34 included studies evaluated the TUG. All direct and indirect comparisons underwent consistency and inconsistency tests, yielding a total of 12 comparisons. One comparison showed significant inconsistency (8.3% inconsistency rate), specifically the CON vs RT comparison. Although minor inconsistencies existed within the network, overall consistency was acceptable, indicating good network consistency.
The network meta-analysis results indicated that compared with conventional interventions in the control group, DT [MD = -3.11, 95% CI = (-4.91, -1.30)], VRT [MD = -2.48, 95% CI = (-4.53, -0.43)], and qigong [MD = -1.54, 95% CI = (-3.08, -0.03)] significantly reduced TUG scores compared to the control group. See Supplementary Table 2 for details. With regard to TUG scores, previous studies have shown that the MCID range for PD is typically 1.7 seconds or more (Mishra et al., 2024). In this study, DT and VRT achieved an improvement in TUG scores that met the established MCID threshold, indicating that this intervention can provide patients with clinically meaningful improvements in motor function. Probabilistic ranking of different exercise interventions showed DT ranked first in SUCRA analysis (89.9%), as illustrated in Figure 4.
Figure 4
The 25 included studies evaluated BBS. All studies underwent consistency and inconsistency tests for direct versus indirect comparisons, with all p-values exceeding 0.05. This indicates acceptable overall consistency among studies and good network consistency.
The network meta-analysis results showed that compared with conventional interventions in the control group, NW [MD = 8.43, 95% CI = (2.98, 13.80)], DT [MD = 6.25, 95% CI = (3.76, 8.75)], and VRT [MD = 5.36, 95% CI = (2.64, 8.15)], qigong [MD = 3.26, 95% CI = (0.33, 6.19)] were significantly lower than the control group in improving BBS scores, as detailed in Supplementary Table 3. With regard to BBS scores, previous studies have indicated that the MCID range for PD is typically 3 points or more (Miyata et al., 2021). In this study, all four interventions achieved the established MCID threshold for BBS improvement, indicating that these interventions can produce clinically perceptible improvements in motor function for patients. Probabilistic ranking of different exercise interventions showed NW ranked first in SUCRA analysis (91.1%), as illustrated in Figure 5.
Figure 5
The 14 included studies evaluated the 6MWT. All studies underwent consistency and inconsistency tests for direct and indirect comparisons, with all P-values exceeding 0.05. This indicates acceptable overall consistency among all studies and good network consistency.
The network meta-analysis results showed that compared with conventional interventions in the control group, no exercise intervention was significantly less effective than the control group in improving 6MWT scores, as detailed in Supplementary Table 4. Probabilistic ranking of different moderate-intensity exercise interventions revealed that DT ranked first in the SUCRA analysis (73.3%), as illustrated in Figure 6.
Figure 6
3.1.2 Non-motor abilities
Non-motor abilities are primarily assessed in terms of cognitive function and quality of life. The Montreal Cognitive Assessment (MoCA) serves as a core tool for screening cognitive function in PD, and its use is supported by a solid evidence-based foundation and broad international consensus (Camargo et al., 2016). Sixteen studies employed the MoCA to assess cognitive function in Parkinson’s patients. All studies underwent consistency and inconsistency tests for direct and indirect comparison P-values, with all P-values exceeding 0.05. This indicates acceptable overall consistency among all studies and good network consistency.
The network meta-analysis results showed that compared with conventional interventions in the control group, RT [MD = 2.91, 95% CI = (1.28, 4.53)] and DT [MD = 1.99, 95% CI = (0.34, 3.61)] were significantly higher than the control group in improving MOCA scores, as detailed in Supplementary Table 5. Probability-based ranking of different moderate-intensity exercise interventions showed NW ranked first in the SUCRA analysis (91.1%), as illustrated in Figure 7.
Figure 7
Thirty-five studies used the PDQ-39 scale to assess quality of life in Parkinson’s disease patients. All studies underwent consistency and inconsistency tests for direct versus indirect comparisons, with all P-values exceeding 0.05. This indicates acceptable overall consistency among studies and good network consistency.
The network meta-analysis results showed that compared with conventional interventions in the control group, no exercise intervention was significantly less effective than the control group in improving PDQ-39 scores, as detailed in Supplementary Table 6. Probability-based ranking of different exercise interventions showed AQT ranked first in the SUCRA analysis (88.5%), followed by BOX (80.2%). See Figure 8 for details.
Figure 8
3.2 Publication bias test
Funnel plots were constructed for all outcome measures to examine potential publication bias. The scatter plots exhibited broadly symmetrical distributions, indicating no significant publication bias (Supplementary Figure 1).
4 Discussion
This study systematically compared the effects of different exercise interventions on improving motor and non-motor abilities in patients with Parkinson’s disease (PD). A total of 83 exercise intervention studies were included, encompassing 11 distinct exercise modalities, with 2,642 PD patients participating—representing a substantial sample size. The results indicated that for reducing UPDRS-III scores, WT was the most effective exercise intervention, followed by DT. The optimal intervention for improving BBS scores was NW, followed by DT. DT was also the best modality for reducing TUG scores and increasing 6MWT scores. Regarding non-motor abilities, NW was the most effective intervention for improving MoCA scores, followed by DT. AQT was the best for improving PDQ-39 scores. In summary, while DT appears to be the most effective single intervention for improving motor performance based on overall analysis, NW demonstrated the highest number of indicators showing significant improvement across both motor and non-motor abilities.
Motor abilities is a critical clinical indicator for assessing recovery in PD patients. The most common symptoms of PD involve a significant decline in muscular control, including impaired balance, reduced muscular endurance, and diminished dynamic mobility (Mancini and Horak, 2016). Therefore, a comprehensive evaluation of motor performance requires multiple metrics, as relying solely on the UPDRS-III scale is often insufficient. This review employed four indicators—UPDRS-III, 6MWT, TUG test, and BBS—to assess motor performance, analyzing it from perspectives of motor impairment severity, endurance, balance, and more. For PD patients, interventions like DT, which require high neural involvement and regulation, emphasize cognitive participation and feedback mechanisms. This activates brain regions associated with movement, promotes neural plasticity, and thereby improves motor abilities (Petzinger et al., 2013). Current research suggests the primary mechanism involves enhancing neural plasticity to restore motor abilities, potentially by remodeling the morphological structure and function of dendritic spines on medium spiny neurons (MSNs) in the striatum of PD models. This may involve upregulating the expression of GluR2 subunit-containing AMPARs in the striatum, reducing calcium ion content in MSN dendritic spines, inhibiting the activity of calcium-dependent proteases and the initiation of downstream intracellular signaling cascades. Consequently, the overactivation of the striatal glutamatergic pathway is suppressed, mitigating a series of pathological changes caused by striatal overactivation in PD from the perspective of neural plasticity (VanLeeuwen et al., 2010; Toy et al., 2014; Shin et al., 2016). This leads to improved static and dynamic balance in patients, as evidenced by significant improvements in BBS and TUG scores. Furthermore, MCID threshold analysis of the TUG indicates that DT produces highly clinically significant improvements in motor function.
Furthermore, studies indicate that such exercise can also increase antioxidant capacity in PD, potentially by strengthening oxidative defense and free radical scavenging. This may prevent the release of cytochrome c from the mitochondrial respiratory chain and apoptosis-inducing factor from the mitochondrial matrix into the cytoplasm, inhibiting the initiation of apoptotic cascades and the expression of inflammatory genes. This curbs inflammatory damage to dopaminergic neurons in the substantia nigra, thereby ameliorating motor dysfunction in PD (Tajiri et al., 2010). Therefore, exercise can indeed significantly improve the motor performance of PD patients across multiple dimensions, and comprehensive analysis suggests DT is the most effective modality for this purpose.
Non-motor abilities is also a major focus in contemporary PD research. Cognitive impairment, as one of the common non-motor symptoms (Goldman and Postuma, 2014), not only severely impacts patients’ quality of life but also imposes a heavy burden on families and society. This review used the MoCA scale to assess cognitive function, and the results showed that DT could also significantly improve cognitive function in PD patients. Several studies have focused on the benefits of aerobic exercise (AE) for improving cognition in PD (Johansson et al., 2022). However, a meta-analysis indicated that while AE showed a trend toward improving cognitive function in PD patients, the difference was not statistically significant (Tang et al., 2023). This may be because cognitive dysfunction is an independent risk factor in PD, and the cognitive benefits of exercise might not be sufficient to counteract the progression of the disease itself (van der Kolk et al., 2019). Additionally, there may be a ceiling effect for aerobic exercise on cognitive improvement, making it suboptimal for addressing PD-related cognitive decline. In contrast, DT promotes neural plasticity and improves functional connectivity in the brain, particularly yielding positive effects on executive function and attention (Combs et al., 2011; Corcos et al., 2012), thereby enhancing cognitive performance.
The PDQ-39, extensively validated, is commonly used to assess patients’ health-related quality of life (Martinez-Martin and Kurtis, 2012). Mitigating the negative impact on health-related quality of life is a core objective in the management of Parkinson’s disease. This study indicates that no single form of exercise can significantly improve patients’ quality of life, which is related to the complexity and multidimensional nature of quality-of-life assessment in Parkinson’s disease. The PDQ-39 covers eight dimensions: mobility, activities of daily living, emotional state, stigma, social support, cognitive function, communication, and physical discomfort (Jones et al., 2014). Any single form of exercise may only have a significant effect on specific dimensions, while its impact on other dimensions may be less pronounced, resulting in statistically non-significant improvements in overall quality of life—a finding highly consistent with the analysis in this paper (Lacy et al., 2023). Therefore, it is particularly important to explore specialized, multidimensional exercise interventions in the future to improve patients’ quality of life.
In summary, DT appears to be an excellent option for improving patients’ motor and cognitive abilities. Furthermore, walking-based exercises (WT and NW) also achieved the recognized MCID thresholds for overall motor function, balance, and cognition. This suggests that the aforementioned exercise interventions not only yield statistically significant differences but also provide patients with clinically perceptible improvements in motor function. Therefore, future exercise programs can be tailored to patients’ specific disease stages and physical capabilities by employing different exercise modalities. This approach allows patients to transition smoothly and rapidly into the exercise phase, preventing exercise resistance in some patients who may be intimidated by the high intensity of DT. Furthermore, for patients lacking lower-limb mobility, modified forms of exercise—such as wheelchair dancing—can be adopted to better serve the needs of PD patients.
5 Strengths and limitations
This article incorporated a substantial number of RCTs, resulting in a large sample size that provides more objective, reliable, and credible data compared to case studies. It is the first to include traditional Chinese exercises for comparison with other exercise types, offering a more objective and comprehensive overview of exercise modalities. Furthermore, this review employed meta-analysis, which not only enhanced the statistical power of the findings but also yielded more precise data. It also helped clarify the reasons for contradictory results across different studies, providing novel and comprehensive evidence-based data.
However, this study has certain limitations. The intervention duration and frequency differed across the various exercise protocols included, with a wide range across studies. Moreover, the dosage of exercise was not analyzed, which may introduce limitations. Some of the included literature did not report the disease stage/severity of PD patients, making it impossible to determine whether the exercise interventions were effective only for those with mild symptoms. The evidence for certain exercise interventions is limited, resulting in a lack of direct comparative data for some outcomes. Therefore, the interpretation of these results should be approached with caution, and future research should aim to increase the number of studies in these areas.
6 Conclusion
Current evidence suggests that exercise involving walking and dance, which stimulates neural regulation, can help improve motor function and cognitive abilities in people with Parkinson’s disease. Patients can choose exercises based on their current fitness level.
Statements
Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.
Author contributions
JT: Data curation, Formal analysis, Writing – original draft. XL: Data curation, Formal analysis, Investigation, Visualization, Writing – review & editing. JG: Data curation, Formal analysis, Software, Writing – original draft. SL: Methodology, Formal analysis, Writing – original draft. ZX: Supervision, Writing – review & editing.
Funding
The author(s) declared that financial support was not received for this work and/or its publication.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fphys.2026.1809614/full#supplementary-material
References
1
AcarerA.KarapolatH.CelebisoyN.OzgenG.ColakogluZ. (2015). Is customized vestibular rehabilitation effective in patients with Parkinson’s? NeuroRehabilitation37, 255–262. doi: 10.3233/NRE-151258
2
AltmannL. J. P.StegemöllerE.HazamyA. A.WilsonJ. P.BowersD.OkunM. S.et al (2016). Aerobic Exercise Improves Mood, Cognition, and Language Function in Parkinson’s Disease: Results of a Controlled Study. J. Int. Neuropsychol. Soc.22, 878–889. doi: 10.1017/S135561771600076X
3
AmanoS.NoceraJ. R.VallabhajosulaS.JuncosJ. L.GregorR. J.WaddellD. E.et al. (2013). The effect of Tai Chi exercise on gait initiation and gait performance in persons with Parkinson’s disease. Parkinsonism Relat. Disord.19, 955–960. doi: 10.1016/j.parkreldis.2013.06.007
4
ArcolinI.FP.CD.MG.MS.AM.et al. (2016). Intensive cycle ergometer training improves gait speed and endurance in patients with Parkinson’s disease: A comparison with treadmill training. Restor. Neurol. Neurosci.34. doi: 10.3233/RNN-150506
5
BangD.-H.ShinW.-S. (2017). Effects of an intensive Nordic walking intervention on the balance function and walking ability of individuals with Parkinson’s disease: a randomized controlled pilot trial. Aging Clin. Exp. Res.29, 993–999. doi: 10.1007/s40520-016-0648-9
6
BelloO.SanchezJ. A.Lopez-AlonsoV.MárquezG.MorenillaL.CastroX.et al. (2013). The effects of treadmill or overground walking training program on gait in Parkinson’s disease. Gait Posture38, 590–595. doi: 10.1016/j.gaitpost.2013.02.005
7
CamargoC. H. F.TolentinoE. S.BronziniA.LadeiraM. A.LimaR.Schultz-PereiraG. L.et al. (2016). Comparison of the use of screening tools for evaluating cognitive impairment in patients with Parkinson’s disease. Dement Neuropsychol.10, 344–350. doi: 10.1590/s1980-5764-2016dn1004015
8
CanningC. G.AllenN. E.DeanC. M.GohL.FungV. S. C. (2012). Home-based treadmill training for individuals with Parkinson’s disease: a randomized controlled pilot trial. Clin. Rehabil.26, 817–826. doi: 10.1177/0269215511432652
9
CarrollL.DV.MeM.JS.AmC. (2017). Aquatic Exercise Therapy for People With Parkinson Disease: A Randomized Controlled Trial. Arch. Phys. Med. Rehabil.98. doi: 10.1016/j.apmr.2016.12.006
10
CarvalhoA.BarbiratoD.AraujoN.MartinsJ. V.CavalcantiJ. L. S.SantosT. M.et al. (2015). Comparison of strength training, aerobic training, and additional physical therapy as supplementary treatments for Parkinson’s disease: pilot study. Clin. Interv. Aging10, 183–191. doi: 10.2147/CIA.S68779
11
CheungC.RB.JfW.JK.LZ.UM.et al. (2018). Effects of yoga on oxidative stress, motor function, and non-motor symptoms in Parkinson’s disease: a pilot randomized controlled trial. Pilot Feasibility Stud.4. doi: 10.1186/s40814-018-0355-8
12
ChoiH.-J. (2016). Effects of therapeutic Tai chi on functional fitness and activities of daily living in patients with Parkinson disease. J. Exerc. Rehabil.12, 499–503. doi: 10.12965/jer.1632654.327
13
CholewaJ.Boczarska-JedynakM.OpalaG. (2013). Influence of physiotherapy on severity of motor symptoms and quality of life in patients with Parkinson disease. Neurol. Neurochir. Pol.47, 256–262. doi: 10.5114/ninp.2013.35774
14
CombsS. A.DiehlM. D.StaplesW. H.ConnL.DavisK.LewisN.et al. (2011). Boxing training for patients with Parkinson disease: a case series. Phys. Ther.91, 132–142. doi: 10.2522/ptj.20100142
15
CombsS. A.DiehlM. D.ChrzastowskiC.DidrickN.McCoinB.MoxN.et al. (2013). Community-based group exercise for persons with Parkinson disease: a randomized controlled trial. NeuroRehabilitation32, 117–124. doi: 10.3233/NRE-130828
16
CorcosD. M.ComellaC. L.GoetzC. G. (2012). Tai chi for patients with Parkinson’s disease. N. Engl. J. Med.366, 1737–1738. doi: 10.1056/NEJMc1202921
17
CorcosD. M.RobichaudJ. A.DavidF. J.LeurgansS. E.VaillancourtD. E.PoonC.et al. (2013). A two-year randomized controlled trial of progressive resistance exercise for Parkinson’s disease. Mov. Disord.28, 1230–1240. doi: 10.1002/mds.25380
18
CugusiL.SollaP.SerpeR.CarzeddaT.PirasL.OggianuM.et al. (2015). Effects of a Nordic Walking program on motor and non-motor symptoms, functional performance and body composition in patients with Parkinson’s disease. NeuroRehabilitation37, 245–254. doi: 10.3233/NRE-151257
19
DemonceauM.MaquetD.JidovtseffB.DonneauA. F.BuryT.CroisierJ. L.et al (2017). Effects of twelve weeks of aerobic or strength training in addition to standard care in Parkinson’s disease: a controlled study. Eur. J. Phys. Rehabil. Med.53, 184–200. doi: 10.23736/S1973-9087.16.04272-6
20
DibbleL. E.ForemanK. B.AddisonO.MarcusR. L.LaStayoP. C. (2015). Exercise and medication effects on persons with Parkinson disease across the domains of disability: a randomized clinical trial. J. Neurol. Phys. Ther.39, 85–92. doi: 10.1097/NPT.0000000000000086
21
DibbleL. E.HaleT. F.MarcusR. L.GerberJ. P.LaStayoP. C. (2009). High intensity eccentric resistance training decreases bradykinesia and improves Quality Of Life in persons with Parkinson’s disease: a preliminary study. Parkinsonism Relat. Disord.15, 752–757. doi: 10.1016/j.parkreldis.2009.04.009
22
DuncanR. P.EarhartG. M. (2012). Randomized controlled trial of community-based dancing to modify disease progression in Parkinson disease. Neurorehabil. Neural Repair26, 132–143. doi: 10.1177/1545968311421614
23
DunckerD. J.BacheR. J. (2008). Regulation of coronary blood flow during exercise. Physiol. Rev.88, 1009–1086. doi: 10.1152/physrev.00045.2006
24
ErnstM.FolkertsA.-K.GollanR.LiekerE.Caro-ValenzuelaJ.AdamsA.et al. (2023). Physical exercise for people with Parkinson’s disease: a systematic review and network meta‐analysis. Cochrane Database Syst. Rev.2023, CD013856. doi: 10.1002/14651858.CD013856.pub2
25
FerrazD. D.TrippoK. V.DuarteG. P.NetoM. G.Bernardes SantosK. O.FilhoJ. O. (2018). The Effects of Functional Training, Bicycle Exercise, and Exergaming on Walking Capacity of Elderly Patients With Parkinson Disease: A Pilot Randomized Controlled Single-blinded Trial. Arch. Phys. Med. Rehabil.99, 826–833. doi: 10.1016/j.apmr.2017.12.014
26
FerreiraR. M.AlvesW. M. G. da C.de LimaT. A.AlvesT. G. G.Alves FilhoP. A. M.PimentelC. P.et al. (2018). The effect of resistance training on the anxiety symptoms and quality of life in elderly people with Parkinson’s disease: a randomized controlled trial. Arq. Neuropsiquiatr.76, 499–506. doi: 10.1590/0004-282X20180071
27
GladfelterB. (2011). “ The effect of Tai Chi exercise on balance and falls in persons with Parkinson’s,” in Evidence-based practice project reports. 2.
28
GoldmanJ. G.PostumaR. (2014). Premotor and nonmotor features of Parkinson’s disease. Curr. Opin. Neurol.27, 434. doi: 10.1097/WCO.0000000000000112
29
GoodwinV. A.RichardsS. H.HenleyW.EwingsP.TaylorA. H.CampbellJ. L. (2011). An exercise intervention to prevent falls in people with Parkinson’s disease: a pragmatic randomised controlled trial. J. Neurol. Neurosurg. Psychiatry82, 1232–1238. doi: 10.1136/jnnp-2011-300919
30
GranzieraS.AlessandriA.LazzaroA.ZaraD.ScarpaA. (2020). Nordic Walking and Walking in Parkinson’s disease: a randomized single-blind controlled trial. Aging Clin. Exp. Res.33, 965–971. doi: 10.1007/s40520-020-01617-w
31
HackneyM. E.EarhartG. M. (2009). Effects of dance on movement control in Parkinson’s disease: a comparison of Argentine tango and American ballroom. J. Rehabil. Med.41, 475–481. doi: 10.2340/16501977-0362
32
HarperS. A.DowdellB. T.KimJ. H.PollockB. S.RidgelA. L. (2019). Non-Motor Symptoms after One Week of High Cadence Cycling in Parkinson’s Disease. Int. J. Environ. Res. Public Health16. doi: 10.3390/ijerph16122104
33
HashimotoH.TakabatakeS.MiyaguchiH.NakanishiH.NaitouY. (2015). Effects of dance on motor functions, cognitive functions, and mental symptoms of Parkinson’s disease: a quasi-randomized pilot trial. Complement. Ther. Med.23, 210–219. doi: 10.1016/j.ctim.2015.01.010
34
HigginsJ. P. T.AltmanD. G.GøtzscheP. C.JüniP.MoherD.OxmanA. D.et al. (2011). The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ343, d5928. doi: 10.1136/bmj.d5928
35
HubbleR. P.NaughtonG.SilburnP. A.ColeM. H. (2018). Trunk Exercises Improve Gait Symmetry in Parkinson Disease: A Blind Phase II Randomized Controlled Trial. Am. J. Phys. Med. Rehabil.97, 151–159. doi: 10.1097/PHM.0000000000000858
36
JohanssonM. E.CameronI. G. M.Van der KolkN. M.de VriesN. M.KlimarsE.ToniI.et al. (2022). Aerobic exercise alters brain function and structure in Parkinson’s disease: a randomized controlled trial. Ann. Neurol.91, 203–216. doi: 10.1002/ana.26291
37
JonesJ. D.HassC.MangalP.LafoJ.OkunM. S.BowersD. (2014). The cognition and emotional well-being indices of the Parkinson’s disease questionnaire-39: what do they really measure? Parkinsonism Relat. Disord.20, 1236–1241. doi: 10.1016/j.parkreldis.2014.09.014
38
KhuzemaA.BrammathaA.Arul SelvanV. (2020). Effect of home-based Tai Chi, Yoga or conventional balance exercise on functional balance and mobility among persons with idiopathic Parkinson’s disease: An experimental study. Hong Kong Physiother. J.40, 39–49. doi: 10.1142/S1013702520500055
39
KurtE. E.BüyükturanB.BüyükturanÖ.ErdemH. R.TuncayF. (2018). Effects of Ai Chi on balance, quality of life, functional mobility, and motor impairment in patients with Parkinson’s disease. Disabil. Rehabil.40, 791–797. doi: 10.1080/09638288.2016.1276972
40
KwokJ. Y. Y.KwanJ. C. Y.AuyeungM.MokV. C. T.LauC. K. Y.ChoiK. C.et al. (2019). Effects of Mindfulness Yoga vs Stretching and Resistance Training Exercises on Anxiety and Depression for People With Parkinson Disease: A Randomized Clinical Trial. JAMA Neurol.76, 755–763. doi: 10.1001/jamaneurol.2019.0534
41
LacyB.PiotrowskiH. J.DeweyR. B.HusainM. M. (2023). Severity of depressive and motor symptoms impacts quality of life in Parkinson’s disease patients at an academic movement clinic: a cross-sectional study. Clin. Park Relat. Disord.8, 100180. doi: 10.1016/j.prdoa.2022.100180
42
LeeN.DkL.HsS. (2015). Effect of virtual reality dance exercise on the balance, activities of daily living, and depressive disorder status of Parkinson’s disease patients. J. Phys. Ther. Sci.27. doi: 10.1589/jpts.27.145
43
LeiS.ChenX.WangZ. (2025). Exercise prevention and treatment of Parkinson’s disease mediated exercise prevention and treatment of Parkinson’s disease mediated. Chin. J. Tissue Eng. Res.29, 5454–5468. doi: 10.12307/2025.503
44
LiF.HarmerP.FitzgeraldK.EckstromE.StockR.GalverJ.et al. (2012). Tai chi and postural stability in patients with Parkinson’s disease. N. Engl. J. Med.366, 511–519. doi: 10.1056/NEJMoa1107911
45
MalkkiH. (2013). Parkinson disease: nonmotor symptoms predict quality of life in patients with early Parkinson disease. Nat. Rev. Neurol.9, 544. doi: 10.1038/nrneurol.2013.176
46
ManciniM.HorakF. B. (2016). Potential of APDM Mobility Lab for the monitoring of the progression of Parkinson’s disease. Expert Rev. Med. Devices13, 455. doi: 10.1586/17434440.2016.1153421
47
Martinez-MartinP.KurtisM. M. (2012). Health-related quality of life as an outcome variable in Parkinson’s disease. Ther. Adv. Neurol. Disord.5, 105–117. doi: 10.1177/1756285611431974
48
MichelsK.DubazO.HornthalE.BegaD. (2018). “Dance Therapy” as a psychotherapeutic movement intervention in Parkinson’s disease. Complement. Ther. Med.40, 248–252. doi: 10.1016/j.ctim.2018.07.005
49
MishraB.SudheerP.RajanR.AgarwalA.SrivastavaM.NilimaN.et al. (2024). Bridging the gap between statistical significance and clinical relevance: A systematic review of minimum clinically important difference (MCID) thresholds of scales reported in movement disorders research. Heliyon10 (5), e26479. doi: 10.1016/j.heliyon.2024.e26479
50
MitarnunW.MitranunW.MitarnunW.PangwongW. (2022). Home-Based Walking Meditation Decreases Disease Severity in Parkinson’s Disease: A Randomized Controlled Trial. J. Integr. Complement. Med.28, 227–233. doi: 10.1089/jicm.2021.0292
51
MiyataK.AsakuraT.ShinoharaT.UsudaS. (2021). Minimal clinically important difference for mini-balance evaluation systems test and Berg balance scale: a systematic review. Japanese J. Rehabil. Med. doi: 10.2490/jjrmc.20032
52
MoratelliJ. A.CorrêaC. L.AndradeA.LyraV. B.GuimarãesA. C. d. A. (2025). Functional training and Mat Pilates have a positive effect on non-motor symptoms improving cognition, depressive symptoms, anxiety, and happiness in people with Parkinson’s disease: a randomized controlled clinical trial with follow-up. Aging Ment. Health. 1–10. doi: 10.1080/13607863.2025.2496728
53
NadeauA.PourcherE.CorbeilP. (2014). Effects of 24 wk of treadmill training on gait performance in Parkinson’s disease. Med. Sci. Sports Exerc.46, 645–655. doi: 10.1249/MSS.0000000000000144
54
NiM.MooneyK.SignorileJ. F. (2016a). Controlled pilot study of the effects of power yoga in Parkinson’s disease. Complement. Ther. Med.25, 126–131. doi: 10.1016/j.ctim.2016.01.007
55
NiM.SignorileJ. F.BalachandranA.PotiaumpaiM. (2016b). Power training induced change in bradykinesia and muscle power in Parkinson’s disease. Parkinsonism Relat. Disord.23, 37–44. doi: 10.1016/j.parkreldis.2015.11.028
56
NoceraJ. R.AmanoS.VallabhajosulaS.HassC. J. (2013). Tai Chi Exercise to Improve Non-Motor Symptoms of Parkinson’s Disease. J. Yoga Phys. Ther.3. doi: 10.4172/2157-7595.1000137
57
PageM. J.McKenzieJ. E.BossuytP. M.BoutronI.HoffmannT. C.MulrowC. D.et al. (2021). The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ372, n71. doi: 10.1136/bmj.n71
58
PangW.ZhaiL.LiuY.ChenX.ZhuB.LiJ.et al. (2022). Analysis of global guidelines for comprehensive treatment of Parkinson’s disease. Chin. J. Clin. Pharmacol.38. doi: 10.13699/j.cnki.1001-6821.2022.21.026
59
PaulS. S.CanningC. G.SongJ.FungV. S. C.SherringtonC. (2014). Leg muscle power is enhanced by training in people with Parkinson’s disease: a randomized controlled trial. Clin. Rehabil.28, 275–288. doi: 10.1177/0269215513507462
60
PazzagliaC.II.ET.CM.DR.RL. M.et al. (2020). Comparison of virtual reality rehabilitation and conventional rehabilitation in Parkinson’s disease: a randomised controlled trial. Physiotherapy106. doi: 10.1016/j.physio.2019.12.007
61
PetzingerG. M.FisherB. E.McEwenS.BeelerJ. A.WalshJ. P.JakowecM. W. (2013). Exercise-enhanced neuroplasticity targeting motor and cognitive circuitry in Parkinson’s disease. Lancet Neurol.12, 716–726. doi: 10.1016/S1474-4422(13)70123-6
62
PicelliA.VaraltaV.MelottiC.ZatezaloV.FonteC.AmatoS.et al. (2016). Effects of treadmill training on cognitive and motor features of patients with mild to moderate Parkinson’s disease: a pilot, single-blind, randomized controlled trial. Funct. Neurol.31, 25–31. doi: 10.11138/fneur/2016.31.1.025
63
PostumaR. B.BergD.SternM.PoeweW.OlanowC. W.OertelW.et al. (2015). MDS clinical diagnostic criteria for Parkinson’s disease. Movement Disord.30 (12), 1591–1601. doi: 10.1002/mds.26424
64
QutubuddinA.ReisT.AlramadhaniR.CifuD. X.TowneA.CarneW. (2013). Parkinson’s disease and forced exercise: a preliminary study. Rehabil. Res. Pract.2013, 375267. doi: 10.1155/2013/375267
65
Rios RomenetsS.AnangJ.FereshtehnejadS.-M.PelletierA.PostumaR. (2015). Tango for treatment of motor and non-motor manifestations in Parkinson’s disease: a randomized control study. Complement. Ther. Med.23, 175–184. doi: 10.1016/j.ctim.2015.01.015
66
SacheliM. A.NevaJ. L.LakhaniB.MurrayD. K.VafaiN.ShahinfardE.et al. (2019). Exercise increases caudate dopamine release and ventral striatal activation in Parkinson’s disease. Mov. Disord.34, 1891–1900. doi: 10.1002/mds.27865
67
SageM.AlmeidaQ. (2009). Symptom and gait changes after sensory attention focused exercise vs aerobic training in Parkinson’s disease. Mov. Disord.24. doi: 10.1002/mds.22469
68
SalantiG.AdesA. E.IoannidisJ. P. A. (2011). Graphical methods and numerical summaries for presenting results from multiple-treatment meta-analysis: an overview and tutorial. J. Clin. Epidemiol.64, 163–171. doi: 10.1016/j.jclinepi.2010.03.016
69
SantosL.Fernandez-RioJ.WingeK.Barragán-PérezB.González-GómezL.Rodríguez-PérezV.et al. (2017). Effects of progressive resistance exercise in akinetic-rigid Parkinson’s disease patients: a randomized controlled trial. Eur. J. Phys. Rehabil. Med.53, 651–663. doi: 10.23736/S1973-9087.17.04572-5
70
SantosP.TM.LS.NR.AM. (2019). Efficacy of the Nintendo Wii combination with Conventional Exercises in the rehabilitation of individuals with Parkinson’s disease: A randomized clinical trial. NeuroRehabil.45. doi: 10.3233/NRE-192771
71
SchenkmanM.HallD. A.BarónA. E.SchwartzR. S.MettlerP.KohrtW. M. (2012). Exercise for people in early- or mid-stage Parkinson disease: a 16-month randomized controlled trial. Phys. Ther.92, 1395–1410. doi: 10.2522/ptj.20110472
72
SchlenstedtC.SP.AK.JR.BW.GD. (2015). Resistance versus Balance Training to Improve Postural Control in Parkinson’s Disease: A Randomized Rater Blinded Controlled Study. PLoS One10. doi: 10.1371/journal.pone.0140584
73
SchootemeijerS.van der KolkN.BloemB.de VriesN. (2020). Current perspectives on aerobic exercise in people with Parkinson’s disease. Neurotherapeutics17 (4), 1418–1433. doi: 10.1007/s13311-020-00904-8
74
ShamseerL.MoherD.ClarkeM.GhersiD.LiberatiA.PetticrewM.et al. (2015). Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ350, g7647. doi: 10.1136/bmj.g7647
75
ShinM.-S.JeongH.-Y.AnD.-I.LeeH.-Y.SungY.-H. (2016). Treadmill exercise facilitates synaptic plasticity on dopaminergic neurons and fibers in the mouse model with Parkinson’s disease. Neurosci. Lett.621, 28–33. doi: 10.1016/j.neulet.2016.04.015
76
SollaP.CugusiL.BertoliM.CereattiA.Della CroceU.PaniD.et al. (2019). Sardinian Folk Dance for Individuals with Parkinson’s Disease: A Randomized Controlled Pilot Trial. J. Altern. Complement. Med.25, 305–316. doi: 10.1089/acm.2018.0413
77
SonH. G.ChoiE. O.. (2018). The Effects of Mindfulness Meditation-based Complex Exercise Program on Motor and Non-Motor Symptoms, and Quality of Life in Patients with Parkinson’s Disease. Asian Nurs. Res. doi: 10.1016/j.anr.2018.06.001
78
SongJ.PaulS. S.CaetanoM. J. D.SmithS.DibbleL. E.LoveR.et al. (2018). Home-based step training using videogame technology in people with Parkinson’s disease: a single-blinded randomised controlled trial. Clin. Rehabil.32, 299–311. doi: 10.1177/0269215517721593
79
TanakaK.QuadrosA. C. de J.SantosR. F.StellaF.GobbiL. T. B.GobbiS. (2009). Benefits of physical exercise on executive functions in older people with Parkinson’s disease. Brain Cogn.69, 435–441. doi: 10.1016/j.bandc.2008.09.008
80
TajiriN.YasuharaT.ShingoT.KondoA.YuanW.KadotaT.et al. (2010). Exercise exerts neuroprotective effects on Parkinson’s disease model of rats. Brain Res.1310, 200–207. doi: 10.1016/j.brainres.2009.10.075
81
TangR.WangY.WangM.YinX.WangZ.GaoZ. (2023). A meta-analysis of the effect of aerobic exercise on cognitive function in patients with Parkinson’s disease. Neural Injury Funct. Reconstruction18, 396–403. doi: 10.16780/j.cnki.sjssgncj.20220791
82
TollárJ.NagyF.HortobágyiT. (2019). Vastly Different Exercise Programs Similarly Improve Parkinsonian Symptoms: A Randomized Clinical Trial. Gerontology65, 120–127. doi: 10.1159/000493127
83
TollárJ.NagyF.KovácsN.HortobágyiT. (2018). A High-Intensity Multicomponent Agility Intervention Improves Parkinson Patients’ Clinical and Motor Symptoms. Arch. Phys. Med. Rehabil.99, 2478–2484.e1. doi: 10.1016/j.apmr.2018.05.007
84
ToyW. A.PetzingerG. M.LeyshonB. J.AkopianG. K.WalshJ. P.HoffmanM. V.et al. (2014). Treadmill exercise reverses dendritic spine loss in direct and indirect striatal medium spiny neurons in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) mouse model of Parkinson’s disease. Neurobiol. Dis.63, 201–209. doi: 10.1016/j.nbd.2013.11.017
85
van der KolkN. M.de VriesN. M.PenkoA. L.van der VlugtM.MulderA. A.PostB.et al. (2018). A remotely supervised home-based aerobic exercise programme is feasible for patients with Parkinson’s disease.
86
van der KolkN. M.de VriesN. M.KesselsR. P. C.JoostenH.ZwindermanA. H.PostB.et al. (2019). Effectiveness of home-based and remotely supervised aerobic exercise in Parkinson’s disease: a double-blind, randomised controlled trial. Lancet Neurol.18, 998–1008. doi: 10.1016/S1474-4422(19)30285-6
87
VanLeeuwenJ.PetzingerG. M.WalshJ. P.AkopianG. K.VuckovicM.JakowecM. W. (2010). Altered AMPA receptor expression with treadmill exercise in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-lesioned mouse model of basal ganglia injury. J. Neurosci. Res.88, 650–668. doi: 10.1002/jnr.22216
88
Vergara-DiazG.OsypiukK.HausdorffJ. M.BonatoP.GowB. J.MirandaJ. G.et al. (2018). Tai Chi for Reducing Dual-task Gait Variability, a Potential Mediator of Fall Risk in Parkinson’s Disease: A Pilot Randomized Controlled Trial. Glob. Adv. Health Med.7, 2164956118775385. doi: 10.1177/2164956118775385
89
Vieira de Moraes FilhoA.ChavesS. N.MartinsW. R.TolentinoG. P.de Cassia Pereira Pinto HomemR.Landim de FariasG.et al. (2020). Progressive Resistance Training Improves Bradykinesia, Motor Symptoms and Functional Performance in Patients with Parkinson’s Disease. Clin. Interv. Aging15, 87–95. doi: 10.2147/CIA.S231359
90
VivasJ.AriasP.CudeiroJ. (2011). Aquatic therapy versus conventional land-based therapy for Parkinson’s disease: an open-label pilot study. Arch. Phys. Med. Rehabil.92. doi: 10.1016/j.apmr.2011.03.017
91
VolpeD.GiantinM. G.MaestriR.FrazzittaG. (2014). Comparing the effects of hydrotherapy and land-based therapy on balance in patients with Parkinson’s disease: a randomized controlled pilot study. Clin. Rehabil.28, 1210–1217. doi: 10.1177/0269215514536060
92
VolpeD.GiantinM. G.ManuelaP.FilippettoC.PelosinE.AbbruzzeseG.et al. (2017). Water-based vs. non-water-based physiotherapy for rehabilitation of postural deformities in Parkinson’s disease: a randomized controlled pilot study. Clin. Rehabil.31, 1107–1115. doi: 10.1177/0269215516664122
93
VolpeD.SignoriniM.MarchettoA.LynchT.MorrisM. E. (2013). A comparison of Irish set dancing and exercises for people with Parkinson’s disease: a phase II feasibility study. BMC Geriatr.13, 54. doi: 10.1186/1471-2318-13-54
94
WangX.DongQ.DongS.CaiZ. (2021). Effects of Baduanjin and Balancer Exercise on Motor and Non-motor Symptoms of Parkinson’s Disease. Chin. J. Rehabil. Theory Pract.27, 111–116.
95
XiaoC.ZhuangY.KangY. (2016). Effect of Health Qigong Baduanjin on Fall Prevention in Individuals with Parkinson’s Disease. J. Am. Geriatr. Soc.64, e227–e228. doi: 10.1111/jgs.14438
96
YanY.XuY.WangX.WangY.HuangC.LinR.et al. (2024). The effect of multi-component exercise intervention in older people with Parkinson’s disease and mild cognitive impairment: A randomized controlled study. Geriatr. Nurs.60, 137–145. doi: 10.1016/j.gerinurse.2024.08.028
97
ZhangJ.YuY.JiangR. (2018). Influence of rehabilitation exercise intervention on limb motor function and self-care ability of patients with Parkinson′s disease. Chin. Nurs. Res.32, 3410–3413. doi: 10.12102/j.issn.1009-6493.2018.21.026
98
ZhouX.ZhaoP.GuoX.WangJ.WangR. (2022). Effectiveness of aerobic and resistance training on the motor symptoms in Parkinson’s disease: systematic review and network meta-analysis. Front. Aging Neurosci.14. doi: 10.3389/fnagi.2022.935176
99
ZhuM.ZhangY.PanJ.FuC.WangY. (2020). Effect of simplified Tai Chi exercise on relieving symptoms of patients with mild to moderate Parkinson’s disease. J. Sports Med. Phys. Fitness60, 282–288. doi: 10.23736/S0022-4707.19.10104-1
Summary
Keywords
aerobic exercise, cognitive level, goal-directed training, motor skills, non-motor skills, Parkinson’s disease, traditional exercise
Citation
Tang J, Liu X, Guan J, Li S and Xing Z (2026) Effects of different exercises on motor and non-motor abilities in patients with Parkinson disease—a network meta-analysis of randomized controlled trials. Front. Physiol. 17:1809614. doi: 10.3389/fphys.2026.1809614
Received
12 February 2026
Revised
17 March 2026
Accepted
01 April 2026
Published
01 May 2026
Volume
17 - 2026
Edited by
Giuseppe D’Antona, University of Pavia, Italy
Reviewed by
Guotuan Wang, Henan University, China
John Luca Palmieri, Icahn School of Medicine at Mount Sinai, United States
Updates
Copyright
© 2026 Tang, Liu, Guan, Li and Xing.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Zenghui Xing, xingzh792@nenu.edu.cn
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.