843174

research-article2019
CRE0010.1177/0269215519843174Clinical RehabilitationWang et al.

CLINICAL
Original article REHABILITATION

Clinical Rehabilitation

Effect of virtual reality on balance 1­–9

© The Author(s) 2019
Article reuse guidelines:
and gait ability in patients with sagepub.com/journals-permissions
DOI: 10.1177/0269215519843174
https://doi.org/10.1177/0269215519843174

Parkinson’s disease: a systematic journals.sagepub.com/home/cre

review and meta-analysis

Bo Wang1, Min Shen2, Yan-xue Wang1, Zhi-wen He1,

Shui-qing Chi3 and Zhao-hui Yang1

Abstract
Objective: The aim of this study was to evaluate the effectiveness of virtual reality interventions for
improving balance and gait in people with Parkinson’s disease.
Design: This is a systematic review and meta-analysis of randomized controlled trials.
Methods: Databases of MEDLINE, Cochran Central Register of Controlled Trials, EMBASE, PEDro,
Web of Science and China Biology Medicine disc were searched from their inception up to 1 March 2019.
Two reviewers individually appraised literatures for inclusion, extracted data and evaluated trial quality.
Results: A total of 12 studies with a median PEDro score of 6.4 and involving 419 participants were
included. This review first demonstrated significant improvements in Berg Balance Scale (mean difference
= 2.69; 95% confidence interval = 1.37 to 4.02; p < 0.0001), Timed Up and Go Test (mean difference =
−2.86; 95% confidence interval = −5.60 to −0.12; p = 0.04) and stride length (mean difference = 9.65;
95% confidence interval = 4.31 to 14.98; p = 0.0004) in Parkinson patients who received virtual reality
compared with controls. However, there was no significant difference in gait velocity and walk distance.
Conclusion: This systematic review and meta-analysis supports the use of virtual reality to enhance the
balance of patients with Parkinson’s disease. However, the review does not find any definite effect upon
gait by the use of virtual reality.

Keywords
Virtual reality, balance, gait, Parkinson’s disease, systematic review

Received: 15 September 2018; accepted: 14 March 2019

1Department of Rehabilitation, Union Hospital, Tongji Medical Corresponding author:

College, Huazhong University of Science and Technology,
Wuhan, China
2Department of Hematology, Tongji Hospital, Tongji Medical
College, Huazhong University of Science and Technology,
Wuhan, China
3Department of Pediatric Surgery, Union Hospital, Tongji
Zhao-hui Yang, Department of Rehabilitation, Union Hospital,
Tongji Medical College, Huazhong University of Science and
Technology, Wuhan 430022, China.
Email: annyhao430@hust.edu.cn

Medical College, Huazhong University of Science and

Technology, Wuhan, China
2 Clinical Rehabilitation 00(0)
Introduction
Parkinson’s disease is a neurodegenerative disease
that is often associated with movement impairments
such as reduced balance and gait control.1,2 The loss
of gait and balance with progression of the disease
has an impact on their ability to perform daily life
activities.3,4 Virtual reality is increasingly being used
in neurological conditions including Parkinson’s
disease, but that its effects are not yet well proven.5,6
Virtual reality is a high-tech computer–human
interface system which can make subjects engaged
in environments that feel similar to real-world
objects and events.7 A recent meta-analysis with 21
randomized controlled trials (RCTs) demonstrated
that virtual reality was more effective in training gait
and balance than conventional methods in patients
with stroke.8 The potential reasons are as follows:
not only it can strengthen feedback by visual, ves-
tibular and somatosensory input but it also enhances
the personalized training of gait and balance control.
However, the effect of virtual reality on gait and bal-
ance of Parkinson’s disease remains unclear.
Recently, a growing number of studies demon-
strated that the ability of gait and balance of
patients with Parkinson’s disease can be signifi-
cantly improved via virtual reality when compared
with conventional therapy.9–12 But, another two
studies pointed out that no significant difference
between virtual reality intervention and conven-
tional therapy was observed in Berg Balance Scale,
Timed Up and Go Test and gait.13,14 Unfortunately,
although there are already two systematic reviews
discussing this topic, this controversy has not been
solved and there exist some problems in these
reviews. First, one review mainly focused on older
adults and only one study involving Parkinson’s
disease was included in this review, thus the evi-
dence was low. Second, another review lacked a
meta-analysis. Finally, in the past year, new RCTs
comparing the effect of virtual reality training with
conventional therapy have been published.
Therefore, we performed this systematic review by
a comprehensive meta-analysis via an updated lit-
erature search and evaluated RCTs in order to
explain the evidence-based effectiveness of virtual
reality for Parkinson’s disease.
Methods
Data sources and searches
All procedures mentioned below were performed in
accordance with PRISMA Checklist and Cochrane
Collaboration protocols.15,16 Databases of MEDLINE
(via PubMed), Web of Science, EMBASE, PEDro,
Cochran Central Register of Controlled Trials and
China Biology Medicine disc were thoroughly exam-
ined using the search term “virtual reality or exer-
gaming or game or gaming,” “Parkinson’s Disease or
Parkinson or Parkinsonism,” “balance or posture or
postural control or mobility” or “gait or ambulation
or walking” and their synonyms, from their inception
to 1 March 2019. The detailed example of the full
electronic search strategy for MEDLINE is provided
in Supplementary Material Appendix A. Additional
records were searched through other sources to sup-
plement the database results. Also, manual search of
reference lists of related reviews was adopted. Two
investigators carried out study selection and data
extraction independently, while any discrepancy was
resolved by mutual discussion. All results involving
title, abstract and reference were imported into
EndNote X8 for screening.
Inclusion and exclusion criteria
Studies that met the following criteria were eventu-
ally included: (1) participants were clinically diag-
nosed as Parkinson’s disease, which was defined by
the UK Parkinson’s Disease Society Brain Bank or
other diagnostic criteria; (2) there were no restrictions
about participants’ age, gender, disease severity and
disease duration; (3) only RCTs that evaluated effects
of virtual reality interventions on balance and gait in
patients with Parkinson’s disease were selected; (4)
trials that compared two or more interventions (at
least one of which was an ongoing virtual reality
training) were also included; (5) at least one quantita-
tive outcome of balance and gait was measured and
no restrictions in frequency and duration of these vir-
tual reality interventions; (6) the effectiveness of vir-
tual reality interventions versus an alternative
intervention such as a placebo control intervention, or
any other exercise intervention without virtual reality
Wang et al. 3
was assessed; and (7) non-English language trials
were also included. The eliminated criteria were as
follows: (1) overlapped or duplicated articles; (2)
reviews, case reports, letters, unpublished articles and
conference abstracts; and (3) inadequate original data.
Data extraction and quality assessment
Two investigators (M.S. and Y.W.) extracted rele-
vant data independently and reached agreement on
all items. The following data were extracted from
selected studies: general characteristics (title, first
author and year published), demographic charac-
teristics of the study sample sizes, age, disease
stage and duration, study characteristics (content of
intervention, training intensity and frequency,
comparators and outcome measures) and signifi-
cant main findings in measures of balance and gait
ability between groups. Two reviewers (Z.H. and
S.C.) independently evaluated the included articles
according to the PEDro scale. A PEDro scale is
composed of 11 items that can contribute one point
to 10 score, except for the first item, eligibility cri-
teria. A PEDro scale is an effective measure of
methodological quality of clinical trials.17,18
Articles with a score of 6 or higher are considered
as high quality, and those with scores of less than 6
are defined as lower quality.17 In case of disagree-
ment in the quality assessment of two reviewers,
consensus was conducted by a third reviewer.
Data synthesis and analysis
Our quantitative meta-analysis was performed using
both fixed- or random-effects model in Review
Manager 5.3 under Cochrane Collaboration proto-
cols. Forest plots were generated to present the
pooled effect, and the mean difference (MD) with
95% confidence interval (CI) was calculated for the
all outcomes because the pooled studies used the
same rating scale or test. Heterogeneity was assessed
by means of the value of the I2 statistic. We acknowl-
edged I2 as a heterogeneity indicator with its value
<25%, 25%–50% and >50% defined as low, mod-
erate and significant heterogeneity, respectively. A
moderate or significant heterogeneity was adjusted
by a random-effects model; otherwise, a fixed-effects
model was preferred. The statistical significance
within all comparisons was mathematically signified
as p < 0.05. The potential publication bias was
assessed using Egger’s test.
The overall quality evidence was appraised using
GRADE (Grading of Recommendations, Assessment,
Development and Evaluation) approach.19 Through
this measure, the evidence was assessed not by indi-
vidual trail but across studies for special outcomes.
Factors that may reduce the quality of the evidence
include the following: study design, risk of bias,
inconsistency of results, indirectness and imprecision.
We adjusted the quality of evidence by a level based
on five factors.20 Levels of evidence were defined as
follows: (1) high-quality evidence: further studies are
very unlikely to change the estimate or the confidence
in the result. There are at least 75% of RCTs with no
limitations of study design, low risk of bias, consist-
ent finding, direct, and precise data and no known or
suspected publication biases; (2) moderate-quality
evidence: further studies are likely to have an impor-
tant impact on our confidence in the estimate of effect
and may change the estimate. One of the factors is not
met; (3) low-quality evidence; (4) very low-quality
evidence; and (5) no evidence.21
Results
Identification of studies
A total of 586 related articles were identified
through database searching. In addition, three stud-
ies were found by manually searching reference
lists. When duplicates were removed and titles and
abstracts of all remaining unique articles were
screened, there were 36 articles needed to be fully
evaluated for the final inclusion. Eventually, only
12 of them were selected for this review.9–14,22–27
The selection flow chart is depicted in Figure 1.
Description of included studies
The participants and study characteristics are shown
in Table 1 and Supplementary Material Appendix B.
In the selected studies, 12 parallel groups rand-
omized controlled studies involving 419 participants
published between 2012 and 2018. Six trials were
4 Clinical Rehabilitation 00(0)

Figure 1.  The selection flow chart of our systematic review.

undertaken in China, three from Brazil, the rest of
three were conducted in the Netherlands, Korea and
Italy, individually. The mean age of the participants
varied between 50.9 and 80.9 years in virtual reality
group and 48.9 and 81.7 years in control group. The
mean Hoehn and Yahr stage of the subjects varied
between 0.9 and 3.6 in virtual reality group and 0.87
and 3.6 in control group. In addition, the mean
Parkinson’s disease duration of the patients varied
between 0.29 and 13 years in virtual reality group
and 0.64 and 12.4 years in control group. Moreover,
only four studies involved sample sizes of less than
30 participants. In virtual reality group, there were
10 trials10–13,22–27 using basic virtual reality where 2
trials9,14 using augmented virtual reality.
Content of the interventions of included
studies
The intervention approaches differed with each other.
Five studies used the Nintendo Wii Fit. Three trials
used Kinect X-BOX or Sensor. The majority of stud-
ies (n = 10) compared the virtual reality intervention
with a comparable alternative intervention. The alter-
native intervention in most of the studies was con-
ventional physiotherapy (n = 7). Two studies
Wang et al. 5

Table 1.  Characteristics of included studies.

References Experiment Control Dosage Outcomes Major findings Quality

Shen (2014)9 BAL Muscle strength 100 min/d, 5 SLS, stride Significant 8
training d/w, 12w length, gait improvement in SLS
velocity and stride length
Lee (2015)10 VR plus NDT NDT and FES 45 min/d, 5 BBS Significant 4
and FES d/w, 6w improvement in BBS
Liao (2015)11 Wii Fit VR Traditional 1 h/d, 2 d/w, LOS, TUG, Significant 7
exercise exercise 6w obstacle improvement in
crossing obstacle crossing
performance velocity, crossing
stride length, TUG
Shih (2016)12 VR balance Balance training 50 min/d, 2 LOS, OLS, Significant 6
training d/w, 8w BBS, TUG improvement in LOS
and OLS
Pompeu Global Global exercise 1 h/d, 2 d/w, BBS, UST No significant 5
(2012)13 exercise plus plus balance 7w differences were
Wii Fit VR exercise found in BBS, UST
Van (2014)14 VFT CBT 1 h/d, 2 d/w, FRT, BBS, SLS, No significant 8
5w 10MWT differences were
found in FRT, BBS,
SLS and 10MWT
Yang (2016)22 VR balance CBT 50 min/d, 2 BBS, TUG, No significant 7
training d/w, 6w Gait differences were
found in BBS, TUG,
gait
Lin (2016)23 VR balance CBT 30 min/d, 5 BBS, TUG Significant 7
training d/w, 4w improvement in BBS,
TUG
Ribas (2017)24 Wii Fit VR Conventional 30 min/d, 2 BBS, 6MWT Significant 7
exercise exercise d/w, 12w improvement in BBS
Gandolfi VR balance SIBT 50 min/d, 3 BBS, ABC, Significant 6
(2017)25 training d/w, 7w 10MWT improvement in BBS
and Dynamic Gait
Index
Chen (2017)26 VR balance CBT 50 min/d, 5 BBS, TUG Significant 6
training d/w, 6w improvement in BBS,
TUG
Melo (2018)27 VR gait Conventional 20 min/d, 3 6MWT, speed Significant 6
training gait training d/w, 4w improvement in
6MWT, gait speed

VR, virtual reality; BBS, Berg Balance Scale; UST, Unipedal Stance Test; VFT, virtual feedback training; CBT, conventional balance
training; FRT, functional reach test; SLS, single leg stance test; 10MWT, 10-meter walk test; BAL, technology-assisted balance and
gait training; NDT, neurodevelopment treatment; FES, functional electrical stimulation; LOS, limits of stability; TUG, Timed Up
and Go Test; OLS: one-leg stance; 6MWT, 6-minute walk test; ABC, Activities Balance Confidence scale.

combined physical therapy as an experimental inter- from 20 to 100 minutes. In addition, the duration of
vention. Moreover, 10 studies focused on balance interventions varied between 4 and 12 weeks and the
interventions and 2 studies focused on gait training. frequency of interventions ranged from 2 to 5 day/
The intensity of different trainings of all trials ranged week. Furthermore, in terms of outcome measures,
6 Clinical Rehabilitation 00(0)

Figure 2.  Meta-analysis results using the random-effects model for (a) BBS and (b) TUG. Study effect, weightage
and confidence interval are represented by box position, size and the horizontal line. The pooled effect of the
studies is shown by the diamond and its width represents the confidence interval.

there were all kinds of methods. Both Berg Balance
Scale and Timed Up and Go Test were used to assess
the function balance. And the gait ability was
appraised by 10-meter walk test or 6-minute walk
test. All studies measured outcomes promptly post-
intervention and post-intervention scores were used
to compare between groups in this review.
PEDro scores
The PEDro scores of the included studies varied
between 4 and 8, and the average score was 6.4
(Supplementary Material Appendix C). In total, 10
studies had a score of 6 or higher and were consid-
ered of high quality. All trials randomly allocated
the subjects. Moreover, the majority of trials
reported eligibility criteria, had similar groups at
baseline, assessor blinding and concealed alloca-
tion and used both point measures and measures of
variability. In summary, 13 of the 132 items from
the PEDro scale were initially scored different by
the two reviewers. After discussion, there was in
agreement for all items.
Outcome measures and main findings of
included studies
Balance.  Nine studies provided Berg Balance Scale
data including 299 participants in this meta-analysis
(Figure 2(a)). The comprehensive results demon-
strated significant effect on Berg Balance Scale
score after a virtual reality intervention compared
with a control intervention (MD = 2.69; 95% CI =
1.37 to 4.02; p < 0.0001). However, substantial het-
erogeneity was indicated with an I2 statistic of 69%.
When excluding the studies by Chen et al. and Shih
et al., low heterogeneity (I2 = 9%) was observed.
The MD was 2.71 and remained significant in favor
of the virtual reality group (95% CI = 1.80 to 3.63;
p < 0.0001) (Supplementary Material Appendix D).
There was no publication bias within the included
cohorts according to the Egger’s test (p = 0.49).
Wang et al. 7

Figure 3.  Meta-analysis results using the fixed-effects model for (a) gait velocity, (b) walk distance and (c) stride
length. Study effect, weightage and confidence interval are represented by box position, size and the horizontal line.
The pooled effect of the studies is shown by the diamond and its width represents the confidence interval.

Five studies involving 144 participants meas-
ured the effect of virtual reality versus control ther-
apy on the Timed Up and Go Test (Figure 2(b)).
The impact was statistical significant: MD =
−2.86; 95% CI = −5.60 to −0.12; p = 0.04.
Heterogeneity was high (I2 = 73%). However,
excluding the studies by Chen et al., no heteroge-
neity (I2 = 0%) was observed. The MD was −1.74
and remained significant in favor of the virtual
reality group (95% CI = −3.21 to −0.27; p = 0.02)
(Supplementary Material Appendix D). Also, there
was no publication bias within the included cohorts
according to the Egger’s test (p = 0.089).
Gait ability.  Gait velocity was recorded in five stud-
ies (n = 203), but there was no significant difference
between groups (MD = −0.00; 95% CI = −0.06 to
0.06; p = 0.98) (Figure 3(a)) and no heterogeneity
(I2 = 0%). Also, two of the six studies (n = 45)
showed no significantly greater increases in walk
distance in virtual reality group compared with the
control group using 6-minute walk test (MD = 8.91,
95% CI = −43.32 to 61.13; p = 0.74) (Figure 3(b)).
No heterogeneity (I2 = 0%) was observed. How-
ever, in terms of stride length, there was a significant
difference when virtual reality group was compared
with the control group (MD = 9.65; 95% CI = 4.31
to 14.98; p = 0.0004) (Figure 3(c)). And there is no
statistical heterogeneity (I2 = 0%). An Egger’s test
was not applied due to the very small number of
studies.
As for subgroup analysis, the different Hoehn
and Yahr stage of included studies were compared
in the Berg Balance Scale and Timed Up and Go
Test, and no significant differences were observed
between groups in Berg Balance Scale (p = 0.27)
and groups in Timed Up and Go Test (p = 0.30)
(Supplementary Material Appendix E).
8 Clinical Rehabilitation 00(0)
In summary, using GRADE criteria, this review
found moderate evidence that virtual reality inter-
ventions are more effective than control interven-
tions in Berg Balance Scale and low evidence that
virtual reality training is more effective than con-
trol training in Timed Up and Go Test in Parkinson’s
disease. Furthermore, there is very low-quality evi-
dence that virtual reality interventions are more
effective than control interventions in stride length
(Supplementary Material Appendix F).
Discussion
This meta-analysis demonstrated significant results
of virtual reality training on the improvement of
Berg Balance Scale based on moderate evidence
according to GRADE criteria and Timed Up and
Go Test with low evidence according to GRADE
criteria. In addition, with respect to gait ability, sig-
nificant improvements in stride length were
reported in two studies, although this finding was
supported only by very low evidence according to
GRADE criteria.
The positive findings of virtual reality training are
consistent with a previous review on the effect of vir-
tual reality on balance including Berg Balance Scale
and Timed Up and Go Test in patients with Parkinson’s
disease.28 However, the systematic review by Dockx
et al.29 did not support these positive results. The
potential conflicts may be due to the fact that the
review included only seven studies for meta-analysis.
And five additional studies were included in our
review, which were published after its publication.
However, there exist some limitations in our
study. First, the majority of 12 included studies
had small sample size and intervention approaches
varied considerably in this review. Not only the
devices of virtual reality come from different cor-
porations but also the contents of virtual reality
differ with each other. Moreover, there are some
variabilities in the number of intervention ses-
sions, frequency and duration in the included stud-
ies. In addition, the majority of post-intervention
follow-up ranges from 1 month to 3 months and
only one study reported the effects after 15 months.
Furthermore, virtual reality is not a very accurate
description, although it is widely used. Thus, we
may have missed some literatures through the
search terms we have used. Finally, although the
majority of the selected studies provided high
methodological quality, the most did not conduct
participant and clinician blinding, which could
lead to bias in the included trials.
Even though there exist some limitations in
this systematic review, there remain some impli-
cations for practice and future research. First, vir-
tual reality therapies are thought to improve
motivation and enjoyment, to increase the activ-
ity adherence in training and to decrease the per-
ception of exertion. However, none of included
studies in this review assessed the motivation.
Further study should consider motivation as an
outcome measure. Second, to assess the duration
of any improvement outcome to be sustained for
virtual reality intervention, future trials should
have longer follow-up duration. Furthermore,
additional large-scale RCTs should be conducted
to appraise the effect of virtual reality interven-
tion because of the small sample size of included
studies in this review. Finally, future researchers
should compare the effects between different vir-
tual reality interventions and investigate the most
effective virtual reality method and dosage.
Clinical Messages
•• This systematic review suggests that
“virtual reality” interventions are effec-
tive at improving balance in people with
Parkinson’s disease.
•• The review did not find any definite
effect upon gait and mobility, but the
total number of studies and the total
number of patients were both low leaving
some uncertainty on this matter.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest
with respect to the research, authorship and/or publica-
tion of this article.
Funding
The author(s) received no financial support for the
research, authorship and/or publication of this article.
Wang et al. 9
Supplemental material
Supplemental material for this article is available online.
References
1. Wang XQ, Pi YL, Chen BL, et al. Cognitive motor inter-
vention for gait and balance in Parkinson’s disease: sys-
tematic review and meta-analysis. Clin Rehabil 2016;
30(2): 134.
2. Canning CG, Paul SS and Nieuwboer A. Prevention of
falls in Parkinson’s disease: a review of fall risk factors
and the role of physical interventions. Neurodegener Dis
Manag 2014; 4(3): 203–221.
3. Pedreira G, Prazeres A, Cruz D, et al. Virtual games and
quality of life in Parkinson’s disease: a randomised con-
trolled trial. Adv Parkinsons Dis 2013; 2: 97–101.
4. Rizos A, Martinez-Martin P, Odin P, et al. Characterizing
motor and non-motor aspects of early-morning off periods
in Parkinson’s disease: an international multicenter study.
Parkinsonism Relat Disord 2014; 20(11): 1231–1235.
5. Kim JH, Jang SH, Kim CS, et al. Use of virtual reality
to enhance balance and ambulation in chronic stroke: a
double-blind, randomized controlled study. Am J Phys
Med Rehabil 2009; 88(9): 693–701.
6. Gatica-Rojas V and Mendez-Rebolledo G. Virtual reality
interface devices in the reorganization of neural networks
in the brain of patients with neurological diseases. Neural
Regen Res 2014; 9(8): 888–896.
7. Imam B and Jarus T. Virtual reality rehabilitation from
social cognitive and motor learning theoretical perspec-
tives in stroke population. Rehabil Res Pract 2014; 2014:
594540.
8. de Rooij IJ, Ig VDP and Meijer JG. The effect of virtual
reality training on balance and gait ability in patients with
stroke: a systematic review and meta-analysis. Phys Ther
2016; 96: 1905–1918.
9. Shen X and Mak MK. Technology-assisted balance and
gait training reduces falls in patients with Parkinson’s dis-
ease: a randomized controlled trial with 12-month follow-
up. Neurorehab Neural Re 2014; 29: 103.
10. Lee NY, Lee DK and Song HS. 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 2015; 27(1): 145.
11. Liao YY, Yang YR, Cheng SJ, et al. Virtual reality-based
training to improve obstacle-crossing performance and
dynamic balance in patients with Parkinson’s disease.
Neurorehabil Neural Repair 2015; 29(7): 658–667.
12. Shih MC, Wang RY, Cheng SJ, et al. Effects of a balance-
based exergaming intervention using the Kinect sensor on
posture stability in individuals with Parkinson’s disease:
a single-blinded randomized controlled trial. J Neuroeng
Rehabil 2016; 13(1): 78.
13. Pompeu JE, Mendes FA, Silva KG, et al. Effect of Nintendo
Wii™-based motor and cognitive training on activities of
daily living in patients with Parkinson’s disease: a ran-
domised clinical trial. Physiotherapy 2012; 98(3): 196.
14. van den Heuvel MR, Kwakkel G, Beek PJ, et al. Effects
of augmented visual feedback during balance training
in Parkinson’s disease: a pilot randomized clinical trial.
Parkinsonism Relat Disord 2014; 20(12): 1352–1358.
15. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA state-
ment for reporting systematic reviews and meta-analyses of
studies that evaluate healthcare interventions: explanation
and elaboration. J Clin Epidemiol 2009; 62: e1–e34.
16. Higgins JP and Green S. Cochrane handbook for sys-
tematic reviews of interventions Version 5.0.0. Naunyn-
Schmiedeberg’s Archiv Für Exp Pathol Pharm 2008; 5: S38.
17. Maher CG, Sherrington C, Herbert RD, et al. Reliability
of the PEDro scale for rating quality of randomized con-
trolled trials. Phys Ther 2003; 83(8): 713–721.
18. de Morton NA. The PEDro scale is a valid measure of the
methodological quality of clinical trials: a demographic
study. Aust J Physiother 2009; 55(2): 129–133.
19. Furlan AD, Malmivaara A, Chou R, et al. 2015 updated
method guideline for systematic reviews in the Cochrane
back and neck group. Spine 2015; 40(21): 1660–1673.
20. Atkins D, Best D, Briss PA, et al. Grading quality of
evidence and strength of recommendations. BMJ 2004;
328: 1490.
21. Romeo A, Vanti C, Boldrini V, et al. Cervical radiculopa-
thy: effectiveness of adding traction to physical therapy
—a systematic review and meta-analysis of randomized
controlled trials. Phys Ther 2018; 98: 231–242.
22. Yang WC, Wang HK, Wu RM, et al. Home-based virtual
reality balance training and conventional balance training
in Parkinson’s disease: a randomized controlled trial. J
Formos Med Assoc 2016; 115(9): 734–743.
23. Lin Z, Chen A and Yijing J. Effects of virtual reality bal-
ance game on balance function for Parkinson’s disease.
Chinese J Rehabil Theor Pract 2016; 9: 1059–1063.
24. Ribas CG, Alves da Silva L, Correa MR, et al.
Effectiveness of exergaming in improving functional bal-
ance, fatigue and quality of life in Parkinson’s disease:
a pilot randomized controlled trial. Parkinsonism Relat
Disord 2017; 38: 13–18.
25. Gandolfi M, Geroin C, Dimitrova E, et al. Virtual real-
ity telerehabilitation for postural instability in Parkinson’s
disease: a multicenter, single-blind, randomized, con-
trolled trial. Biomed Res Int 2017; 2017: 7962826.
26. Chen S, Liu J and Shun L. Effects of virtual reality reha-
bilitation on balance for patients with Parkinson’s disease.
Chinese J Rehabil Theor Pract 2017; 9: 1091–1095.
27. de Melo GEL, Kleiner AFR, Lopes JBP, et al. Effect of vir-
tual reality training on walking distance and physical fitness
in individuals with Parkinson’s disease. Neurorehabilitation
2018; 42(4): 473–480.
28. Harris DM, Rantalainen T, Muthalib M, et al. Exergaming
as a viable therapeutic tool to improve static and dynamic
balance among older adults and people with idiopathic
Parkinson’s disease: a systematic review and meta-anal-
ysis. Front Aging Neurosci 2015; 7: 167.
29. Dockx K, Bekkers EM, Van den Bergh V, et al. Virtual
reality for rehabilitation in Parkinson’s disease. Cochrane
Database Syst Rev 2016; 12: CD010760.