Study design and methodological quality

Table 2 summarises the methodological quality of each study. Adequate details were provided to replicate the studies. The aims, inclusion and exclusion criteria, and exergaming interventions were well described. Six studies used commercial Wii games [19–24] and one developed a game designed for PD using the Sony Playstation Eye™ camera [25]. Participant demographic characteristics and the severity of their PD were generally well reported although the use of several different scales of disease severity (Hoehn and Yahr [26], UPDRS [27] and disease duration) made comparisons between studies difficult and may limit future meta-analyses. The rationale for sample size was provided in three studies.

Sample Characteristics

The number of people with PD in each study ranged between 1 and 32 (see Table 3). A broad age range was represented across studies with participants ranging from 44-91 years, and both males and females were included. People with PD had relatively mild symptoms (Hoehn & Yahr stage I & II) and were tested whilst on their medication, although medication status was not reported by Assad et al. [25].

Safety

Feasibility

Three studies measured performance in gameplay before and after an exergaming intervention (14 sessions [20, 21] or 16 sessions [22]). In all three studies, people with PD improved their performance on the games. Two of these studies tested participants 60 days after the cessation of the intervention and reported that improvements in gameplay were retained [20, 21]. However, Mendes et al. also found that people with PD failed to improve in two games that required fast reactions in response to virtual targets and did not learn as quickly as controls in those games that required dual tasking [20].

Only one study offered a detailed review of the game playing experience [25]. Good levels of motivation during game play were reported although difficulties with the fast pace and cognitive complexity of some games were raised. Esculier et al., presented the only study to use home-based exergaming and used participant logbooks to assess satisfaction of the exergaming intervention [19]. Fifty percent liked the exercise program very much, 33% liked the program, and 17% rated the program as neutral, with no subjects disliking the program.

Effectiveness

The only randomised controlled trial included in our review tested the effects of playing commerical games on the Nintendo Wii for 14 sessions (over 7 weeks) compared to traditional balance training in 32 people with mild PD (16 in each group) [21]. The primary outcome was section II of the Unified Parkinson’s Disease Rating Scale (UPDRS) and secondary outcomes included balance and cognitive tests. Both groups also received stretching, strengthening and axial mobility exercises. There were no differences between the two training groups at baseline and both groups improved by the same amount after 7 weeks on the primary or secondary measures. When data from both groups were combined, small improvements were observed in the UPDRS II, Berg balance scale, single leg stance (eyes open and eyes closed, but not eyes open with a dual task) and Montreal Cognitive Assessment (MoCA) immediately after training. These improvements were retained 60 days after the final training session.

Five further studies reported improvements on clinical tests in response to an exergaming intervention using commercial Nintendo Wii balance board (Wii Fit) games [19, 20, 22–24]. However, none of these studies included a control intervention. First, Esculier et al. showed that 18 sessions (over 6 weeks) of exergaming resulted in people with PD being able to stand on one leg for longer (with eyes open but not closed), improved bipedal standing balance with their eyes open (but not closed), and faster performance on the timed up and go, sit to stand and 10 m walk tests [19]. Participants also scored better on the Community Balance and Mobility scale after the intervention. Second, Mendes et al. found 14 sessions of exergaming elicited improvements in the functional reach test immediately after the last training sessions, which was retained 60 days after the last session [20]. Zettergren et al. showed that the gait speed, timed up and go, and Berg balance scale all improved in a case study of one 69 year old man with PD [22]. Mhatre et al. found significant improvements in Berg Balance Scale, Dynamic Gait Index and static balance over an 8 week intervention, [23]. Hertz found significant improvements in in the Nottingham Extended Activities of Daily Living Test (NEADL) at post intervention (12 sessions), however these improvements were short lived as the NEADL scores returned to baseline levels four weeks post intervention. Quality of life as assessed by the PDQ-39 showed significant improvement post intervention for ADL, emotion, communication, bodily discomfort and total score. At 4 week post intervention only ADL and emotion remained significantly improved with the addition of mobility significantly improving. The UPDRS motor score showed a significant decrease post exercise and improvements were retained 4 weeks post intervention. Other outcome measures which showed significant improvement post intervention were the timed up and go (TUG), timed tap (right side, Purdue score (left-side) and 9 – hole peg test score (right side).

Safety

More evidence is needed regarding the safety of people with PD playing exergames before it can be recommended for wide spread clinical use, particularly in home-based settings. For this reason, we recommend future studies report both objective and subjective measures of safety. Importantly, six of the seven studies we reviewed used the Nintendo Wii and Wii fit balance board [19–24]. The Wii fit is a raised platform and so may present a trip risk for people with PD, particularly when they are focused on the television screen. New commercial exergaming systems are now available, such as the XBOX Kinect™, that do not require a raised platform. Using these systems may improve the safety of exergaming for people with PD, however this remains to be tested, as does the clinical efficacy of these new exergaming systems.

Feasibility

Three studies reported improvement in gameplay performance in PD participants using commercial Nintendo Wii exergames [20–24], suggesting commercial games are feasible. Two studies that sought participant feedback about gameplay and reported that people with PD enjoyed playing the games and were motivated to play them [19, 25]. Despite these promising results and the potential of commercial exergames as a means of low-cost home-based exercise, there is concern that the commercial games are too difficult for some people with PD [28]. This reflects the need for appropriate game selection and design. A key advantage of exergames is that they can provide immediate biofeedback of performance. This is a useful attribute that may be exploited to improve motor learning in people with PD, especially given their increased reliance on visual cues. Setting an appropriate threshold of difficulty is essential. If feedback is too negative, for example when playing games that are too fast or complex, motivation, adherence and safety may suffer. This was particularly highlighted when Mendes et al. assessed motor and cognitive ability of game play using the Nintendo Wii for PD and healthy elderly adults [20], and showed that PD participants failed to improve on games that required fast decision making and movements to avoid virtual obstacles.

Of the reviewed papers, Assad and colleagues were the only authors to design their own exergame specifically tailored towards people with PD (WuppDi) [25]. Five games were developed to rehabilitate upper body movements and elements of cognition. Games were played, either with a hand held marker (wooden stick) or with no markers. The results showed that the majority of the participants enjoyed the experience and gained positive feedback from playing the games, and would enjoy playing with others especially their grandchildren. However, the authors noted some of the games were too complex and needed to be made easier for people with PD. Some participants also had difficulty with the hand held controllers. Hand held devoices may not be appropriate for people with severe tremor or dyskinesia who could find it difficult to keep a hold of and manipulate the controller.

Effectiveness

There is also little known regarding the efficacy of exergaming as compared to traditional rehabilitation. Pompeu et al. has conducted the only randomised controlled trial of exergaming for people with PD [21]. The findings of the study suggest that 14 sessions (over 7 weeks) of playing the Nintendo Wii™ can improve clinical measures of balance to the same extent as traditional balance training in people with PD. A potential confounder to this study was that participants in both groups also received stretching, strengthening and axial mobility exercises before each session. As such, it is difficult to conclude whether changes in the clinical tests can be ascribed to the balance training (either traditional or exergaming). Bateni found similar results when analysing the effects of the Wii on balance in healthy elderly adults, in that a combination of the Wii and standard balance training had the greatest improvement in balance outcomes at post test compared to Wii or balance training alone [29]. Therefore, it is possible that exergaming may be more useful for PD as an adjunct to standard clinical treatment than as a stand alone intervention.

Six of the seven studies examined whether people with PD improved on clinical tests after an exergaming intervention [19–24]. All six showed that people with PD improved on various clinical measures of balance (Berg balance score, single leg stand, functional reach test), motor function (Sit to stand, Timed up and go, 10 m walk, timed tapping) and severity of PD motor symptoms (UPDRS II). Two of these studies also showed that improvements in clinical tests were retained 60 days after the intervention [20, 21]. With Hertz et al. showing improvements were attained 4 weeks post intervention for PDQ-39 ADL and emotion, and motor scores for UPDRS [24]. These results indicate that exergaming may be effective for the rehabilitation of motor and balance symptoms in people with PD. However, more rigorous research designs need to be adopted in future trials to confirm whether improvements in these tests are clinically meaningful and are not due to increased familiarity with the clinical tests.

One of the potential benefits of exergaming interventions is that they can be administered in the home [17]. Although not formally tested, the resources required for exergaming are likely to be less than those required by formal rehabilitation services. To date, there has only been one home-based exergaming intervention reported in people with PD [19]. The authors were able to show that people with PD improved on several clinical tests and on gameplay over the 6 week training period. Questions remain, however, whether home-based exergaming interventions are safe and as effective as traditional clinic based interventions for people with PD.

Limitations

We included only full articles in this review, which resulted in the exclusion of two relevant abstracts relating to exergaming and PD [30, 31]. Dos Santos et al., showed that 14 sessions of dance based exergaming using a balance board resulted in improved balance and reduced motor symptoms (UPDRS) in people with PD [30]. Alvarez et al. trained 12 people with PD using three commercial XBOX Kinect™ games. Twenty-four sessions over eight weeks resulted in improvements in the UPDRS (II&III) and Tinetti scale for gait and balance [31]. Another limitation of this review was the heterogeneity of the different exergames included in each intervention limited our ability to distinguish which particular games, and which aspects of those games, are useful for rehabilitation of specific symptoms in PD.

Recommendations

This systematic review indicates that exergaming appears feasible for people with PD, however it also highlights the current paucity of evidence for exergaming as a rehabilitation tool. The studies we reviewed adopted relatively weak designs and small sample sizes. As such, we can not state with confidence whether or not exergaming is a safe and effective rehabilitation tool until larger randomised controlled trails are conducted, especially in a home-based setting. In addition, exergaming has only been tested in people with mild PD. To address these concerns, we recommend that large, robust randomised controlled trials be conducted in order to establish the safety and effectiveness of exergaming for rehabilitation in people with PD with a range of disease severity. In addition, to facilitate future reviews and meta-analysis of exergaming in PD we recommend i) future studies should report standard measures of disease severity (Hoehn and Yahr, and UPDRS III), medication status (levodopa equivalent daily dosage) and cognitive status (MMSE or MoCA); and ii) objective and subjective measures of safety, and the participants’ ability to play the games should be reported.

This review has also identified several important considerations when designing games for people with PD. To improve the effectiveness and adherence to exergaming interventions in people with PD, we recommend games designed for PD should: i) target specific clinical features of PD; ii) be easier than commercial games; iii) avoid negative feedback; iv) include very clear instructions and goals; iv) introduce cognitively demanding aspects slowly and sparingly; and v) examine the use of new exergaming systems that do not require balance platforms or handheld controller.