Research question

The research question guiding this scoping review was: what is known about GBT to enable physical therapy in home settings for people with PD? Through this review, we aimed to: (1) explore the key features of technology systems that have been used with the PD population; (2) describe their effectiveness for PD patients; and (3) identify gaps and challenges in the adoption of GBT physical therapy solutions for PD. As noted, no studies have focused on the implementation considerations of using GBT paired with physical therapy with the PD population. Mapping out the different technologies, approaches, and end-user experiences is critical for informing clinical practice and for supporting the design of future studies. Hence, a scoping review centered on implementation considerations can offer valuable insights regarding the feasibility of utilizing GBT physical therapy-based approaches in home settings for individuals with PD, and for identifying strategies to optimize their adoption by both patients and healthcare providers.

Information source and search strategy

As the topic is related to both the medical field and engineering, relevant studies were identified through the searching of four major electronic databases (OVID Medline, OVID EMBASE, CINAHL EBSCO, and Scopus) between January 2010 and May 2024 In order to retrieve information from the electronic databases, all possible combinations of the following key word strings were used: Game OR gamification OR user-computer interface OR software, mobile OR cellphone OR iOS OR Android, app OR application OR program, Rehabilitation OR Physiotherapy OR Physical therapy, Tele OR Remote OR Virtual OR home exercises OR Unsupervised. An expert librarian supervised the development of the search strategy. Details of the search terms and search strategies in all four databases are shown in S2 Appendix. Our results from each database were exported to a systematic review software system, Covidence, which automatically de-duplicated the extracted data [23].

Inclusion and exclusion criteria

The inclusion criteria were studies of any design (quantitative, qualitative), published in English with adult PD individuals (18 years old and older), and that developed and/or evaluated technologies (software and/or hardware) to facilitate home-based physical therapy exercises. As well, the study needed to incorporate a gamification element. To ensure alignment with current standards in the rapidly evolving technological landscape, our inclusion criteria focused on studies published from 2010 onwards.

We excluded (1) book chapters, conference papers, commentaries, gray literature, review papers, mini-reviews, and letters; (2) studies that did not include gamification features/elements (e.g., reward collection, interactive, engaging, fun, brain stimulation to boost motivation); (3) immersive-VR; and (4) studies that only presented the conceptual designs of their systems (i.e., no PD patient involvement or no device development).

Study selection process

The selection of studies was conducted in two phases: (1) screening of titles and abstracts and (2) full-text screening. Both phases were conducted independently by three authors (SNG, SM, ZL), who undertook an inter-rater screen of the first 80 abstracts (Phase I) and the first 30 full-text articles (Phase II) to test their rates of agreement. An agreement of 80% was set as the target, which was obtained in the title and abstract screening and full-text review phases. Any disagreements were discussed in consultation with a pre-identified fourth reviewer (SLH).

Data extraction

A review of a subset of articles was conducted to ensure the accuracy of data extracted by the authors (SNG, SM, ZL). The remaining articles were then divided across the three reviewers for data extraction. The extracted data encompassed study purpose and design, features of the software and hardware systems (i.e., game system, sensors, functional features, etc.), characteristics of the patient population (i.e., age, sex, etc.), intervention (exercise programs, type of exercises, etc.), adverse effects associated with using the technologies, and study outcomes.

After reviewing the abstracts and identifying papers for inclusion, three reviewers (ZL, MA, SS) conducted a quality assessment of each relevant article using the Downs and Black (D&B) tool [24]. Any disagreements in scoring were resolved through consultation with the fourth author (SM). This tool is comprised of 27 questions designed to assess external and internal validity, with a focus on evaluating bias and confounding factors. The last question in the tool, originally scored on a 0–5 scale, was adjusted to a 0–1 scale. A score of 1 was given if the study included a power or sample size calculation, while a score of 0 was assigned if neither was provided nor any explanation on the appropriateness of the sample size. Consequently, the maximum score any reviewed article could achieve was 28, with a higher score reflecting higher methodological quality. The Sackett Scale was then used to assign a level of evidence to each study (See Table 1), with level 1 representing the highest level and level 5 representing the lowest [25].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Sackett Scale level of evidence.

https://doi.org/10.1371/journal.pone.0326705.t001

Study characteristics

All 24 studies used a quantitative methodology. This included: five clinical studies [35,36,40–42], three observational clinical studies [26,27,34], five pilot clinical studies [30–32,39,49], one non-randomized pilot study [37], three single-blind randomized controlled trials (RCT) [45,46,48], three pre-post observational studies [28,29,33], one cross-sectional study [44], two case studies [38,43] and one comparative study [47]. These studies were conducted across eight countries, with the majority (66%) done in Europe. The breakdown of countries was: France (n = 4) [29,30,34,48], Spain (n = 3) [26,31,49], Italy (n = 4) [27,32,46,47], Poland (n = 1) [35], Slovenia (n = 3) [28,36,37], Brazil (n = 2) [33,44], Canada (n = 2) [39,42], The United States (n = 2) [41,43], Iran (n = 1) [40], Germany (n = 1) [45], and Israel (n = 1) [38].

Participant characteristics

Across all studies (see Table 3), the sample size ranged from 2 to 76, with an average of 20 participants. All studies reported participants’ ages, with a mean age range between 46 and 80 years of age. Twenty studies reported on participants’ disease severity using the Hoehn and Yahr (H&Y) scale [50], where classifications ranged between I and IV, and the majority (73%) were classified in stages II-III. Only one study used the Unified Parkinson’s Disease Rating Scale (UPDRS) [51] to report on the disease severity, while twelve studies used both H&Y and the UPDRS to characterize severity. The H&Y stage-range marks the transition from bilateral symptoms without balance impairment (Stage II) to the onset of postural instability with preserved physical independence (Stage III). Similarly, mean UPDRS III and Movement Disorder Society-UPDRS (MDS-UPDRS) III scores largely fell between approximately 27 and 34, except for one study that had a score as high as 51 [46], which suggested that participants generally exhibited mild to moderate motor symptom severity. Cognitive status was assessed in 16 studies using the Mini-Mental State Examination (MMSE) [52] or Montreal Cognitive Assessment (MoCA) [53], with mean MoCA scores ranging from 25.8 (standard deviation [SD]=2.8) to 26.77 (SD = 1.48), indicating intact or mildly impaired cognition. Twelve studies excluded participants with severe cognitive impairments (i.e., MMSE < 24).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Intervention and patient population characteristics.

https://doi.org/10.1371/journal.pone.0326705.t003

Rehabilitation process and outcomes

Eighteen articles [4 level 2b, 6 level 3b, 5 level 4, and 3 level 1b] focused on whole-body rehabilitation that predominantly involved balance-focused exercises with controlled lower extremity and trunk/postural movements (weight shifting or bearing, stepping, leaning etc.) prompted by the VR technology [26,27,30,32–35,38,39,41–49]. All 18 of these studies aimed to treat gait and balance disorders, and to evaluate improvements in postural stability, independence, confidence, as well as perceived balance in performing activities of daily living in PD patients. Except for one study [34], all the studies that targeted whole-body rehabilitation reported intervention durations ranging from several weeks to a year [26,27,30,32,33,35,38,39,41–49]. The planned frequency of the sessions ranged from every day to four times per week and they were scheduled as 30- to 75-minute long sessions [26,27,30,32–35,38,39,41–48].

The most commonly employed outcome measures used in these studies to assess functional mobility, postural stability and perceived balance were the Berg Balance Scale (BBS) [54], MDS-UPDRS [51], the Tinetti Performance-Oriented Mobility Assessment (POMA) [55], the Activities-Specific Balance Confidence (ABC) Scale [56], and the United Theory of Acceptance and Use of Technology (UTAUT) questionnaire [57]. All 18 studies demonstrated positive outcomes in terms of improved balance, balance confidence, gait, motor function, mood, and acceptance rates among persons with PD [26,27,30,32–35,38,39,41–49].

In addition to physical outcomes, the study by Del Pino and colleagues [49] undertook a cost-utility analysis comparing their GBT solution (n = 10) to a control group of patients (n = 10) who underwent standard clinical care. They found that the cost of GBT was €2243.07 while standard care cost €5108.26 over a 3 month period. The authors attributed the cost savings to not requiring the physical presence of the professional for the GBT participants. Similarly, Gandolfi et al. [46] examined costs between their GBT participants who underwent a TeleWii intervention (n = 38) versus those who participated in an in-clinic sensory integration balance training (SIBT; n = 38) program, which found the total treatment cost was €246.75 per patient for the GBT group and €493.50 per patient for the SIBT group. When taking into account indirect costs as well as equipment costs, the total cost for GBT was €383.55 per patient for the TeleWii group and €602.1 per patient for the SIBT group. One case study [38] using a customizable game estimated their system would cost approximately $2,000 (currency not specified). Other included studies highlighted the cost-effectiveness of GBT compared to traditional therapy (e.g., Negrini et al. [47]) but did not provide data to support these claims or did not discuss costs at all.

Five studies targeted upper-limb rehabilitation [1 level 2b, 1 level 3b, and 3 level 4], and all of these specifically focused on exergames that involved rotating and picking up objects with the hands and arms [28,29,36,37,40]. Only one study involved fine motor movements of the hand (finger tapping) [29]. All five studies were intervention-based, with the primary objective of designing targeted tele-rehabilitation programs for the upper extremities. The effectiveness, adherence, usability, and acceptance of these designed exergame tele-rehabilitation services were evaluated in each study [28,29,36,37,40].

The duration of the interventions varied from 2 weeks to 8 weeks, with sessions occurring three times a week, and each ranging from 30 to 75 minutes per session [28,29,36,37,40]. The most frequently utilized outcome measurement tools across these five studies were the Box and Block Test (BBT) [58], Nine-Hole Peg Hole Test (9HPT) [59], the Jebsen Hand Function Test (JHFT) [60], and the MDS-UPDRS [51]. Three of the five studies reported significant improvements in gross and fine manual dexterity and grip strength, as well as the range of motion and hand movements [29,37,40]. However, two studies that utilized the 9HPT to assess gross and fine manual dexterity did not demonstrate statistically significant improvements [28,36].

Two level 4 studies [31,38] focused on lower-limb rehabilitation. Lower-limb impairments in PD are often associated with mid to late stages of the disease. One study [38] employed lateral leg movements as the primary exercise, which was delivered four times per week over five weeks, and totaled 10 hours of treatment. The other study [31] incorporated obstacle negotiation tasks on an obstacle-lined pathway, with 45-minute sessions conducted weekly over a 12-month period, using game-based elements to enhance engagement.

Across all the included studies, there were overall positive outcomes associated with GBT interventions for individuals with PD, and the interventions appeared to be feasible in the majority of cases (see Table 4) [33–40,42–56]. Home-based implementation was often deemed feasible and well-received [28–32,37,42,47]. Similarly, the feasibility of GBT is represented by findings on system usability, where patients and physical therapists reported that they were easy to use and well-accepted. [31,34,39]. Some noted threats to feasibility however were related to game difficulty and repetitiveness, which suggested a challenge to sustained engagement [29,34], as well as the need for individualized software enhancements [45].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Clinical impact and implementation outcomes.

https://doi.org/10.1371/journal.pone.0326705.t004

While the findings from the studies suggest that the interventions were well-accepted and feasible, the consistency and magnitude of the therapeutic effects of GBT varied across studies. For instance, some interventions demonstrated either only a trend towards improvement [26], short-lived effects [49], or non-significant differences compared to control groups for primary outcomes [45,52,55]. Additionally, four studies provided insights into the effectiveness of GBT beyond the active intervention period. One study [41] showed some longer-term effectiveness, whereby improvements in motor function and quality of life were maintained at a one-month follow-up. Overall, positive impacts were observed but the therapeutic gains over time varied in those studies where documented.

Safety and feasibility

Of the 24 studies, 8 studies of varied methodological quality indicated that the GBT interventions were safe, with no adverse events reported [30,31,38,42,44–46,49]. Only one study [48] reported on nine events in eight of their participants (four in the active group and four in the control group). However, these events were deemed unrelated to the intervention, including three serious events (e.g., falls, medication-related hospitalizations) and five non-serious events. The remaining included studies did not report on safety.

There are several factors reported across the included studies that contribute to the feasibility of GBT to support physical therapy in home settings, such as end-user experience, adherence, perceived use, and cost-effectiveness. For instance, participants across studies rated the systems to be highly enjoyable [26], and exhibited high levels of motivation and adherence [27,28,37,38,41–43,48], which included high rates of game completion [28]. Relatedly, home-based implementation was often deemed feasible and well-received [28–32,37,42,47], and that systems were easy-to-use and well-accepted by both patients and physical therapists [31,34,39]. Importantly, participants expressed satisfaction with the interventions [30,31,39]. As noted above, studies also provided evidence on their cost-effectiveness compared to in-person approaches [32,46,47,49].

Among the seven studies that reported on the adherence levels of individuals with PD to GBT programs, four studies [29,34,37] utilized customized gamification modules without specifying a product name except for one [49], while one study [42] used the Nintendo Wii. Two studies [38,48] used the Microsoft Xbox 360 Kinect gaming console. Additionally, among the seven studies, two reported low adherence levels and decreased intentions to use the tele-rehabilitation program. The lack of adherence was attributed to factors such as the difficulty and repetitiveness of the games [34], as well as the lack of evolution of the used tele-rehabilitation tool (comprising bracelets equipped with motion sensors and a tablet) and the absence of after-sales service [29]. The other five studies reported that their participants maintained high levels of motivation and adhered to the exergaming programs consistently throughout the duration of the studies [37,38,42,48,49].

Hardware and software modules

A comprehensive list of the software modules and hardware devices utilized in the included studies are listed in Tables 5 and 6, respectively. Eleven of the 24 included studies [1 level 1b, 2 level 2b, 4 level 3b, and 4 level 4] used commercially available games [26,29,32,33,36,39,41–43,46,47], with most being those produced by the Nintendo Co., Ltd. (www.nintendo.com), which was the Nintendo Wii game console accompanied by the Wii Balance Board (7 out of 11) [26,32,39,41,42,46,47]. This was followed by systems developed by Microsoft’s Xbox 360 game console (www.xbox.com; 3 out of 11) paired with the Kinect camera [33,36,43].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Software features.

https://doi.org/10.1371/journal.pone.0326705.t005

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 6. Hardware features.

https://doi.org/10.1371/journal.pone.0326705.t006

The Nintendo Wii is a gaming console renowned for its innovative motion control system, which is showcased by the Wii Remote, Wii Nunchuk, and Wii Balance Board. The Wii Balance Board is an accessory that is designed to be used with the Wii console, enabling players to participate in games using their balance and physical movements. The Wii Balance Board enhances interactivity and immersion by accurately detecting and responding to players’ movements. The Xbox 360, developed by Microsoft, is a gaming console known for its extensive gaming library and immersive experiences. One notable addition to the Xbox 360 is the Kinect camera, which is a motion-sensing device. The Kinect camera, with its advanced depth sensing, enables controller-free interaction, by tracking players’ body movements, and translates them into game actions.

Among the 13 studies [2 level 1b, 2 level 2b, 3 level 3b, and 6 level 4] with custom-made games, 9 studies [27,28,30,31,35,36,38,45,48] developed their games utilizing the Kinect Camera sensory module. The Kinect camera offers depth sensing, skeletal tracking, and motion capture capabilities for game development. Microsoft provides software development kits (SDKs) and tools that grant access to the Kinect’s sensor data. Developers can incorporate gesture recognition, body tracking, and interactive features into their custom games, enabling the creation of unique and personalized gaming experiences. Additionally, one study developed its own bracelet-type sensory module [34] using an inertial motion sensor, one study [29] utilized the sensory inputs of a touch screen, one [49] used a tablet-based virtual coaching system called vCare that delivered customized games without the use of motion sensors or depth cameras, and one [37] relied on the sensory data of a Leap motion controller. The Leap Motion controller, introduced by Leap Motion Inc. (www.ultraleap.com/), is a compact device known for its precise hand and finger tracking. It enables intuitive gesture-based interaction with computers and digital devices, offering fine-grained control and immersive experiences. The commonly included hardware platforms used to run the software applications to process the data and to display it on a monitor were computers, tablets, and Nintendo Wii consoles.

Game-based solutions adapted in the rehabilitation interventions in all 24 studies were non-immersive VR-based serious exergames, except in one level 4 study [30], and one level 2b study [49] that used a combination of both non-immersive VR and virtual exercise coaching with the gamification techniques. Eleven of the designed gaming interventions included offline or real-time interactive communication and connection, as well as streaming software applications (i.e., Skype, Zoom), which allowed for the therapist/ medical professional to provide ongoing supervision and/or real-time visual feedback to patients on their performance, full tracking of patients’ performance while completing the session, and monitoring if the correct implementation of exercises was being performed [26,27,32,34,35,38,40,43,46–48]. In contrast, the other 12 studies [1 level 1b, 1 level 2b, 4 level 3b, and 6 level 4] consisted of tele-rehabilitation systems designed as a client-server model, where the users/patients self-run the session at their convenience, and the system records and stores the data; enabling the therapist/ medical professional to review data remotely and plan/ follow exergames-based therapy [28–31,33,36,37,39,41,42,44,45].

Applied gamification elements

Several game designs used combinations of different elements and techniques. For instance, some exergame designs introduced persons with PD as avatar characters within the game and assigned them target-based tasks (i.e., fruit picking, finger tapping task, and moving small virtual objects) with the simultaneous addition of auditory and/or visual feedback. For instance, the level 4 study by Nuic and colleagues [30] designed a video game consisting of a scenario of the patient moving to induce displacement of a small animal (avatar) in real-time environments to avoid obstacles and collect points related to visual and auditory feedback. Ten out of 24 studies [1 level 2b, 3 level 3b, 6 level 4] focused on the implementation of a single game [26,28–31,34,36–38,44], while the other 14 studies [3 level 1b, 4 level 2b, 4 level 3b, 3 level 4] provided multiple different games aiming to increase motivation and acceptance of users by inhibiting boredom [27,32,33,35,39–43,45–49].

Game dynamics

One study [49] did not provide a clear description of the implemented game dynamics by participants when undergoing GBT other than to state the activities were personalized. For the remaining 23 studies, all implemented tasks were based on six main categories of game dynamics (activities) for therapeutic games: (1) free movements (5/23 or 22%) [38,42–44,46]; (2) reach and hit the target (8/23 or 35%) [30–33,35,40,41,43]; (3) catch and pick up the target (8/23 or 35%) [33–39,48]; (4) carry and move the target (12/23 or 52%) [26,28,33,35,36,39,41–44,46,48]; (5) follow and track the moving target (3/12 or 25%) [35,40,44]; and (6) shift body weight right or left (5/23 or 22%) [27,35,39,42,46]. All six categories of game dynamics were used to design the therapeutic exergames that focused on whole-body rehabilitation. Activities such as reaching and hitting the target, catching and picking up the target, carrying and moving the target, and following and tracking the target were used in the design of upper-limb rehabilitation-based exergames. In one study [31] that specifically targeted lower-limb rehabilitation, the therapeutic exergames were designed around a single game activity involving reaching and hitting the target.

Limitations

Several limitations should be acknowledged in this scoping review. Firstly, the inclusion criteria were limited to English-written articles, potentially leading to the omission of relevant studies in other languages. Secondly, our search strategy may not have captured all relevant literature, potentially affecting our interpretations of the results. It is important to note that the primary purpose of this scoping review was not to assess the quality of evidence per se, but rather to gain insights into the current approaches used in the PD population and extract key implementation considerations for home-based settings. As such, the findings of this review can serve as a valuable roadmap for the design of future research studies in PD and offer broader implementation considerations to support the adoption and integration of GBT to enhance physical therapy outcomes.