Evaluating the impact of virtual reality on reducing restrictive practices in psychiatric wards

Abstract

This manuscript critically evaluates the randomized controlled trial (RCT) conducted by Phiri et al, which assessed the effectiveness of virtual reality (VR) training for psychiatric staff in reducing restrictive practices (RPs). Specifically, this RCT investigated the impact of VR on the handling of aggressive patients by psychiatric staff compared to traditional training methods. Despite significant reductions in perceived discrimination in the VR group, there were no major improvements in self-efficacy or anxiety levels. The system usability scale rated the VR platform highly, but it did not consistently outperform traditional training methods. Indeed, the study shows the potential for VR to reduce RPs, although fluctuations in RP rates suggest that external factors, such as staff turnover, influenced the outcomes. This manuscript evaluates the study’s methodology, results, and broader implications for mental health training. Additionally, it highlights the need for more comprehensive research to establish VR as a standard tool for psychiatric staff education, focusing on patient care outcomes and real-world applicability. Finally, this study explores future research directions, technological improvements, and the potential impact of policies that could enhance the integration of VR in clinical training.

Key Words: Virtual reality; Restrictive practices; Psychiatric wards; De-escalation techniques; Staff training; Generalized anxiety disorder; System usability scale; Burnout prevention

Core Tip: This manuscript highlights the potential of virtual reality (VR) as an innovative training tool for reducing restrictive practices (RPs) in psychiatric wards. Although VR training showed promise in lowering perceived discrimination and enhancing staff engagement, it had limited effects on self-efficacy and anxiety levels. The high usability of VR suggests its suitability for immersive learning, though challenges such as fluctuating RPs rates and cost may limit its widespread adoption. Further research is needed to refine VR simulations and explore long-term outcomes, positioning VR as a valuable tool in mental health care training.

TO THE EDITOR

The study by Phiri et al[1] evaluated the potential of virtual reality (VR) as an innovative tool to reduce the use of restrictive practices (RPs) in psychiatric settings. RPs are coercive interventions, including sedation, seclusion, and mechanical or physical restraints, commonly used to control psychiatric patients against their will. However, these practices pose ethical and safety concerns for both patients and staff[1,2]. VR training provides immersive simulations for healthcare professionals to manage aggressive behavior without resorting to RPs. This safe and cost-effective method allows staff to repeatedly practice appropriate management of uncooperative patients while receiving feedback from trainers on the correct approach and response. This aligns with growing calls to reduce coercive interventions in mental health care[1]. Previous research highlighted the benefits of VR in medical education, where it fosters better engagement and learning outcomes compared to traditional methods[3,4]. However, the effectiveness of VR in psychiatric training remains underexplored. The study by Phiri et al[1] addresses this gap by investigating VR usability and its impact on reducing RPs in psychiatric wards[2,3].

CRITICAL APPRAISAL OF THE STUDY

Study design and methods

Phiri et al[1] conducted an exploratory, cluster-randomized controlled trial (RCT) to assess the feasibility of VR training for reducing RPs in psychiatric wards. The study was performed according to the CONSORT guidelines, ensuring the reporting adhered to recognized standards for RCTs. Conducted within the Southern Health NHS Foundation Trust (SHFT), the trial randomized three inpatient wards to intervention and control groups each. This cluster randomization minimized contamination between the groups, ensuring that findings reflect real-world training settings.

Eligible participants included clinical staff aged ≥ 18 years who were fluent in English and required to complete RP training as part of their role. Exclusion criteria comprised prior experience with the Virti VR platform, a history of cybersickness, or conditions such as epilepsy. Participants in the VR group underwent immersive training scenarios that included managing violent patients and de-escalating conflicts, while the control group received standard mandatory RP reduction training. Follow-up assessments were conducted in six months to capture long-term effects.

Primary outcomes were measured using validated tools, such as the generalized anxiety disorder scale (GAD-7) for anxiety, the burnout assessment tool for emotional strain, and the compassionate engagement and action scale for compassion. The system usability scale (SUS) was employed to evaluate the usability of the Virti VR platform. Despite the robust methodological framework, certain CONSORT checklist elements, such as detailed participant flow, were not explicitly reported, limiting the ability to assess bias related to dropouts or exclusions.

Findings and interpretation

The study revealed notable differences in perceived discrimination between VR and control groups, particularly in items related to insults and harassment (P = 0.023 and P = 0.040, respectively). However, self-efficacy or anxiety levels did not show significant improvements. This aligns partially with the hypothesis, suggesting that VR training may enhance specific dimensions, such as empathy, while being less effective in addressing broader emotional outcomes, such as anxiety and confidence.

The high usability score of the Virti VR platform (average SUS score of 71.79) supports the potential for VR as an engaging training tool. However, fluctuations in RP rates in the VR group, influenced by external factors such as staff turnover, highlight challenges in consistently applying the intervention in clinical settings.

DISCUSSION OF THE FINDINGS OF THE STUDY

The study revealed significant differences in perceived discrimination between VR and control groups, particularly regarding items related to insults and harassment (P = 0.023 and P = 0.040, respectively). However, no major differences were found in self-efficacy or anxiety measured by the GAD-7 scale. Although baseline GAD-7 scores indicated minimal anxiety (< 5), the intervention group demonstrated slight reductions, potentially reflecting subtle emotional benefits. However, the absence of repeated assessments across multiple time points limits the ability to comprehensively evaluate anxiety trends; hence, we recommend this approach in future research. These results only partially align with the study's hypothesis, which anticipated improvements in these areas. Although VR could enhance empathy and reduce discrimination, it may not effectively impact broader emotional outcomes, such as anxiety and self-confidence, within the brief intervention period.

The study also reported that most participants rated the VR system favorably on the SUS, with an average score of 71.79, indicating high usability. Thus, this supports the hypothesis that VR could serve as an engaging and accessible tool for training. However, despite its usability, the VR intervention did not outperform traditional training methods in terms of reducing anxiety or increasing self-efficacy. Therefore, the content or duration of VR simulations may need refinement to better address these psychological outcomes.

Another key outcome was the reduction in RPs in the VR group compared to the control group. However, RP rates fluctuated in the VR group, possibly due to external factors such as staff turnover and ward dynamics. While the initial decrease in RP rates in the VR group is promising, the observed variability underscores the complexity of applying VR training consistently in clinical environments. Further studies are needed to examine the sustained impact of VR on RP reduction and determine the best method to integrate this technology into routine practice.

STRENGTHS OF THE STUDY

The innovative application of VR technology in clinical training is among the most notable strengths of the study. The use of VR represents a forward-thinking approach to addressing the longstanding issue of RPs in psychiatric wards. Furthermore, this approach allows staff to engage with realistic simulations of challenging situations, such as managing aggressive patients or conflicts over restricted areas, without the pressure or potential harm associated with real-life encounters.

The ability of the VR platform to simulate complex interactions provides a unique opportunity for repetitive practice and experiential learning, which are critical for skill acquisition and retention in high-stress environments such as psychiatric care[5]. This innovative use of VR also aligns with broader trends in medical education, in which simulation-based learning is increasingly recognized as an effective tool for enhancing clinical competence and improving patient outcomes[6].

The design is another key strength of this study, particularly the RCT framework, which is considered the gold standard in clinical research. The RCT design enhances the internal validity of the study by reducing biases that could affect the outcomes. Randomization, conducted at the ward level, ensured that entire wards were assigned either to VR intervention or control group, thereby minimizing the risk of contamination between the groups. This approach was particularly appropriate given the nature of the intervention since individual randomization could have led to cross-exposure of training methods among staff working in the same environment.

Furthermore, the methodology included follow-up assessments at one and six months, enabling the researchers to observe the sustainability of the effects of the intervention over time. Although the study was conducted at a single site, the rigor of the RCT design, combined with the detailed analysis of multiple outcomes, enhances the credibility and relevance of the findings.

LIMITATIONS OF THE STUDY

A primary limitation of the study is its small sample size and single-site setting. With only 56 participants (34 in the VR group and 22 in the control group), the findings may not fully capture the diversity of experiences and outcomes that could be observed in a larger, more varied population. The small sample also limits the study’s statistical power, reducing the ability to detect subtle yet meaningful differences between groups.

The single-site setting, conducted at the SHFT, further limits the generalizability of the results. These findings may not apply to other psychiatric settings with different patient demographics, staff structures, or organizational cultures. Without multi-site validation, it remains uncertain whether the VR intervention would be equally effective in other contexts.

Additionally, the four-week intervention period raises questions about the long-term impact of VR training on reducing RPs and improving staff psychological outcomes. While the study included follow-up assessments at one and six months, the short intervention period may not have been sufficient to produce lasting behavioral changes, potentially explaining the lack of significant findings in areas such as anxiety and burnout.

Another key limitation of the study is the interpretation of GAD-7 scale results. Baseline scores indicated minimal anxiety levels (mean score < 5), which limits the capacity to demonstrate significant reductions in anxiety. While minor improvements in GAD-7 scores were observed, they remain within the "normal" range and may not fully reflect meaningful psychological changes. Additionally, the study design did not include repeated assessments across multiple time points (e.g., baseline, during, and after training). This omission restricts the ability to track dynamic changes in anxiety levels throughout the intervention period, reducing the depth of analysis.

While the reduction in RPs suggests a positive impact, the lack of improvement in anxiety and burnout, along with an unexpected increase in perceived discrimination, points to potential unintended consequences. This discrepancy between behavioral and psychological outcomes emphasizes the need for a deeper exploration of how VR training interacts with the complexities of psychiatric care. Future research should focus on refining the training content or extending its duration to better address the interconnected factors of behavior and well-being in such high-pressure environments.

COMPARISON WITH EXISTING LITERATURE

Using VR for healthcare training, as demonstrated by Phiri et al[1], aligns with a growing body of evidence supporting simulation-based learning. For instance, Kyaw et al[3] conducted a systematic review highlighting that VR improves knowledge retention and engagement among healthcare professionals, although with variable effects on clinical competencies. Similarly, Hasan[4] emphasized the potential of VR for enhancing experiential learning in high-stakes environments, such as psychiatric care, where real-world practice carries significant risks.

However, the findings by Phiri et al[1] diverge from earlier studies reporting significant reductions in staff anxiety following VR training. Studies, such as that by Bell et al[7], showed moderate success in reducing anxiety among healthcare staff, contrasting with the limited impact observed in Phiri et al’s study[1]. This discrepancy may be attributed to differences in intervention design, duration, or participant demographics.

Furthermore, the focus of the study on RPs aligns with research by Chavulak et al[2], which explored staff perceptions of training interventions for reducing coercive practices. While Phiri et al[1] demonstrated reduced perceived discrimination, the variability in RP rates suggests the need for further refinement of VR scenarios to address broader systemic challenges, as emphasized in the literature.

Overall, the findings of Phiri et al[1] contribute to the broader conversation on the role of VR in healthcare training but underscore the need for larger-scale, multi-site studies to validate its long-term effectiveness. Similarly, studies on reducing RPs in mental health, such as that by Bell et al[7], emphasize the need for innovative approaches to train de-escalation techniques for the psychiatric staff. Despite showing moderate success, these studies often lack long-term follow-up data[7]. According to Phiri et al[1], VR, while holding promise in psychiatric settings, did not significantly reduce anxiety or burnout, confirming the mixed outcomes observed in other trials[2].

A more comprehensive comparison with the existing literature reveals both alignment with and divergences from prior research, as detailed below.

Effectiveness of VR in mental health training

Studies consistently demonstrated that VR-based interventions improve learning outcomes and engagement among healthcare professionals. For instance, the systematic review of VR applications in psychiatric disorders by Cieślik et al[8] showed that immersive simulations improve cognitive and behavioral training outcomes, although with variations in clinical effectiveness. Similarly, Kim et al[9] emphasized the efficacy of VR in training mental health professionals to address conditions such as post-traumatic stress disorder, highlighting its role in improving staff preparedness and therapeutic delivery.

Pottle[10] identified VR as a transformative tool in medical education, demonstrating its ability to foster empathy and reduce stress in healthcare environments. Collectively, these studies support the potential of VR to deliver targeted training in high-pressure environments such as psychiatric wards.

Reducing RPs through training interventions

The reduction in coercive practices, such as RPs, requires staff training interventions tailored to real-world scenarios. Chavulak et al[2] emphasized the importance of scenario-based training for reducing coercive behaviors, particularly when interventions are supported by immersive technologies such as VR. However, these studies also underscore the need for longitudinal data to validate the effectiveness of VR in sustaining reductions in coercive practices.

Staff training and emotional outcomes

Although Phiri et al[1] found mixed results regarding anxiety improvement, the existing literature suggests that VR may hold promise for addressing emotional outcomes in healthcare professionals. Bell et al[7] reported significant reductions in burnout following VR-based interventions, although improvements in anxiety were inconsistent, mirroring the findings in Phiri et al’s study[1].

CLINICAL IMPLICATIONS AND FUTURE DIRECTIONS

Integrating VR into standard training protocols for psychiatric staff holds potential, especially for reducing RPs. VR can provide immersive simulations allowing staff to repeatedly practice de-escalation techniques in a controlled, risk-free environment, which could improve training efficacy[7]. However, barriers such as cost, accessibility, and the need for regular updates to VR software could limit its widespread adoption[3]. Enhancing interactivity and tailoring scenarios to real-world complex situations would further boost the utility of VR in mental health care settings.

Future studies should incorporate repeated assessments across multiple time points, including baseline, mid-intervention, post-intervention, and follow-up periods, to enhance the evaluation of anxiety and other emotional outcomes. This approach would provide a more comprehensive understanding of how interventions such as VR training impact anxiety trajectories over time. Furthermore, expanding the focus to include qualitative assessments could provide deeper insights into participants' experiences and emotional responses, particularly with low baseline anxiety levels.

Larger-scale studies with longer follow-ups are required to determine the sustained impact of VR on both staff behavior and RP rates in psychiatric wards. Existing studies, such as those by Chavulak et al[2] and Phiri et al[1], show that despite promising initial findings, variability in outcomes suggests the need for more detailed research to comprehensively understand the long-term effectiveness of VR. Additionally, further exploration into how VR influences different patient demographics and staff roles would be valuable.

Moreover, enhancing the interactivity of VR simulations and developing more complex, real-world scenarios could further refine the effectiveness of VR interventions in healthcare settings[3,7]. Investigating the scalability of VR across institutions is critical for broader implementation[2].

CONCLUSION

The study by Phiri et al[1] demonstrated the feasibility of using VR to train psychiatric staff in reducing RPs. Despite showing significant potential in lowering perceived discrimination and improving staff engagement, VR did not significantly impact self-efficacy or anxiety. The strengths of this study lie in its real-world setting and innovative use of VR. However, challenges such as fluctuating RP rates and usability limitations were identified.

VR technology presents a promising tool for training in mental health care by offering immersive, repeatable simulations that foster safe patient-staff interactions. However, its effectiveness in consistently reducing RPs requires further refinement, particularly regarding cost, accessibility, and the development of more complex, real-life scenarios. Future research with larger sample sizes and longer follow-ups is essential to validate these findings and expand the potential of VR in psychiatric training.