2. Materials and Methods 2.1. DesignThis systematic review was performed according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) 2020 recommendations for systematic reviews and meta-analyses [28]. 2.2. Search StrategyA search of the scientific literature was carried out in the following databases: PubMed, Web of Science, Scopus, CINAHL, and PsycInfo. The “Grey literature” was not consulted. The search was carried out up to October 2022. The following keywords were used for the searches (Table 1): (“physical therapy” OR “physiotherapy”) AND (“teaching” OR “learning”) AND (“virtual reality” OR “augmented reality” OR “mixed reality” OR “virtual reality exposure therapy” OR “virtual system”). No filters were applied in terms of date of publication or study design. 2.3. Eligibility Criteria

The inclusion criteria for the selection of the articles included in this review were as follows: the study population was made up of graduate or undergraduate students of physiotherapy; the teaching method was carried out using VR/AR-based devices and was compared to traditional teaching methods; the results are related to the learning satisfaction/perception with the teaching model, and the academic performance regarding the acquisition of theoretical knowledge and/or physiotherapy-related skills. Concerning the study types, included studies were randomised and non-randomised controlled studies, written in the English or Spanish language, and published in peer-reviewed journals and conference proceedings.

Regarding the exclusion criteria, we excluded those studies that included a population of physiotherapy students but did not provide detailed and separate results from the other study populations. In addition, studies focused on the acquisition of professional skills were excluded, i.e., those that take professional trainees as the study population.

2.4. Selection Process and Data Extraction

First, the search was performed by combining the keywords previously described in the different databases. Potentially relevant articles were identified after reading the title and abstract, and duplicate articles were eliminated. Subsequently, a thorough check for compliance with the inclusion criteria was performed to obtain the articles included in the systematic review. Two reviewers (D.L.A. and J.A.M.M.) actively participated in the selection and review process.

Data collection involving the systematic data extracted by two independent reviewers (D.L.A. and J.A.M.M.) was manually conducted according to the Joana Briggs Institute (JBI) data extraction form [29]. The following information was extracted from each article included in the systematic review: author, year of publication, country, study design, characteristics of the participants, characteristics of the teaching model, number of lessons received, instruments used to assess the impact produced by the teaching model, and results obtained. The synthesis of the information was visually displayed as a table. In case of discrepancies, two additional reviewers (J.C.F.L. and A.I.P.S.) were consulted. 2.5. Methodological Quality and Risk of Bias AssessmentThe JBI Critical Appraisal Checklist for Quasi-Experimental Studies (ranged from 0 to 9), the JBI Checklist for Cohort Studies (ranged from 0 to 11), and the JBI Checklist for Randomised Controlled Trials (ranged from 0 to 13) [30] were used for conducting the quality appraisal. In addition, the Cochrane Collaboration’s risk of bias tool (RoB 2.0) [31] and the risk of bias in non-randomised studies of interventions (ROBINS-I) tool [32] were used to assess the risk of bias of randomised and non-randomised controlled studies, respectively.

Two reviewers independently (C.G.M. and D.L.A.) conducted the quality appraisal and risk of bias assessment of the retrieved papers. An agreement was reached through discussion, and two additional reviewers (J.A.M.M. and A.I.P.S.) were consulted in case of discrepancies.

4. Discussion

The present systematic review provides an overview of using VR/AR for teaching physiotherapy among graduate and undergraduate students. To the best of our knowledge, this is the first systematic review assessing this innovative approach in physiotherapy education. A total of seven studies were included in the review. It should be highlighted that a total of 737 students were involved in the analysed studies. After analysing the results of the different studies included in this review, we cannot conclude that using VR/AR as a teaching tool was more effective than traditional teaching methods in terms of learning satisfaction and academic performance, as contradictory results were obtained.

Due to the lack of systematic review or meta-analysis analysing the use of VR/AR for teaching physiotherapy, the present findings will be compared to other systematic reviews and meta-analyses analysing the topic in health-related education. Contrary to our findings, Zhao et al. [39], stated that the pass rate of medical students trained using VR technology was higher than those using traditional education, resulting in a higher acquisition of specialised knowledge. Another recent systematic review carried out by Barteit et al. [40] suggested that using VR/AR teaching models had beneficial effects on medical education, showing positive results on enthusiasm and enjoyment. In addition, Shorey and Debby Ng [23] analysed the use of virtual environments as a teaching tool in nursing education, stating that this teaching model was effective at improving the theoretical knowledge acquisition, even suggesting it as an alternative method to teaching in nursing education. Although our hypothesis was that using VR/AR as a teaching tool would be more useful than traditional methods for teaching specific special skills and theoretical knowledge during physiotherapy teaching, due to its advantages in terms of multisensory stimulation, interaction, playful environment, and feedback, our results are contradictory for both learning satisfaction and academic performance, so no solid conclusion can be drawn about the usefulness of one method over another. In spite of some studies that used immersive VR devices, most studies included in the review used simulations instead of games, which could have less impact on learning due to the lack of interactive participation [15,18]. In this way, according to Ulrich et al. [34], the non-positive results found in this review could be due to the lack of interactivity induction across the different VR/AR-based teaching models used by the studies. In addition, there were heterogeneous factors, such as the teaching content and duration, VR/AR devices used, and academic degree, which could influence the obtained results.Concerning the teaching content, positive results were found for teaching anatomy (neck, spine, and gross dissection anatomy) [25,36,37], and for acquiring skills related to movement analysis [35], so it could be considered as a useful complement to traditional methods for teaching anatomy content in physiotherapy education. When analysing the results related to the clinical decision-making skills and specific physiotherapy tasks, we found using VR/AR-based teaching methods to be just as effective as traditional methods. A possible explanation for the lack of positive results may be the short teaching period used in the studies, in which improvement trends but non-significant results were observed in the VR/AR group, which could become significant if measured over a greater time period [33]. Therefore, future studies analysing the use of VR/AR to teach these specific skills among physiotherapy students are needed.Interestingly enough, the positive results obtained for teaching anatomy match with those of previous studies [39,41] analysing the use of VR for teaching anatomy in health-related education. However, we found no positive results for clinical decision-making and specific clinical skills, in contrast to a previous systematic review [42] analysing the use of VR for health profession education. In view of these contradictory results, we can suggest that teaching anatomy may be similar in the different health-related education disciplines, but the specific clinical skills to be developed in each discipline may be different. Finally, considering that there was a limited number of studies analysing each teaching content, the results should be taken with caution.The overview provided by the present study about the use of VR/AR for teaching physiotherapy could be considered to assess the inclusion of VR/AR in other health-related areas of knowledge. In this way, and according to Zhao et al. [39], we suggest using VR/AR systems as a complement to traditional models instead of replacing traditional models to improve the education of physiotherapy students.Finally, according to Rickel [43], the inclusion in the university education system of teaching models based on the use of VR would imply an important economic effort for the supply of the infrastructure needed in the different teaching centres, as well as a great effort to offer adequate training and instruction in the use of these new technologies to university teaching staff. In addition, virtual learning environments should be continuously adapted to the needs of teaching staff and students. Study Limitations

First, the limited number of studies included, their low methodological quality and high risk of bias, should be highlighted. In addition, several studies used a single teaching lesson, so no strong conclusions can be drawn. Thus, studies implementing VR/AR-based teaching models for improving teaching and learning experience in the medium and long term are needed. In this line, the development of a greater number of experimental studies with higher methodological quality is also needed, as well as studies that encourage the creation of virtual learning environments using user-centred designs, in order to provide a solid conclusion on the use of VR/AR for teaching physiotherapy.

5. Conclusions

This systematic review analysed the results of seven studies that examined the use of VR/AR for teaching physiotherapy, with a total of 737 students. Despite the potential benefits of using VR/AR for teaching purposes, such as using a motivating context, the sustained focus of attention on the task performed, as well as the opportunity to provide feedback, which are key factors to induce a higher achievement of the objectives in the learning process, our results are contradictory in terms of learning satisfaction and academic performance. We can conclude that VR/AR-based teaching models seem to be equally effective as traditional methods for teaching physiotherapy. In this way, the scientific evidence of using VR/AR-based teaching methods for teaching physiotherapy is still in its first stages, so we cannot strongly recommend its inclusion in physiotherapy education. In any case, we could recommend it as a complement rather than a replacement for traditional teaching.

We encourage teachers and researchers to conduct future research analysing the use of VR based not only on simulation or virtual environments, but also on games, which could enhance interaction and active learning. In addition, future studies including a larger number of sessions and higher methodological quality are needed to provide solid conclusions on the use of VR/AR for teaching undergraduate and graduate curricular content in physiotherapy.