Virtual Reality Rehabilitation and Exergames—Physical and Psychological Impact on Fall Prevention among the Elderly—A Literature Review

: The present review is aimed at the effectiveness of virtual reality (VR) and exergames in the prevention of falls among the elderly. Falls become a signiﬁcant problem in the aging population and lead to psychological, social, and physical impairment. Prevention of falls is crucial to the well-being of the elderly population and is one of the challenges of contemporary rehabilitation. Recently, in view of the threat of the SARS-CoV-2 pandemic, contactless methods of rehabilitation, including telerehabilitation, appear as valuable rehabilitation tools. This review is based on the PRISMA guidelines and was carried out in ﬁve databases: PubMed, Medline, Web of Science, Scopus, and PEDro. Twenty-one randomized controlled trials, focused on the application of VR and exergames in the prevention of falls, were included. This review suggests that VR training in rehabilitation appears to be a promising complement to traditional techniques of physiotherapy to improve speciﬁc physical outcomes. VR and exergames could be considered as a complement of standard physiotherapy and its possible continuation at home for elderly. However, further high-quality studies, with carefully designed protocols and proper blinding, are needed.


Introduction
Falls become a significant problem in the aging population. Falls are prominent factors of accidental or unintentional injuries implied by external causes [1,2]. The incidence of falls amongst elderly people, according to the World Health Organization, varies from 28-35% of individuals aged 65 and over, increasing to 32-42% for those over 70 years of age [1]. Falls result in the fear of falling, loss of independence, institutionalization, and eventually death [3]. Strategies to prevent falls are crucial to the well-being of the elderly population, and rehabilitation plays an important role among them.
Standard rehabilitation protocol focused on the prevention of falls comprises kinesiotherapy (exercises) and education. Exercises should include training of body balance, coordination, gait, changing the body position, minimizing the possibility of an injury during a fall, and techniques of getting up properly after a fall. Exercises are effective in reducing the risk of falling [4].
Advances in medical technology resulted in the popularization of computer-assisted interventions in rehabilitation treatment. New technologies in rehabilitation comprise of a variety of biofeedback implementations into hardware platforms, motion capture systems augmented by biofeedback, augmented reality (AR) systems, and virtual reality (VR) systems. Technological appliances vary from adoptions of popular gaming hardware platforms to specialized systems with dedicated hardware and software. Software protocols, presuming adequate hardware implementation, can be integrated into a complete computer game, which may result in better adherence. An application of virtual reality Recently, in view of the threat of the SARS-CoV-2 pandemic, contactless methods of rehabilitation, including telerehabilitation, appear as valuable rehabilitation tools [20]. Rehabilitation is an important and necessary part of the treatment and prevention of many diseases. Therefore, its safe continuation during the pandemic prefers remote methods. Nowadays, healthcare has to deal with the challenges of providing patient care using modern technologies [21]. Their employment involves economic costs, employee training, and adaptation of the equipment to the treatment goals and the age profile groups of patients. The use of new technologies in rehabilitation carries the risk of exclusion of elderly people. Therefore, knowledge about the possibilities of the use and the adaptation of virtual reality in such a case appears to be particularly important.
The influence of interventions based on new technologies on the course of rehabilitation, regarding their hypothetical advantage over standard rehabilitation, still remains an open question. This statement is also valid in the challenging problem of the prevention of falls among elderly people.
The present study was conducted to determine, based on the current literature review, whether the virtual reality and exergames employed in rehabilitation significantly reduce the incidence of falls among elderly patients. An additional aim of this review was to assess the impact of this type of intervention on patients cognitive functions, quality of life, adherence, usability, and enjoyment of exergames in the process of rehabilitation.

Literature Search Strategy
The aim of the literature review was the evaluation of the effectiveness of virtual reality rehabilitation among elderly patients in the prevention of fall risk. The methodology was based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and conducted following the PRISMA checklist [22]. A literature review was carried out in six databases: Medline, PubMed, Scopus, Embase, Web of Sciences, and PEDro.
The tool used to create the research objective and search strategy was PICOT(T) [23].
• Population-healthy elderly • Intervention-virtual reality, exergames rehabilitation • Comparison-any intervention (standard rehabilitation, physical exercises, daily living activities) or no intervention • Outcomes-fall risk and prevention, body balance, walking, functional improvement, pain, cognition, quality of life • Time-last 5 years (January 2015-December 2020) • (Type of study)-randomized controlled trials The primary search keywords were "elderly", "virtual reality", "exergames" "rehabilitation", "falls", and their synonyms. Keywords were combined using the Boolean operators "AND" and "OR". The search strategy was flexible and adapted to the database search engines. For the PubMed, we used the following search scheme: ((((virtual reality OR augmented reality OR video games OR game * OR computer games OR wii OR exergames OR vr))) AND (((rehabilitation OR exercise * OR physical therapy OR physiotherapy OR physical activity OR activity OR training))) AND ((elderly OR aged OR older OR elder OR geriatric OR elderly people OR old people OR senior))) AND (((fall prevention OR preventing falls OR prevent falls OR falls OR falling))) AND (Randomized Controlled Trial[ptyp]). Each search strategy was checked by two authors.

Study Selection and Data Extra Action
Firstly, two authors created the criteria for eligibility. When both authors had accepted the search strategy, one of them searched the databases (last search: January 2021). Then, two researchers reviewed all the studies (title, abstract, and full text) independently, using an automation tool-Rayyan [24]. The results were compared and all occurring conflicts at every stage of study selection and data extraction were solved through discussion.
Articles meeting the following criteria were included in the review: (1) published in the last 5 years (January 2015-December 2020); (2) English language; (3) randomized controlled trials (RCT); and (4) targeted for the healthy elderly, rehabilitation, and exergames. Studies were excluded due to: (1) publishing before 2015; (2) writing in a language other than English; (3) reports from books, case reports, observational, as well as interventional (non-randomized) clinical trials, and protocols of RCTs; and (4) do not concern the elderly, rehabilitation, and exergames.
The data was extracted from articles by two authors. Quantitative variables were presented as mean, minimum, and maximum values. Qualitative variables were described as percentage frequencies.

Quality Assessment
The methodological quality was assessed using the PEDro scale, which is considered relevant for RCTs related to physical therapy [25,26]. The scale evaluates internal validity of the trials (criteria 2-9) and information on statistics (criteria 10-11). Criterion 1 relates to external validity but is not considered when calculating the final score. The PEDro scale is only used to evaluate the research methodology. Therefore, the PEDro results do not provide evidence of clinical utility and cannot assess the effectiveness of an intervention in clinical practice [27].
The methodological evaluation of the included articles that were available in the PEDro database was taken from the website https://www.pedro.org.au/ (accessed on 5 January 2021).The remaining three articles [28][29][30] were rated on the 11-points PEDro scale by two authors independently, then the results were compared, and potential differences were resolved during the discussion. PEDros' items were scored 0 if not reported or unclear, 1 when reported adequate. Research that received from 1-3 points was rated as poor quality, from 4-6 points was moderate quality, and above 7 points was assessed as high quality.

Searching Results
The selection of articles was carried out in accordance with PRISMA guidelines. After searching the databases and removing duplicates, 682 articles were screened by titles and abstracts. As a result of the selection, 114 articles were included in the analysis of the full texts. After the full-text selection, 93 publications were rejected and 21 studies were included in the qualitative synthesis. For details on the selection of studies and the reasons for their rejection, see Figure 1. PRISMA flowchart.

Characteristics of Included Studies and Participants
Twenty-one RCTs were included in the analysis, involving a total of 1557 participants who were randomized to the study or control group, respectively. These were the healthy elderly with a mean age of 75.7 years who were examined for the risk of falls. Among the patients, the majority were women (female-to-male ratio of 1.6:1). Participants were recruited mainly from seniors residing in retirement homes and among rehabilitation ward inpatients. Post-intervention follow-up was used in 6 out of 21 studies and ranged from 6 weeks [31] to 1 year [32].

Technology of Intervention
The main tool used for rehabilitation was virtual reality games. The exercise program was aimed at increasing muscle strength and improving body balance. In most cases, the training was based on Nintendo Wii balance boards [33][34][35][36][37], biofeedback [32,38], VR dancing [32,39], and exergames obtained for various physical exercises [28,[39][40][41][42][43]. In addition to physical exercises, some studies also used exercises aimed at cognitive function and memory [32,39,40,43]. The experimental group received training under the supervision of a physiotherapist or trainer in 13/21 of the included studies. Unsupervised training focused on interventions at the patient′s home [30,39,42]. The control group also consisted of the elderly who were usually treated with standard physiotherapy including balance exercises or preventive leaflets. The duration of the intervention ranged from a minimum of 10 days [44] to a maximum of 6 months [32], an average of 8.9 weeks. It should be noted that in 8 out of the 21 studies, an intervention in the experimental group was based on VR training and the same standard intervention as in the control group (e.g., exercises or education leaflets). Other research works compared VR training standalone in the experimental group and usual care in the control group. Table 1 presents more information on the intervention use.

Technology of Intervention
The main tool used for rehabilitation was virtual reality games. The exercise program was aimed at increasing muscle strength and improving body balance. In most cases, the training was based on Nintendo Wii balance boards [33][34][35][36][37], biofeedback [32,38], VR dancing [32,39], and exergames obtained for various physical exercises [28,[39][40][41][42][43]. In addition to physical exercises, some studies also used exercises aimed at cognitive function and memory [32,39,40,43]. The experimental group received training under the supervision of a physiotherapist or trainer in 13/21 of the included studies. Unsupervised training focused on interventions at the patient s home [30,39,42]. The control group also consisted of the elderly who were usually treated with standard physiotherapy including balance exercises or preventive leaflets. The duration of the intervention ranged from a minimum of 10 days [44] to a maximum of 6 months [32], an average of 8.9 weeks. It should be noted that in 8 out of the 21 studies, an intervention in the experimental group was based on VR training and the same standard intervention as in the control group (e.g., exercises or education leaflets). Other research works compared VR training standalone in the experimental group and usual care in the control group. Table 1 presents more information on the intervention use.

Quantitative Outcomes
Quantitative outcomes were assessed as the physical performance of patients (Table 2) and mainly contain the following activities: balance, gait ability, risk of falling, and muscle strength. The main tools used to evaluate the above variables were the Berg Balance Scale in 8/21 studies, the Timed Up and Go Test and its variants in 10/21 studies, the Falls Efficacy Scale-International and its variations (FES-I) in 6/21 studies, and the 10-to-2-min march test in 4/21 studies. Short Physical Performance Battery (SPPB) was used in 6/21 studies to evaluate physical activity and risk of disability in activities of daily living.
The assessment of the impact of VR rehabilitation and exergames on physical function among the elderly varies in the analyzed studies. In 11 out of 21 studies, there was a significantly greater improvement among assessed specific outcomes for VR training than conventional therapy. Moreover, 8 of the 21 pieces of research assessed VR-based intervention as being effective as traditional treatment. On the other hand, 2 of the 21 analyzed studies [39,44] indicate no special benefits for VR intervention in comparison to standard therapy. These results indicate that new technologies used in rehabilitation could be equally effective in the improvement of specific physical function among elderly patients. However, the diversity in technologies used and research tools among included studies makes impossible to draw a clear conclusion that any intervention based on VR will be effective in the elderly in terms of improving physical function. Additionally, in the assessment of statistical significance of the results, many studies used only p-value to misinterpret the effectiveness of the application of the intervention. However, p-value only shows that an effect exists and does not reveal the size of the effect. In quantitative research, both p-value and size effect should be reported. Therefore, the lack of information on the size effect does not allow a full assessment of the effectiveness of the intervention. Both interventions improved physical outcomes, but there was no significant difference between the SG and CG in measured outcomes.
Adcock M. et al. [39] Gait analysis, single task walking and dual task walking, balance, leg strength and endurance, brain plasticity/brain volume Step length, step speed, Delbroek T. et al. [43] Balance, gait, dual-task performance (cognitive-motor) Tinetti-POMA scale, iTUG, iTUG with visual task No changes in the Tinetti-POMA. VR training is more effective than conventional treatment only in dynamic balance in single task walking-iTUG improved after 6 weeks training in the IG (17.2 s vs. 15.8 s, p = 0.02).

Qualitative Outcomes
Among the studies included, 14 of them (63.6%) assessed qualitative variables such as cognitive function, quality of life, adherence, useability, and enjoyment of exergames. Table 3 presents detailed results of qualitative outcomes. It has been noted that the use of virtual reality rehabilitation is cost-effective, as compared to standard geriatric rehabilitation in the OTAGO system, due to the reduction of costs related to personnel work, their training, and community [42]. Additional virtual reality equipment does not significantly increase the costs associated with this type of therapy [38]. Most authors highlight the safety and absence of adverse effects in VR training for geriatric patients [31,33,34,39,40,42,47]. However, some adverse events were reported, such as pain during exercises [44], knee joint and tight soreness [45], or those unrelated directly to interventions [30]. Exergames were assessed as an acceptable and pleasant form of treatment in five out of seven studies. In five of nine studies, the VR rehabilitation significantly improved various cognitive functions among the elderly. An assessment of the quality of life among participants using VR training indicates that they mostly achieved greater improvement in QoL parameters than controls (four from five studies). This could be caused by the attractiveness of this kind of intervention. The conclusions formulated above suggest a positive [clear] influence of VR training on those qualitative variables. However, there was no influence of exergames on decreasing the symptoms of depression. This section may be divided by subheadings. It should provide a concise and precise description of the experimental results, their interpretation, as well as the experimental conclusions that can be drawn. The adherence, motivation, and enjoyment were significantly higher in standard exercise training than exergames.
Levy F. et al. [29] Fear of falling, depression, anxiety BDI (21-item version), STAI-Y-A, STAI-Y-B No significant difference between groups in depression, trait anxiety, or improvement in VR group compared to controls only in the case of state anxiety.
van den Berg M. et al. [31] Quality of life, feasibility, safety, usability, enjoyment Euro-QoL EQ-5D, RNLI, SUS, PACES Participants were comfortable with the technology and that the equipment was easy to use and enjoyed the intervention. Significant improvement of quality of life in VR group compared controls.
Mirelman A. et al. [38] Cognition, health-related quality of life EFI, AIS, SF-36 Cognitive function outcomes improved similarly in both training groups. However, the quality of life improved more in VR group. Eggenberger P. et al. [32] Depression GDS No significant difference between groups.
Gschwind Y.J. et al. [47] Quality of life, depression, cognitive measures, executive function, adherence, usability Euro-QoL EQ-5D, PHQ-9, VST, DSB, ANT, SUS, PACES, DART Significantly higher adherence, improvement in quality of life, and executive functioning were found in VR group in comparison controls. The intervention was enjoyed by the participants, and its usability was acceptable.

Quality Assessment
The PEDro scale was used to evaluate the methodological quality of the included studies (Table 4). Among the 21 studies, 4 out of 21 (19%) were of high quality (10-7 points) and 17 out of 21 (81%) were of moderate quality (6-4 points). The average quality rating of the analyzed articles was 5.8 points (moderate quality). The main sources of bias were the lack of blindness for participants and therapists (all studies), which was impossible to obtain for this intervention, and the lack of application of intention to treat data analysis (only 7 from 21, 33.3%).

Discussion
The aim of the present review was to determine the effectiveness of a rehabilitation accommodating virtual reality and exergames in view of fall prevention among elderly people.
The progress of technology resulted in the utilization of virtual reality in medical appliances [49]. Virtual reality techniques have been widely adopted in other fields of medicine, including specialist training like simulations of surgical procedures [50], pain management [51,52], treatment of phobias or anxiety [53], and diagnostics [54]. Growing interest regarding virtual reality appliances and a wide variety of available software applications and hardware devices resulted in wider adoption in rehabilitation in view of making the rehabilitation more attractive, reducing total costs related to human and infrastructural factors (telerehabilitation), as well as considering maximizing the effects of traditional physiotherapeutic proceedings.
Limitations of motor and cognitive functions, aggravating with age, result in a vast representation of elderly people among patients who are a subject of rehabilitation [55]. Nowadays, virtual reality rehabilitation is considered as an intervention targeted at children [56], adolescents and adults [57], athletes [58], as well as the elderly. However, varying motivation was observed among the age groups-children and adolescents are motivated by a challenging part of exergames like competition and gaining rewards/points in a game.
On the other hand, older adults focus on perceived health benefits and enjoyment without competition [59]. Similarly, gaming was perceived as a domain of young people. Besides, one should remember that our contemporary digital world was created by a generation of people, who are currently in their 50s. Recent studies have proved that older people draw substantial satisfaction and present good acceptance of virtual reality applications in the process of rehabilitation [33,[42][43][44][45]47].
Moreover, virtual reality gives many opportunities to provide patient rehabilitation. It appears as a valuable rehabilitation tool regarding older people used in nursing homes, activity centers for the elderly, hospitals, and rehabilitation clinics, where qualified staff take care of the proper adoption and safety of interventions. The development of modern technology allows using VR rehabilitation in patients' homes successfully by means of telerehabilitation [39,42,47]. Telerehabilitation increases motivation to home exercises and continuity of rehabilitation after discharge from the hospital [60,61]. This is especially important during the coronavirus pandemic where many people must stay at home and access to indirect rehabilitation is limited. Additionally, VR rehabilitation seems to be effective in reducing the cost and time of personal staff and the daily commute to the health center [42]. However, most previous implementations relied on telerehabilitation, hence there is only scarce evidence of virtual reality in this field.
This review suggests that virtual reality training in rehabilitation appears to be promising complementation or even an alternative to traditional techniques of physiotherapy. However, several factors, which can limit its wider application among elderly people, should be considered. Seniors are frequently reluctant in unassisted habituation of remote rehabilitation, which could be partially explained by still cursory knowledge of modern technologies. Such a situation can be solved by the implementation of possibly simple systems, adopted to requirements and capabilities of elderly people, provided with capabilities to reversely monitor the activity of patients, proper execution of exercises, and supporting mechanisms of motivation.
Regarding the quantitative outcomes of this review, the question of whether virtual reality offers greater effectiveness over traditional rehabilitation shifts towards a positive answer. The analysis of studies included in this review suggests that rehabilitation accommodating virtual reality, aimed at improving physical functions, including body balance, gait, and muscular strength, has some effectiveness. However, due to differences in applied technology and protocols, it is difficult to draw clear conclusions. These findings are in compliance with the research works of other authors. Pacheco et al. [12] found exergames to improve balance and mobility in older adults with impairments without neurological diseases. Accommodation of those elements of motor activity in exergames in the rehabilitation resulted in an improvement of the body balance among elderly people [62]; however, the significant effect was related to the dynamic and perceived balance, but not to the static balance.
Several studies show the benefits of tai-chi, yoga, and rehabilitation exercises programs regarding the decreased risk of falls among elderly people [63][64][65]. Although their effectiveness is undebatable, elderly patients currently need remote intervention to keep their mobility safe. Therefore, virtual reality rehabilitation and exergames begin playing a role in telerehabilitation, and as is proved in this review, they constitute an equally as effective form of intervention as traditional treatment in view of mobility and balance improvement, as well as reduction of the risk of falls.
An important factor of rehabilitation of elderly people, which should be considered, is a deterioration of cognitive functions during aging. This review mostly found exergames to be an acceptable and pleasant treatment. In most studies, exergames significantly improved various cognitive functions, as well as the quality of life among elderly with a risk of falling.
Benefits regarding the health-related quality of life were also observed in elderly people who underwent virtual reality training. The level of satisfaction and enjoyment of such interventions was appraised positively [66]. There is also evidence of the positive motivational aspect of exergames among older adults [67].
In view of these qualitative outcomes, only a few authors have reported adverse effects, which were identified chiefly as non-related to intervention [38,44,45]. Incidentally, the comparison of adverse effects during VR rehabilitation in relation to traditional rehabilitation builds up another interesting research question.
Although exergames were assessed as an effective intervention in alleviating symptoms of depression [68], the studies in this review do not confirm this thesis. However, differences in the results could be caused by the participants' level of functional abilitiesthere were elderly with decreased mobility; number of sessions-the duration was long; and the games' profile was centered around rehabilitation and functional performance, not enjoyment. Li et al. [68] indicated that the influence of exergames on depression is more effective in patients with relatively good mobility who play highly playful games in small numbers of sessions.
Contemporary literature reviews regarding the effectiveness of virtual reality in orthopedic rehabilitation suggest that the variety of devices used, the variety of their functions, and a lack of coherence in applied protocols result are significant problems related to the comparison of the acquired results [69]. The analysis of the papers included in the current review also varied regarding the period of the intervention, adopted devices, and trained functions. A multitude of applied interventions and equipment, as well as a variety of tools and methods of evaluation of the effectiveness of those interventions, make it difficult to conduct a metanalysis.
Conclusions of the analyzed studies do not allow to draw up any precise clinical guidelines regarding the duration, frequency, types of devices, and software implementations, as well as particular ailments and diseases in which virtual rehabilitation and exergames should be indicated.
This review has some limitations because it concerned only randomized controlled trials published in the last 5 years in English. The reason is that RCTs are considered a gold standard in medical sciences and give reliable information about the effectiveness of the applied intervention [70]. In the last decade, there has been a rapid growth of VR clinical application papers published in PubMed [71]. Also, the quick development of modern technology meant that only the newest papers gave the opportunity to assess the current state of art in this topic. Therefore, there is a risk that other valuable reports would not be included in this review.
In many studies, only clinical tests were used to assess physical outcomes, which provide valid results, but are fraught with a risk of bias due to human mistakes. Regarding those factors, future studies should use standardized biomechanical methods of assessment based on objective diagnostics tools.
Common limitations of the included trials were the small sample size [29,33,37,43,45,46] and the lack of a long-term follow-up (6/21 studies with follow-up), as well as no blindness among participants and therapists (all studies). However, blindness in this kind of intervention is difficult to achieve but long-term follow-up and a proper study sample size are necessary to obtain reliable results.
Exergames can be considered as an evolution to virtual reality, and vice versa. Virtual reality techniques allow deeper immersion during training. Exergames involve better interest in training [5]. Besides, it is interesting to assess whether virtual reality and gaming simply add to interest and immersion or whether they act in a supra-addition.
The current review supports the hypothesis that virtual reality and exergame application among elderly people can positively influence physical functions, as well as cognitive functions of seniors. However, virtual reality and exergame rehabilitation were obligatorily implemented in parallel with traditional rehabilitation and thus are mostly considered as a complement of standard physiotherapy [11,43], as well as its continuation at home [35,36,45].

Conclusions
There is an evidence that interventions based on virtual reality and exergames are effective in improving specific physical and psychological outcomes among elderly.
Interventions based on virtual reality and exergames are promising in view of multifactor management, regarding motor and cognitive functions, as well as the quality of life and enjoyment. The use of virtual reality is cost-effective, safe, and bereft of adverse effects. Future research studies, notably randomized controlled trials, with well-defined research protocol, are desirable to determine the effectiveness of virtual reality in fall prevention.