Can Wii improve balance?
Abstract: The interest in gaming systems as rehabilitation interventions is increasing rapidly and anecdotal evidence suggests that use of gaming systems is relatively common in clinical practice. This paper considers the evidence for the use of the Nintendo Wii balance board with Wii Fit software as a possible intervention for improving balance in adults and older adults. There are currently no high quality randomised controlled trials to support or refute the use of the Wii systems. A few pre-post trials present equivocal results in older adult samples but little is known in relation to people with neurological impairments. At this point in time it would be judicious to be cautious in the use of these gaming systems to treat adults with balance problems.

Taylor D (2011): Can Wii improve balance? New Zealand Journal of Physiotherapy 39(3) 131-133.
Subject: Aged (Research)
Physical therapy for the aged (Research)
Physical therapy (Research)
Therapeutics, Physiological (Research)
Computer entertainment systems (Research)
Software (Research)
Adults (Research)
Video game industry (Research)
Author: Taylor, Denise
Pub Date: 11/01/2011
Publication: Name: New Zealand Journal of Physiotherapy Publisher: New Zealand Society of Physiotherapists Audience: Academic Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2011 New Zealand Society of Physiotherapists ISSN: 0303-7193
Issue: Date: Nov, 2011 Source Volume: 39 Source Issue: 3
Topic: Event Code: 310 Science & research Computer Subject: Computer-based entertainment system; Software quality
Product: Product Code: 7372000 Computer Software; E121940 Adults; 3651920 Electronic Games NAICS Code: 51121 Software Publishers; 339932 Game, Toy, and Children's Vehicle Manufacturing SIC Code: 3944 Games, toys, and children's vehicles; 7372 Prepackaged software
Accession Number: 288538004
Full Text: There is a great deal of enthusiasm afoot for the use of gaming systems in rehabilitation, as demonstrated by the thousands of hits on search engines for information about gaming and rehabilitation. Whilst there are other gaming systems this paper will focus on the Nintendo Wii[R] (Nintendo, Redmond, Washington) as it is the only system that includes a balance platform, the Wii balance board. Although the games developed by Nintendo were meant for the entertainment market it is worth considering if they can be used in the rehabilitation field with any efficacy and effectiveness. Anecdotal evidence suggests that many physiotherapists in New Zealand are using the systems in their practice. For this reason it seems valuable to consider the currently available evidence supporting the use of the Wii balance board (Wii BB) along with the associated software, the Wii Fit software in balance rehabilitation for adults and older adults. This seems to be a viable and anecdotally common use of the Nintendo Wii system.

The Wii BB collects information about postural sway and uses this to drive on-screen images. A quick look at the website, http://www. wiifit.com, will allow you to see the types of activities and games available that utilise the Wii BB. In the Wii Fit package there are a number of activities and games available such as Yoga, balance training, strength training and aerobics. There is also a measurement section, which produces assessment of centre of gravity movement and weight bearing symmetry. A newly added section called 'training plus' includes a variety of activities and games; one of which, Perfect 10, could act as a balance task with a secondary cognitive component. This may be useful in patients for whom you want to challenge the primary task of balance by giving a secondary cognitive task. Of course the risk of falling or losing balance may well be increased during this activity, so caution should be taken.

At first glance it looks as if the Wii BB with Wii Fit software is an interesting and potentially useful balance rehabilitation tool. The next step is to consider the research evidence for the use of the Wii BB in balance rehabilitation. The strongest form of evidence we have is, of course, the randomised controlled trial. I conducted a search to find trials in which the Wii BB had been used to improve balance in adults and older adults.

A systematic search of Scopus, Science Direct, Cinahl, and Medline was conducted. The search terms were broad to avoid missing relevant trials, the search period included 2007-present (as the Nintendo Wii BB was not introduced until 2007). The terms 'Nintendo Wii' and 'balance' were used in all search engines to identify publications. Titles and abstracts were read and publications were included that fulfilled the following criteria a) used the Wii balance board, b) measured balance, c) included a population of adults or older adults and d) used a randomised controlled trial design.

Of the 127 articles identified in the database search and in the search of the reference lists of key references there was only one pilot randomised controlled trial that investigated the effect of Wii BB as a rehabilitation tool to improve balance in adults with acquired brain injury, including stroke, traumatic brain injury and benign cerebral neoplasm (Gil-Gomez et al 2011)(see Table 1). In this trial the Wii BB was coupled with custom made software that had been designed with input from experts in balance rehabilitation; the Wii Fit activities and games were not used. This has the effect of reducing the applicability of this particular trial. However, there are some interesting considerations arising out of this trial. The participants in the trial were more than six months since injury, were able to walk indoors, with or without an aid for 10m, and were cognitively intact without neglect or cerebellar symptoms. From a potential pool of 79 patients, 20 fitted the criteria and were entered into the trial, three dropped out and their results were not included in the analysis. Participants were randomly allocated to either a traditional physiotherapy group or a virtual rehabilitation group. This was a small study with only eight people in the control group and nine in the experimental group. Participants underwent 20 hours of rehabilitation. The content of the rehabilitation for the virtual rehabilitation group is well explained but there is no information about the content of the traditional rehabilitation group. A total of 11 different measures of balance were taken. Of these only two showed a between group difference at the end of the trial, the Berg Balance Score (BBS) and the forward Functional Reach. The BBS improved from 41.2 (SD 10.6) to 45.4 (SD 8.6) in the virtual rehabilitation group and from 45.4 (SD 7.4) to 46.9 (SD 6.2) in the traditional therapy group. It is not surprising that the forward reach test improved in the virtual rehabilitation group as it mimics tasks trained within the intervention; there is sufficient evidence to support the notion of specificity of practice; that we get better at the things we practise (Rabadi 2011; Hornby et al 2011). There are some significant limitations to the trial; a division into high and low risk for falling was undertaken prior to randomisation, although it is unclear about how this was used. It is not known whether stratification or minimisation procedures were used to even out the distribution of participants with high or low risk of falling. There was no indication of the power of the study and importantly, there was no follow up to determine if these differences in the Berg Balance and the Functional Reach scores persisted over time. It is relevant to note that the scores on the Berg Balance Score were on average four points lower in the intervention group at the start of the trial, therefore this group had, in effect, more room to improve. The authors do state that there were no baseline differences according to a t-test of the baseline scores. However, the practice of testing baseline scores in this way has been criticised in the methodology literature and CONSORT guidelines (guidelines for conducting clinical trials) suggest that baseline comparison should not be made (www.consort-statement.org). A glance at the means and standard deviations of this rather small sample would lead one to suspect that the data were not normally distributed, thus implying that the parametric statistical analysis used was possibly inappropriate. It would have been beneficial for the authors to report the distribution of the data with some clarity.

On the basis of this single pilot randomised controlled trial, there is insufficient high quality evidence to support the use of the Wii BB as a rehabilitation tool to improve balance in adults and older adults. So, can we gain any insights from other research that has been conducted? Table 1 presents a summary of the research on the Nintendo Wii and balance. This time it is not restricted to randomised controlled trials but includes all trials that are looking for an effect (usability, feasibility and single case studies were excluded). There were three pre-post trials that fulfilled the criteria, which are summarised in Table 1.

In the trial by Nitz et al (2010) participants selected their own exercises and filled in a log to record their exercising habits. Of six measures of balance reported on in the pre-post trial, only one, the single leg stand test, showed a statistically significant improvement from pre to post intervention. However, it is unclear how this was measured as the results are reported in degrees per second, which implies that it was centre of pressure (CoP) excursion velocity, rather than time in single leg stance. It is not easy to consider if an improvement of 0.13deg/s is a clinically important change. The women in this study were relatively young with a mean age of 46.6 years (SD 9.9) and of the eight participants six were exercising several times a week at a self-paced or vigorous level. This is likely to reduce the applicability of this trial for older adults who are frailer or who have a falls risk, or a neurological condition. A glance at the raw scores on the balance tests prior to the participants commencing the trial would indicate that they were already performing at a high level in all balance measures except for the single leg stand. The authors suggested that larger, adequately powered trials should be completed to rigorously test the assumption that use of the Wii BB and Wii Fit games improves balance.

In the Williams et al (2011) trial participants (n=22) completed 10 minutes of Wii Fit balance exercises followed by 10 minutes of Wii Fit aerobic exercises. Participants in this trial had a mean age of 84 (SD 6), and so were likely to represent a group that would have some deficit in balance and some falls risk factors compared to the sample in the Nitz et al (2010) study. In the Williams et al trial the attendance rate was excellent, exceeding 98%. A further strength of the study was that the two assessors were checked for interrater reliability of the BBS. High reliability was reported (95%), although the method of determining reliability was not adequately described and unfortunately the assessors were not blinded to the study aims. The group mean for the BBS moved from a medium falls risk category (score of 21-40) into a low falls risk category (above 40), with a mean change in excess of 8 points indicating a likelihood of clinical significance. It was not clear if the researchers who were monitoring the rehabilitation sessions provided any stand-by assistance, or whether participants independently used the Wii systems. Overall this was a well-considered study and provides some indication that Wii BB and Wii Fit training may result in balance improvements for older adults.

Bainbridge et al's (2011) study included 8 community dwelling adults over the age of 65 years with a self-identified balance limitation. The training programme was standardised by the researchers and varied over the weeks to give different and increasingly difficult challenges. There was no statistically significant change in BBS scores in this trial. One of the possible explanations for this may be that four of the six people who completed the trial had high BBS scores to start with, so there was little room to improve. It is interesting to consider if it is more appropriate to select participants who have a falls risk that is carefully determined by the researchers, and therefore only include people who have greater capacity to improve on these types of pre-post intervention trials. The alternative would be to use outcome measures that challenge balance to a greater extent than the BBS does. The two participants who withdrew from this trial did so because of increased pain from hip osteoarthritis and increased low-back pain, assumed to have been related to the training programme. Few of the other trials reported adverse events.

There are some important limitations to the Wii BB system. The platform is relatively narrow and short and requires the participant to step up onto it (about 5cm in height) and therefore poses a falls risk. Some of the studies mentioned in this review had a therapist standing alongside the person exercising on the Wii BB for safety reasons (Bainbridge et al 2011; Gil-Gomez et al 2011). If this is the case then the argument that the Wii saves staff time is moot. Feedback is provided by the Wii system, but if performance has been consistently poor, according to the criteria set by the game rather than the therapist, then the feedback to the patient is negative and can cause disappointment, which may not be desirable as it can lead to the reinforcement of the sense of failure (Schubert 2010).

A key issue is to determine whether the Wii BB used with the WiiFit activities provides a reliable assessment of a person's balance performance. Sway measures, often depicted as movement of the CoP, are frequently used in research as measures of balance performance. Commercial force plates are usually used to measure CoP movement; these produce reliable, valid and precise measures of CoP movement and balance performance. In two studies Clark et al have found that the Wii BB was reliable for measures of CoP when compared to a commercial force plate (Clark et al, 2010; Clark et al 2011). However, in both these studies the data that were analysed were from customised software, rather than from the Wii Fit software. Interestingly, when the Wii Fit outputs were compared to the gold standard measure of posturography, the EquiTest, the Wii system did not fare well. The results showed that the CoP measures from the Wii Fit system were not reliable (ICC = 0.253) nor were they correlated with those from the EquiTest (r = 0.08) (Gras 2009). This suggests that the Wii Fit measures of CoP movement are neither reliable nor valid.

It is important to avoid confusing 'lack of evidence' with 'evidence against'. Research into virtual reality applications in rehabilitation is a new and growing area and there is limited research at this point. Not surprisingly, when we consider the use of a specific device, Nintendo Wii BB, and a specific type of intervention, balance rehabilitation in adults, we really are considering very few trials. From the small amount of moderate to low quality evidence currently available we have to be cautious if we choose to use the Wii BB and Wii Fit software in rehabilitation of balance disorders. It is important to closely monitor patient performance and measure outcomes using appropriate clinical measures. The Wii BB cannot replace good assessment, as it does not provide comprehensive assessment of the person's balance problem--at best it informs the therapist about a single factor relating to balance (movement of the vertical component of the ground reaction force) and at worst it provides unreliable information about CoP movement. Considering that balance is a multi-system function, it is not enough to use the Wii BB and Wii Fit software as the sole assessment and treatment tool. To correctly evaluate balance and determine which systems are demonstrating deficits a full balance assessment should be undertaken (Mancini and Horak 2010). There is evidence to suggest caution if using the Wii BB and Wii Fit software for assessment and treatment of balance disorders in adults and older adults.

KEY POINTS

* Whilst the Nintendo Wii balance board and associated games appear to offer activities that challenge balance there is currently no high quality evidence to support the use of these systems to improve balance in adults and older adults with balance deficits.

* There are some concerns about the reliability of the balance assessment components of the Wii system and evidence demonstrating poor correlation between computerised posturography measures and Wii system measures.

* If using the Wii systems in your rehabilitation practice it is vital to conduct a full clinical balance assessment and use the results from this to indicate change in your client rather than information from the gaming system.

REFERENCES

Bainbridge E, Bevans S, Keeley Band Oriel K. (2011). The effects of the Nintendo Wii Fit on community-dwelling older adults with perceived balance deficits: A pilot study. Physical and Occupational Therapy in Geriatrics 29(2): 126-135.

Clark RA, Bryant AL, Pua Y, McCrory P, Bennell K and Hunt M. (2010). Validity and reliability of the Nintendo Wii Balance Board for assessment of standing balance. Gait and Posture 31(3): 307-310.

Clark RA, McGough R and Paterson K. (2011). Reliability of an inexpensive and portable dynamic weight bearing asymmetry assessment system incorporating dual Nintendo Wii Balance Boards. Gait and Posture 34(2): 288-291.

Gil-Gomez JA, Llorens R, Alcniz M and Colomer C. (2011). Effectiveness of a Wii balance board-based system (eBaViR) for balance rehabilitation: A pilot randomized clinical trial in patients with acquired brain injury. Journal of NeuroEngineering and Rehabilitation 8: 30-39.

Gras LZ, Hummer AD and Hine ER. (2009). Reliability and validity of the nintendo[R] Wii fitTM. Journal of Cyber Therapy and Rehabilitation 2(4): 329-336.

Hornby TG, Straube DS, Kinnaird CR, Holleran CL, Echauz AJ, Rodriguez K, Wagner EJ and Narducci EA. (2011). Importance of specificity, amount, and intensity of locomotor training to improve ambulatory function in patients poststroke. Topics in Stroke Rehabilitation 18(4): 293-307.

Mancini M and Horak FB. (2010). The relevance of clinical balance assessment tools to differentiate balance deficits. European Journal of Physical & Rehabilitation Medicine 46 (2): 239-248.

Nitz JC, Kuys S, Isles R and Fu S. (2010). Is the Wii Fit a new-generation tool for improving balance, health and well-being? A pilot study. Climacteric: the Journal of the International Menopause Society 13 (5): 487-491.

Rabadi MH. (2011). Review of the randomized clinical stroke rehabilitation trials in 2009. Medical Science Monitor 17 (2): Ra25-Ra43.

Schubert T. (2010). The use of commercial health video games to promote physical activity in older adults. Annals of Long Term Care 18 (5): 27-32.

Williams B, Doherty NL, Bender A, Mattox H and Tibbs JR. (2011). The effect of Nintendo Wii on balance: A pilot study supporting the use of the Wii in occupational therapy for the well elderly. Occupational Therapy in Health Care 25 (2-3): 131-139.

Denise Taylor, PhD, MSc, NZSP

Associate Professor, Health & Rehabilitation Research Institute, Faculty of Health & Environmental Science, AUT University
Table 1: Table of key studies

Authors           Design           Participants

Gil-Gomez et      Pilot RCT        Acquired brain injury
al 2011                            n=17. Stroke (n=11),
                                   TBI (n=3) and benign
                                   cerebral neoplasm
                                   (n=3)
Nitz et al 2010   Pre-post trial   Healthy women
                                   aged 30-60 years
                                   n=10
Williams et al    Pre-post trial   Healthy older adults
2011                               n=22
                                   Aged 74-94
Bainbridge et     Pre-post trial   Community dwelling
al 2011                            older adults with
                                   balance problems
                                   n=8

Authors           Intervention

Gil-Gomez et      Wii BB with custom
al 2011           made software
                  1hour 3-5x per week.
                  Total of 20 sessions.
Nitz et al 2010   Wii BB and Wii Fit
                  software
                  30mins, 2x week for 10
                  weeks
Williams et al    Wii BB and Wii Fit
2011              software
                  20mins 3x week for 4
                  weeks
Bainbridge et     Wii BB and Wii Fit
al 2011           software
                  30mins 2x week for 6
                  weeks

Authors           Outcome *

Gil-Gomez et      BBS, BBA, FR (sitting), FR (stand)
al 2011           ST, TSC, 1MWT, 10MWT, TUG,
                  30CST
                  Significant difference between
                  groups on BBS and FR only.
                           Pre           Post
                  BBS      E=41 (11)     E=45 (9)
                           C=45 (7)      C=47 (6)
                  FR       E=24 (8)      E=27 (10)
                           C=25 (9)      C=26 (10)
Nitz et al 2010   TUG, TUGcog, ST, standing on
                  foam with eyes closed, SLS, LOS
                  Statistically significant
                  improvement on SLS only
Williams et al    BBS.
2011              Pre mean = 39.4 (6.3)
                  Post mean = 48.6 (4.6)

Bainbridge et     BBS, ABC, MDRT, COP excursion
al 2011           No statistically significant
                  difference on any test.

Authors           Comments

Gil-Gomez et      3 participants
al 2011           dropped out
                  and were not
                  included in the
                  analysis.
Nitz et al 2010   8 completed
                  the study and
                  included in
                  analysis.
Williams et al    Participants
2011              selected their
                  own games.
Bainbridge et     2 participants
al 2011           dropped out
                  and were not
                  included in the
                  analysis.

* only balance related outcomes reported

BBS = Berg Balance Scale, BBA = Brunel Balance Assessment,
FR = Functional Reach (sitting and standing), ST = Step Test,
TST = Timed Stair Test, 1MWT = 1 minute walk test,
10MWT = 10 metre walk test, TUG = Timed Up and Go,
TUGcog = Timed Up and Go with secondary cognitive task,
30CST = 30second Chair Stand Test, ABC = Activities-specific
Balance Confidence Scale, MDRT = Multi-Directional Reach Test,
COP = centre of pressure, SLS = single leg stand,
LOS = Limits of Stability, SD = standard deviation,
E= experimental group, C= control group
Gale Copyright: Copyright 2011 Gale, Cengage Learning. All rights reserved.