Practice placements are important activities in the training of health care students. They promote the application of knowledge to a practical setting for developing the skills, attitudes, and behaviors expected of a health care professional [1-3]. Placements allow active involvement in care delivery under supervision, and the opportunity to receive feedback on student performance [4]. In other words, student learning on placement is contextualized to future practice.

Simulation-based placements present an alternative to traditional practice placements. In traditional placements, students enter a workplace and learn through observation and participation in actual clinical events. In contrast, health care simulation is a technique that produces a scenario designed to represent a real-life practice situation for experiential learning [5,6]. Compared with traditional placements, simulation can ensure that low-frequency and high-risk cases or situations receive sufficient practice in a safer space, without mistakes causing harm to real persons [7]. Thus, the advantage of simulation is the ability to control and direct case-based learning.

With advances in technology, simulation-based education is expanding into web-based environments, a trend accelerated during the pandemic. The increasing complexity of health care also requires an agile workforce of lifelong learners, capable of substituting skills across professions [8,9]. Consequently, health care training must keep pace with technology developments, and virtual simulations could support the training of these skills [10-12]. Furthermore, virtual simulations offer greater flexibility and scalability compared with using standardized patients (people play acting the role of a service user) [13].

As virtual simulated placements (VSPs) are an emerging field, mapping the literature across health care and analyzing gaps is recommended before more specific research questions are defined [14,15]. Therefore, our research team chose a scoping review method to conduct a broader search across medicine, nursing, midwifery, and allied health, for undergraduate and postgraduate students who undertook VSPs. Considering the importance of practice placement, the advantages of simulation-based learning, and recent advances in technology, this topic was relevant for the review. We define virtual simulations as computer-based activities according to the Healthcare Simulation Dictionary [16], and our aim was to determine the scope of the VSP literature to inform future research questions. Our objective was to assess the range and types of evidence related to VSPs, across the health care professions.

First, what is the scope of the literature relating to VSPs for health care students? Second, what outcomes are reported in relation to the students undertaking VSPs? Third, what are the patterns and gaps in the literature and the reported outcomes? Finally, what are the implications of the review findings for future directions in VSP research?

This study followed the stages detailed in a framework for scoping reviews [14]: (1) identify the research question; (2) identify relevant studies; (3) study selection; (4) charting the data; and (5) collating, summarizing, and reporting the results.

A preliminary search of MEDLINE, the Cochrane Database of Systematic Reviews, and Joanna Briggs Institute Evidence Synthesis was conducted on June 17, 2022 to locate any existing or underway reviews on the topic. One systematic review [13] was identified and focused on digital placements for undergraduate nursing and medical students. The review also included experiences such as telemedicine and on-screen role-play. While their search located 16 studies in April 2021, the increased trend toward implementing VSPs within undergraduate and postgraduate programs across the wider health professions justified this review.

An a priori protocol used the Joanna Briggs Institute template for scoping reviews [15] and was registered with the Open Science Framework (DOI 10.17605/OSF.IO/AY5GH) [17]. The PRISMA-ScR (Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews) checklist (Checklist 1) ensured methodological rigor when reporting this review [18].

The eligibility criteria are listed in Table 1 using the SPIDER (sample, phenomenon of interest, design, evaluation, research type) [19] and PCC (population, concept, and context) [15] formats:

The selection criteria were piloted by screening 50 randomly selected titles and abstracts, independently by 2 reviewers (JS and MG). This process generated 94% agreement (Cohen κ=0.6) and served to clarify the selection criteria. In discussion with a third reviewer (NT), the Health and Care Professions Council definition for allied health [20] was adopted in place of the National Health Service (NHS) criteria [21], since this definition includes practitioner psychologists—a population potentially well-suited to VSPs, with the emphasis on talking therapies.

An initial limited search of MEDLINE and CINAHL was undertaken on June 28, 2022 to identify articles on the topic. The text words contained in the titles and abstracts of relevant articles and index terms were used to develop a full search strategy. This was checked by a health care research librarian and run on MEDLINE on August 3, 2022 (Multimedia Appendix 1). The search strategy was then adapted for each database. The databases searched included MEDLINE, CINAHL, Allied and Complementary Medicine Database, Cochrane Database, PsychINFO, Education Resources Information Center, SCOPUS, ScienceDirect, and Biomed Central. Gray literature sources include PubMed, Electronic Theses Online Service, ProQuest (dissertations), Google Scholar, and Institute of Electrical and Electronics Engineers Xplore. Searches were limited to English language and dated from January 1, 2020. The date limitation was justified given that VSP research has essentially emerged postpandemic.

Supplementary search strategies were employed using existing knowledge and networks, contacting relevant organizations, hand-searching journals, and checking the reference list of all included sources and relevant reviews. Advances in Simulation, British Medical Journal: Simulation and Technology Enhanced Learning (BMJ STEL) and Clinical Simulation in Nursing were hand-searched. These supplementary searches were conducted by one reviewer (JS) and checked by another (NT).

An updated database search was conducted on July 9, 2023. A second reviewer (MG) checked the title, abstract, and full-text selection decisions. Registries (Clinical Trials.gov, World Health Organization International Clinical Trials Registry Platform, and the Cochrane Database) were searched for additional papers [22]. Updated hand searches were performed in Advances in Simulation and Clinical Simulation in Nursing (BMJ STEL had since discontinued). A second reviewer (NT) checked these supplementary searches.

Following the database searches, all identified citations were uploaded into EndNote (Clarivate) [23], and duplicates were removed. Each potential duplicate was confirmed separately, rather than using batch automation to prevent the removal of false positives [24]. Citations were exported to Rayyan and rechecked for any missed duplicates [25].

Once pilot screening was complete, the remaining titles and abstracts were screened independently by 2 reviewers (JS and MG) against the revised criteria, and potentially relevant sources were retrieved in full text. These were assessed in detail against the inclusion criteria by 2 independent reviewers (JS and MG), blinded in Rayyan. There was 83% agreement (Cohen κ=0.5) between reviewers. A 100% agreement was reached through discussion. Further details of the source selection, including a list of references excluded at full text screening are detailed in Multimedia Appendix 2.

A Microsoft Excel spreadsheet was used as a data charting tool to standardize obtaining information from the papers. Furthermore, 2 independent reviewers (JS and MG) conducted a pilot of 5 included papers to assess the utility of the information charted and generate emerging themes. Consensus was reached between reviewers (JS and MG) on the charting method, and modifications were made to the spreadsheet, to improve the quality of charted data (Multimedia Appendix 3). Following this, one reviewer (JS) charted the remaining data, which was checked by another (NT).

A table of included study characteristics was collated, and numerical analysis in Microsoft Excel was undertaken to provide descriptive statistics. The size of the dataset was manageable enough to organize findings across the PAGER (patterns, advances, gaps, evidence for practice, and research recommendations) domains [26], for synthesis, without the use of NVivo software (Lumivero; as was planned in the protocol).

The Coventry University Ethical Approval process has been completed and the project has been confirmed and approved as low risk (project reference P139783). Date of approval is August 12, 2022.

The search results and selection process are reported in the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram (Figure 1).

The characteristics of the 28 included papers are summarized in Multimedia Appendix 4. Overall, VSPs were a combination of videoconferencing sessions with educators and peers, as well a variety of web-based material, including videos, reading, modules, and assignments. Most VSPs included some form of case-based learning that required problem-based activities to complete. Teaching methods ranged from didactic lecture-style sessions to peer learning and flipped classrooms. Session delivery featured more formal case conference-style sessions, as well as small group learning, and the use of online chat, polls, and quizzes.

The PAGER themes across the papers are summarized in Table 2. Key patterns and gaps are mapped across all included studies in Tables 3 and 4. The global distribution of publications is illustrated in Figure 2.

Patterns are mapped across all included studies in Table 3 and gaps are mapped in Table 4.

In total, 86% (24/28) of the included papers were published in high-income countries, as defined by the Organisation for Economic Co-operation and Development [56]. The VSP research was located primarily in the United States and the Northern Hemisphere.

The literature was predominantly medical and nursing research, constituting 93% (26/28) of the included papers. The distribution by profession and breakdowns by specialty are illustrated in Multimedia Appendix 5. Diagnostic radiology rotations were the most prevalent VSPs in medicine and pediatrics in nursing. Where stated, learners were often in their latter stages of training, or undertaking these VSPs as postgraduates.

All the VSPs in the included papers were developed in response to COVID-19 restrictions, which aligns with the time span of the scoping search. The context at the time was that the pandemic necessitated that face-to-face (FTF) practice placements were often discontinued. VSPs were implemented to provide alternative placement hours, enabling students to progress toward professional registration and graduation.

The most basic study design was a single group, with a postintervention measure, featuring in 16 papers. In total, 7 papers compared measures pre- and postintervention [34,37,39,41,48,50,54]. Furthermore, 5 papers compared VSP outcomes with a previous cohort of students who completed FTF placements prepandemic [28,33,45,51,55].

Practice partners (clinicians working in practice) were involved in the VSP course development with faculty in 8 studies [29,32,37,41,44,48,50,55] and students were involved in 4. Furthermore, 3 studies developed a needs assessment from student surveys [34,43,52]. None involved service users.

Conceptual underpinnings include pedagogy, theoretical frameworks, and professional standards. Although no single paper covered all elements, underpinning concepts are evident across the literature, summarized in Multimedia Appendix 6. Pedagogies employed, focused on adult student learners, case-based activities, and experiential and web-based learning. The frameworks structured the VSP development, and the professional standards guided curriculum, simulation, and placement.

All studies used generic software such as Zoom (Zoom Communications) or Microsoft Teams for screen-based communication, and many used existing learning management systems to host files and activities. Others adopted commercial software applications, allowing students to conduct a history by selecting from a menu of interview questions. None used conversational artificial intelligence (AI) systems (computer-generated conversation, assisted by AI). Some applications presented virtual reality (VR) patient avatars with which the student could direct a physical examination, although this was delivered via a screen [31,38,40,41,51] and 1 study provided an interactive community setting in screen-based VR [53]. All software resources are outlined in Multimedia Appendix 7.

The focus of most VSPs was clinical cases, through which knowledge, reasoning, decision-making, and communication skills (both verbal and written) were developed. Practical skills training was rarely practiced, with 1 study including home practice surgical kits as the exception [1]. Instead, skill learning was visualized through virtual patient encounters and instructional or walk-through procedure videos. The social determinants of health were the focus in 2 studies [50,53] and another facilitated students in teaching roles [42].

The most common outcome measures were custom-developed student evaluation questionnaires, followed by exam marks. Custom questionnaires provided positive feedback for student experience, satisfaction, and usability, although some technical issues and Zoom fatigue were cited [31,50]. In total, 3 papers reported a 100% pass rate on their VSPs [35,49,52], and 4 used a standardized exam to demonstrate comparable outcomes with FTF cohorts [45,51], or the national average [30,32].

Table 5 summarizes the outcomes of research that employed a repeated measures design or group comparisons.

When measures were compared pre- and post-VSP, there was a trend of improvement in self-rated competencies, knowledge scores, and critical thinking skills. However, when the comparison is made with traditional FTF placements, the pattern is less clear. There were no differences in grades when post-VSP exam scores were compared with previous cohorts’ who attended an FTF placement prepandemic. Student satisfaction was comparable in a study conducted in medical general practice, but professional exchange and learning scored higher in the VSP, while the attainment of new skills and attitudes scored higher in the FTF placement [33]. Furthermore, 1 paper compared students who participated in web-based readouts (the radiology equivalent of patient rounds) with students who attended workplace readouts prepandemic [28]. The educational value was comparable in survey results, although students on the VSP rated slightly higher for perceived interaction. That FTF students that were mostly observing on their placement might explain this finding. Conversely, FTF students had greater confidence in using the workstations, considered the case because the VSP students were unable to operate Picture Archiving Communication System workstations remotely.

This study mapped the literature describing VSPs across health care. All 28 papers were pandemic responses, primarily from medicine and nursing in high income countries . Selecting studies that conducted a web-based simulation, rather than employing a hybrid or blended approach may explain why all papers in this review were pandemic responses, and why the student populations were in their latter stages of training or postgraduates. COVID-19 necessitated a rapid shift to provide VSPs as a replacement for lost clinical hours to allow students to progress toward graduation [57]. However, these VSPs were often produced in a short time frame, under emergency situations, and may explain why few papers featured robust experimental designs and conceptual frameworks.

Replacing FTF placement hours with simulation is a contentious issue. Accordingly, a Delphi study considered the benefits and limitations of this approach [58]. Expert consensus across multiple professions agreed that between 11%‐30% of hours replaced with simulation would be acceptable, and this aligns with the current allocation set by the Nursing and Midwifery Medical Council [59]. VSPs in the curriculum may offset some pressure on workplace settings as they attempt to fulfill the NHS long-term plan to recruit and train more health care learners [11]. However, this does not diminish the importance of building further workplace placement capacity [58]. VSPs can be considered an additional pedagogy that offers a different, yet complimentary experience to traditional FTF placements.

In general, VSPs had a teleconferencing and a web-based learning component. The teleconferencing was commonly conducted with educators and peers over Zoom or Microsoft Teams, and the web-based learning activities included, but were not limited to videos, reading, modules, and assignments. There were a few examples of immersive learning with VR patient avatars, and these were delivered via a screen [31,38,40,41,51,53].

Disciplines that rely on image-based diagnoses may be more easily adapted to screen-based delivery, and consistent with this, diagnostic radiology, and pathology VSPs together constituted over 30% (6/19) of the medical papers in this review. In the development of this scoping review, we anticipated that psychology might be suited to VSPs due to the nature of talking-based therapies over physical skills, although it is possible that psychological presentations were considered too complex to portray accurately in computer-based simulations. With future developments in conversational AI systems and the growing acceptance of this technology, this situation may change. Similarly, professions that rely heavily on hands-on assessment, such as physiotherapy, may feature more in extended reality spaces with haptics, as further research and development into these technologies emerge. In the meantime, VSPs that require complex conversations are likely to include telecast or telemedicine simulations. Likewise, VSPs that teach advanced handling skills might adopt a hybrid or blended approach, thus combining the strengths of both web-based and FTF approaches.

VSPs have the potential to break down silos between professions, by delivering interprofessional education (IPE) over a web-based platform. IPE is defined as 2 or more professions, “learning with from and about one another to improve collaborative practice and quality of care” [P4] [60]. The intended outcome is to improve mutual understanding, teamwork, and leadership among different professionals [61]. VSPs have advantages over FTF training in building asynchronous activities for flexibility in timetabling and hosting synchronous activities without geographical constraints [62]. Given the relevance of IPE to quality care and the fit with web-based technologies, IPE-VSPs may be an important area for future research.

Elements of thoughtful VSP design are evident across several papers. Frameworks, such as ADDIE (analysis, design, development, implementation, and evaluation), ensure that there is structure to the process and stakeholder needs are met [30]. Existing curricula [54,55], or processes such as Kern’s 6-step model for curricular development could be used [43,45,46,50,52]. If framed within existing standards [40,49], VSPs can align with specified learning outcomes. Principles in pedagogy, such as andragogy [29,30] and web-based learning [33,50], ensure that VSPs build features that engage students with experiential learning [30] and promote problem-solving [29,30,39] and active reflection [49]. The conceptual underpinnings documented across this body of literature could provide a blueprint for best practice in VSP design.

Stakeholder involvement is a key process to inform the design of a VSP. Service users could inform the content, which is especially important in computer-based simulations, yet no service user involvement was documented. Students are the end users of a VSP, yet they were involved in a minority of studies. When students were involved, surveys informed a needs assessment, or they were consulted early in the process. This is a tokenistic approach compared with cocreation, the preferred method of engaging with stakeholders. Cocreation involves a collective effort with all stakeholders to collaborate across the entire design, development, implementation, and testing phases [63]. One noteworthy research report provided an overview of VSP development within a nursing program, which included input from students, service users, and other universities throughout [64]. Their working group comprised of academics, clinicians, a service user, a carer involvement lead, and an education technology lead. Therefore, in addition to underpinning VSP design with the relevant conceptual frameworks (pedagogical principles, theoretical frameworks, and published standards), broad stakeholder cocreation is optimal.

The pattern of positive student evaluation, improvement from baseline measures post VSP, and equivalence in exam scores, compared with in-person cohorts, appears promising, although, it should be remembered that the objective of a scoping review is to map the literature for patterns and gaps, rather than in-depth appraisal of the quality of the papers.

The findings compare with a systematic review that examined digital clinical education more broadly [13]. Stand-alone digital education was reported to be as effective as conventional learning for knowledge and practice, in nursing and medicine. However, there are some methodological concerns with this systematic review [13]. There was no a priori protocol, and the study lacked a pilot to test the methods. A librarian’s involvement in verifying the search strategy was not reported, gray literature was not searched, and duplicate processes were absent for the study selection and data extraction stages.

There are several barriers to conducting a systematic review of VSPs across health care. First, there is insufficient research across midwifery and allied health [34,49]. Another consideration is that all student evaluations in this scoping review were custom-designed. Therefore, the inconsistency of outcome measures might prevent meaningful comparisons across papers. One study used previously researched scales for clinical thinking ability, academic self-efficacy, and student engagement, which demonstrated good reliability [41]. Some of the exams were standardized [30,32,45,51], but none compared the baseline marks of each group to determine whether there were differences at the outset. In all cases, VSP exam scores were compared with a previous cohort that attended placement FTF prepandemic, or the national average, rather than adopting a prospective design.

It is clear from the paucity of research outside nursing and medicine, the lack of prospective research designs and inconsistent, nonvalidated outcome measures, that research into VSPs is in its infancy. It is tempting to recommend greater consistency of outcome measures and more robust experimental designs to improve the evidence base. However, such approaches may not fit the study of complex educational interventions such as VSPs. More suitable approaches include quasi-experimental, qualitative, and evaluative designs to examine conceptual underpinnings, VSP cocreation, the mechanisms mediating learning responses, and individual case trends over time.

The strengths of this study relate to the methodology. A structured process for defining search terms was undertaken, and a librarian was consulted for the search strategy. A range of databases were searched across medical and technology specialties. Gray literature sources were searched, and an updated search included trial registries. An a priori protocol was registered, and a subset of data was piloted to determine the declared changes. Duplicate processes in study selection and data charting were employed, and existing guidelines were used to design the protocol, synthesize the findings, and report the paper.

One weakness is that many health care educators may have implemented VSPs without documenting their practices. As such, a scoping review of the literature will always underestimate the scale and depth of innovation in practice. Limiting the search to English language increased the risk of language bias. While the limited number of publications from low or middle income countries could reflect the language limitation, it is also likely that countries with greater resources were better positioned to make the rapid shift to web-based education and publish their research during a global health emergency. Web-based platforms are suited to sharing resources and overcoming geographical constraints to access expertise, and VSPs present an opportunity to address inequality in health care education moving forward.

This scoping review mapped the VSP evidence across health care, highlighting patterns and gaps in the evidence base. All papers documented pandemic responses, primarily in medicine and nursing in high income countries. There are notable gaps in the midwifery and allied health research. Although emerging trends for VSPs in this review demonstrate some positive outcomes, this review highlights the need for improvements in VSP design. These include cocreation with a wider range of stakeholders and underpinning by pedagogical principles, theoretical frameworks, and published standards. Research into student engagement using VR headsets, haptics, and conversational AI systems in VSPs, are areas for future research, as immersive technologies and their use cases develop. The pandemic has revealed an opportunity to augment placement capacity through VSPs. There is the potential for future VSPs to feature IPE, thus promoting joined-up care in health care graduates. There is also the opportunity for VSPs to improve local and global access to quality clinical education experiences.