Healthcare Professionals’ Perspectives on Immersive Virtual Reality for Rehabilitation and Patient Engagement in a Post-Acute Community Setting: A Qualitative Descriptive Study

Abstract

Background: Immersive virtual reality (VR) may support rehabilitation and patient engagement; however, little is known about how frontline healthcare professionals perceive its clinical relevance and feasibility in post-acute community care settings. This study explored healthcare professionals’ perceptions of the potential clinical uses, usability, and implementation requirements of patient-facing immersive VR in a community hospital providing post-acute care in Singapore. Methods: Using purposive sampling, nineteen healthcare professionals from a subacute hospital in Singapore participated in individual semi-structured interviews. Participants discussed their expectations of VR, experienced approximately 10–15 min of a reminiscence-oriented virtual tour of a local (now defunct) village, using an Oculus Quest head-mounted display, and subsequently reflected on its usability and potential applications. The Unified Theory of Acceptance and Use of Technology 2 informed the interview guide and was used as a sensitising framework. Interviews were transcribed verbatim and analysed using hybrid deductive–inductive framework analysis. Results: Four themes were identified: (1) clinical value was conditional on therapeutic relevance; (2) anticipated patient acceptability depended on person–content fit and social context; (3) simple navigation did not remove physical and sensory barriers; and (4) clinical integration required a supporting service model. Conclusions: Healthcare professionals viewed immersive VR as potentially useful but not yet ready for routine clinical implementation. Adoption in post-acute care should be guided by therapeutic purpose, accessibility, and integration into existing workflows rather than technological novelty alone. Future research and development should prioritise co-designed therapeutic content, patient selection, accessibility, workflow integration, and implementation support.

1. Introduction

Post-acute care plays an increasingly important role in helping patients recover function, regain independence, and prepare for a safe return to the community [1,2]. Within the post-acute care continuum, community hospitals provide sub-acute rehabilitation, particularly to populations such as stroke, orthopaedic, neurological, and geriatric rehabilitation patients. In community hospitals, rehabilitation extends beyond scheduled therapy sessions. With the ageing population, demand for community hospital resources is rising, and the need to sustain rehabilitation beyond inpatient settings becomes vital for recovery. Patients spend substantial periods outside formal treatment, during which opportunities for meaningful activity, cognitive stimulation, and continued practice may be limited. Technologies that can complement staff-delivered care may add value if they are clinically relevant, acceptable to patients, and feasible to integrate into existing workflows.

Immersive virtual reality (VR), delivered through a head-mounted display, is becoming increasingly popular in healthcare and offers one possible approach towards improving rehabilitation outcomes [3,4]. By creating interactive and context-rich experiences beyond the physical boundaries of the ward, immersive VR has been explored for rehabilitation, symptom management, cognitive stimulation, and patient engagement [3,4]. Studies in selected clinical populations, including people recovering from stroke and patients receiving oncology care, suggest that VR can be feasible and acceptable and may help sustain attention, encourage participation, or provide distraction from symptoms [5,6]. For older adults, applications have also included reminiscence, cognitive stimulation, virtual travel, and activities intended to support psychosocial well-being and quality of life [7].

However, most evidence on the use of VR technology comes from feasibility studies conducted in controlled or small-scale settings, which does not necessarily translate into routine clinical usefulness [8]. The value of VR depends not only on the immersive technology itself but also on whether its content is relevant to patients’ goals, capabilities, and lived experiences. This is particularly important in post-acute care, where patients may have multimorbidity, frailty, sensory impairment, cognitive difficulties, reduced mobility, or limited familiarity with digital technology. Cybersickness, visual fatigue, headset weight, discomfort, and difficulties operating the equipment may further limit its use [9,10]. Other practical aspects of usage and implementation should also be considered, such as the need for staff to screen suitable patients, set up and clean the device, provide supervision, troubleshoot technical problems, and integrate VR activities into rehabilitation plans. A technology that is engaging in isolation may therefore have limited clinical value if it creates additional workload for staff, lacks a clear therapeutic purpose, or cannot be used safely and consistently.

This gap between technological potential and real-world implementation remains insufficiently addressed. A recent scoping review of VR implementation in healthcare found that many studies focused on acceptability or short-term outcomes, while paying less attention to contextual fit, staff responsibilities, technical infrastructure, workflow integration, and sustained use [8]. These implementation considerations are especially pertinent in community hospitals, where care is interdisciplinary and directed towards functional recovery, discharge preparation, and community reintegration [11,12]. Compared with acute and specialist settings, community hospitals generally care for a high proportion of older and frailer patients over longer periods of admission. This creates potential opportunities for VR-supported rehabilitation and meaningful inpatient activity but also raises distinct questions regarding patient selection, accessibility, staff supervision, infection control, content relevance, and operational ownership.

Healthcare professionals are a critical stakeholder group to examine because they are likely to act as both gatekeepers and facilitators of implementation. Their perceptions may influence patient access to the technology, its integration into clinical practice, the perceived therapeutic value of content, and the extent to which operational requirements are regarded as manageable. Gaining insights into healthcare professionals’ perspectives before wider deployment may reveal potential discrepancies between technological innovation and clinical needs and clarify the contextual conditions that support the successful adoption and use of VR in healthcare settings.

Existing research on healthcare professionals’ attitudes towards VR suggests cautious optimism. A recent systematic review found that mental health professionals held generally positive attitudes towards the use of VR therapies, yet clinical adoption remained limited [13]. Similarly, reviews of the usability and adoption of VR and related technologies indicate that uptake can be constrained by time pressures, low confidence or self-efficacy in operating the devices, and the perception that VR is suitable only for selected patient groups, such as those without significant cognitive impairment [14,15]. Moreover, much of this research has been conducted in acute, specialist, or mental health settings; healthcare professionals in community care settings remain an underrepresented group, despite serving predominantly older populations for whom usability constraints may be greatest. Taken together, the literature suggests that positive attitudes do not necessarily translate into higher uptake of VR, and that integration into usual workflows may be necessary to increase adoption [8]. Exploring the views of healthcare professionals in a community hospital therefore addresses a recognised gap.

The Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) is the theoretical model that underpins this study [16]. UTAUT2 describes several factors that shape technology acceptance, including performance expectancy, effort expectancy, social influence, facilitating conditions, hedonic motivation, price value, and habit [16]. Although UTAUT2 was originally developed to explain an individual user’s behavioural intention and technology use, its constructs have also been used as sensitising concepts to guide data collection and analysis in qualitative health-technology research [17,18,19]. UTAUT2 was used to guide the framing of interview questions and provided an initial scaffold for examining perceived usefulness, usability, social influences, and implementation conditions. It was not used to assume that actual adoption or sustained use had occurred.

Therefore, the aim of this study was to explore healthcare professionals’ perceptions of the potential clinical uses, usability, patient acceptability, and implementation requirements of patient-facing immersive VR after exposure to a reminiscence-oriented prototype in a Singapore community hospital. The prototype served as a concrete, shared stimulus for eliciting perceptions of immersive VR rather than as a candidate rehabilitation intervention under evaluation.

2. Materials and Methods

2.1. Study Design and Reporting

This qualitative descriptive study explored healthcare professionals’ perspectives on immersive VR in the community hospital setting. The study was conducted at Outram Community Hospital (OCH), a 448-bed sub-acute facility in Singapore that primarily serves older adults undergoing rehabilitation or continuing care following acute hospitalisation. Interviews were conducted in a private meeting room where only the interviewer and the participant were present. Data collection took place between July 2024 and May 2025. A qualitative descriptive design using naturalistic inquiry is appropriate for generating a practice-oriented account of perceived barriers, facilitators, and implementation considerations. The study is reported with reference to the Consolidated Criteria for Reporting Qualitative Research (COREQ) [20].

2.2. Participants and Recruitment

Eligible participants were 21 years or older, employed as full-time staff at Outram Community Hospital at the time of recruitment, able to communicate in English or Mandarin, and able to provide written informed consent. Participants were excluded if they had a history of severe motion sickness associated with nausea, vomiting, or headache; a previous head injury or condition that could restrict headset use; or were unable to provide consent independently.

Potentially eligible patient-facing healthcare professionals were identified from relevant clinical and allied health departments at OCH. Purposive sampling was used to obtain representation across professional groups involved in patient care and rehabilitation. Sampling deliberately prioritised rehabilitation and therapy staff (physiotherapists, occupational therapists, and therapy assistants), as they were envisaged to be the primary users, gatekeepers, and facilitators of any patient-facing VR service, while other patient-facing roles (patient experience ambassadors and a dietitian) were included to broaden the range of perspectives; nurses, doctors, and technical and operational staff were not represented in the final sample, which is acknowledged as a limitation. Eligible staff were approached progressively, in successive waves and in parallel with concurrent data analysis, by a member of the research team and invited to participate; all staff approached agreed to take part, and no further staff were approached once data saturation was judged to have been reached. Written informed consent was obtained by a team member who was not the participant’s supervisor or direct reporting officer. Data saturation was assessed concurrently with data collection, through recurrent analysis of the accumulating transcripts and coding framework [21]. During data analysis, the research team observed that no substantively new codes or themes were emerging after approximately 14 interviews. Five additional interviews were conducted to confirm and refine the thematic structure and to ensure that perspectives from the various healthcare professional groups were adequately captured. Recruitment ceased after 19 interviews, when the coding framework required no further changes and data saturation was therefore deemed to have been achieved.

2.3. Data Collection

The interview guide was informed by UTAUT2 and developed through iterative team discussions; it contained open-ended questions that covered topics relating to prior experience with technology, perceived clinical usefulness, ease of use, enjoyment, anticipated patient and caregiver responses, affordability and value, and conditions required for implementation. The guide had been piloted with two participants to improve the clarity of the questions. The two participants involved in piloting the interview guide were not included in the final sample or analysis. Each interview question was mapped a priori to the UTAUT2 construct it was intended to explore, and interviewers were provided with the corresponding construct definitions; the full interview guide, including the correspondence between each question and its UTAUT2 construct, is presented in Supplementary Table S2. Questions for healthcare professionals focused on clinical applications, patient suitability, workflow, and implementation support.

Prior to each interview, participants were informed that the study aimed to explore healthcare professionals’ perceptions of immersive VR in the community hospital setting, and that findings would be used to inform future development and implementation. Each interview began with a discussion of the participant’s previous experience and expectations. The interviewer then explained how to use the Oculus Quest head-mounted display (models MH-A64 and MH-A32; Oculus, Menlo Park, CA, USA), before participants experienced a prototype virtual tour of Kampung Lorong Buangkok developed by the Singapore Institute of Technology in collaboration with FX Media Internet Pte Ltd (Singapore). The application allowed users to view photographs and videos, navigate using gaze-based prompts, explore different areas of the virtual environment, and engage with recorded narratives. The prototype was deliberately employed as an elicitation stimulus, grounding the interview in a concrete, shared experience of immersive VR, rather than as a candidate rehabilitation intervention in itself. Each session was delivered in a standardised manner: the headset was pre-loaded with the virtual environment, the same navigation instructions were given by the interviewer, and participants used the application for approximately 10–15 min before resuming the interview to discuss their experience, possible clinical uses, limitations, and suggested improvements. No participant terminated the VR session prematurely. Transient sensations such as giddiness or heaviness reported during or after use were explored within the interview itself; adverse reactions were captured through participants’ interview accounts rather than a separate structured log.

Interviews lasted 30 min on average, excluding the VR exposure session. All interviews were conducted in English and audio-recorded with participants’ consent. Recordings were transcribed verbatim for analysis. Quotations presented in the manuscript were lightly edited for readability by removing verbal fillers, repeated words, false starts, and colloquial discourse particles, and by making minor grammatical corrections where necessary. These edits did not alter the substantive meaning of the quotations.

2.4. Data Analysis

The findings were organised in NVivo 12 and a hybrid deductive–inductive framework analysis was undertaken [22]. P.J.J.L. and M.H.T. first familiarised themselves with the transcripts. All transcripts were coded by P.J.J.L. and M.H.T. To promote consistency, an initial subset of transcripts was coded independently by both researchers, and differences in code application were discussed and resolved with S.S.Y.S. before the remaining transcripts were divided between the two coders. Inductive codes were proposed where data could not be adequately indexed against the UTAUT2-informed framework; each new code was defined and documented in the NVivo codebook and reviewed at team meetings. Framework revisions and the rationale for merging or reclassifying codes were recorded in meeting minutes, which served as an analytical audit trail. UTAUT2 and the interview guide informed an initial analytical framework, but codes that were not adequately represented by UTAUT2 were added inductively. Transcripts were indexed systematically, and coded data were charted across participants to examine for recurring patterns, differences, and less common or contradictory views. The analysis distinguished between participants’ own experiences of using the headset and their anticipated views about patients, caregivers, workflows, and organisational adoption.

After 11 interviews had been analysed, P.J.J.L., M.H.T., and S.S.Y.S. reviewed the coding structure, revised theme and subtheme groupings, and reclassified codes where necessary. The team met weekly to discuss interpretation and resolve differences by consensus. S.S.Y.S. supervised the analysis. The final themes were confirmed with the whole research team, who brought expertise from medicine and public health, and interpreted in relation to UTAUT2 while retaining findings that extended beyond, or did not fit neatly within, its original individual-user constructs. Habit was not treated as an established finding because participants had only a single exposure to the prototype. The full coding structure, including initial codes, subthemes, and their mapping to UTAUT2 constructs, is presented in Supplementary Table S1.

2.5. Research Team, Reflexivity and Positionality

Interviews were conducted by F.F.Y.L., P.J.J.L., M.H.T. and F.F.Y.L. is a male physiotherapist with a Bachelor of Science (Honours) in Physiotherapy. P.J.J.L. is a male occupational therapist with a Bachelor of Science (Honours) in Occupational Therapy. M.H.T. is a female research officer with a Bachelor of Arts (Honours) in Sociology. S.S.Y.S. is a female research fellow with a PhD in Public Health and training in programme evaluation. F.F.Y.L. and P.J.J.L. worked in the same hospital as participants and may have been known to some of them through professional interactions, although neither was a direct supervisor of any participant. This prior familiarity may have influenced how freely participants expressed critical views and was considered when interpreting the findings. M.H.T. and S.S.Y.S. had no prior professional relationship with the participants. Several steps were taken to mitigate the risk of response bias arising from professional familiarity: written informed consent was obtained by a team member who was not the participant’s supervisor or direct reporting officer; interviews were conducted in a private room with only the interviewer and participant present; participants were assured that their responses were confidential, would not be shared with colleagues or supervisors, and would have no bearing on their employment; and the coding and thematic analysis were led by team members with no prior professional relationship with participants.

M.H.T. and S.S.Y.S. had formal training in qualitative research, and M.H.T. supervised the interview protocol. P.J.J.L. and F.F.Y.L., while not formally trained in qualitative research prior to the study, were trained in the interview protocol by M.H.T. before data collection commenced. The VR prototype was developed by the Singapore Institute of Technology in collaboration with FX Media Internet Pte Ltd., prior to and independently of this study. Although several authors are affiliated with the Singapore Institute of Technology, no member of the research team was involved in commissioning, developing, funding, or advocating for the VR prototype, and the study funder had no role in the design of the study, the collection, analysis, or interpretation of data, or the decision to publish.

The team approached the study with an expectation that VR might offer clinical value in the community hospital setting, given the broader evidence base for VR in rehabilitation and patient engagement. This prior orientation was acknowledged as a potential source of confirmation bias, particularly during coding and theme development. Reflexive field notes and analytic memos were not systematically maintained, which is acknowledged as a limitation of the study’s reflexive practice. This risk was partially mitigated by weekly team meetings at which interpretations were deliberately challenged and alternative explanations considered, by the documentation of analytical decisions in meeting minutes, and by the deliberate retention of negative and contradictory cases during analysis.

2.6. Rigour and Trustworthiness

Rigour and trustworthiness were addressed with reference to established criteria for qualitative research [23]. Credibility was supported through systematic comparison within and across cases, attention to less common or contradictory accounts, and collaborative interpretation, with interpretive differences discussed until consensus was reached. Dependability was enhanced by involving multiple members of the team in reviewing the data and coding structure and in checking analytical decisions for alignment: consistent with Mays and Pope’s emphasis on transparent exposition of data collection and analysis and consideration of alternative explanations [23], the provisional analytical framework was reviewed after 11 interviews, and codes and theme groupings were revised where necessary. To support transferability, the study setting and participants are described, and verbatim participant quotations are reported alongside the findings. Inductive findings that did not fit UTAUT2 were retained rather than forced into the framework. Transcripts were not returned to participants for correction, and formal member checking of the findings was not undertaken.

2.7. Ethical Considerations

The research was conducted in accordance with the Declaration of Helsinki, and the study was approved by the SingHealth Centralised Institutional Review Board (reference 2023-2266). All participants provided written informed consent before participation.

3. Results

3.1. Participant Characteristics

Nineteen eligible healthcare professionals were approached; all agreed to participate, provided written informed consent, and completed an interview. There were no refusals or withdrawals. Most were women (n = 13; 68.4%) and of Chinese ethnicity (n = 14; 73.7%). Participants were aged 21 to 50 years. Participants were staff from rehabilitation and therapy services, including therapy assistants, physiotherapists, and occupational therapists, who together accounted for 16 of 19 participants (84.2%). Participant characteristics are shown in Table 1.

Table 1. Participant characteristics (N = 19).

Characteristic	Category	n (%)
Gender	Male	6 (31.6)
	Female	13 (68.4)
Ethnicity	Chinese	14 (73.7)
	Malay	3 (15.8)
	Indian	1 (5.3)
	Other	1 (5.3)
Age group, y	21–30	8 (42.1)
	31–40	6 (31.6)
	41–50	5 (26.3)
Education attainment	Secondary	2 (10.5)
	Post-secondary	2 (10.5)
	‘A’ levels or Diploma	4 (21.1)
	University	11 (57.9)
Profession	Physiotherapist	5 (26.3)
	Occupational therapist	3 (15.8)
	Patient experience ambassador	2 (10.5)
	Dietitian	1 (5.3)
	Therapy assistant	8 (42.1)

More than half of the participants (63.1%) had no prior experience with using technology in patient care. No participant reported prior regular use of an immersive VR head-mounted display in either personal or clinical contexts. The rest reported the use of ‘Silver Pad’, a tablet for conducting application-based cognitive activities with patients (21.1%); assistive devices such as a continuous passive motion (CPM) knee machine and EsoGlove during patients’ rehabilitation sessions (10.5%); and DEXIE, a socially assistive robot for engaging patients living with dementia (5.3%).

3.2. Overview of Themes

Four themes captured how participants perceived the prototype: (i) clinical value was conditional on therapeutic relevance; (ii) anticipated patient acceptability depended on person–content fit and social context; (iii) simple navigation did not remove physical and sensory barriers; and (iv) clinical integration required a supporting service model. To reduce repetition across sections, the detailed subtheme content is presented once: Table 2 summarises each theme, its analytical interpretation, and its relationship to the UTAUT2 constructs, and Section 3.3, Section 3.4, Section 3.5 and Section 3.6 elaborate each theme with illustrative quotations.

Table 2. Themes, analytical interpretation, and relationship to UTAUT2.

Theme	Analytical Interpretation and UTAUT2 Relationship	Illustrative Quotation
1. Clinical value was conditional on therapeutic relevance	Participants perceived possible benefits for reminiscence and meaningful activity, but the current prototype was not viewed as sufficient for rehabilitation. Primarily related to performance expectancy and hedonic motivation.	“If they are going to buy and it is going to be the same pictures or videos, then I think it does not add any value.” (OCH 304, Occupational therapist, >20 years of experience)
2. Anticipated acceptability depended on person–content fit and social context	Staff anticipated that receptivity would vary with prior familiarity with technology, relevance of the content, and encouragement from peers or caregivers. Related to social influence and anticipated performance expectancy, but represents staff perceptions rather than patient data.	“It really depends on how the patient embraces new technology.” (OCH 313, Therapy assistant, 3–5 years of experience)
3. Simple navigation did not remove physical and sensory barriers	Gaze-based navigation was generally straightforward, but giddiness, headset weight, visual clarity, and spectacle compatibility could limit use. Related to effort expectancy and facilitating conditions.	“Overall, it was quite understandable—you see the white dot, move to that cursor, and then it enters.” (OCH 311, Physiotherapist, <1 year of experience)
4. Clinical integration required a supporting service model	Participants wanted multilingual and rehabilitation-oriented content, staff training, troubleshooting support, and a clearer value proposition. Related to facilitating conditions, performance expectancy, hedonic motivation, and hypothetical price value.	“They will need to know and be confident enough to troubleshoot it.” (OCH 303, Physiotherapist, 1–3 years of experience)

Supplementary Table S1 provides the complete thematic structure, including subthemes, illustrative quotations, and secondary UTAUT2 mappings.

3.3. Theme 1: Clinical Value Was Conditional on Therapeutic Relevance

Participants identified possible roles for immersive VR in reminiscence and meaningful inpatient activity. The Kampung Lorong Buangkok environment prompted discussion of culturally familiar places and experiences, and some participants anticipated that similar personalised environments could support conversation or memory-oriented activities.

“We can use it for maybe reminiscence, like bring the patient out of the hospital to visit places that they are familiar with. We can check with the patient where they frequently go, and then if we can find something and put it in, maybe it can help them in terms of memory.”

(OCH 304, Occupational therapist, >20 years of experience)

Participants also described VR as a possible way to provide stimulation during prolonged periods of inactivity in hospital. These comments reflected perceived potential rather than observed patient outcomes.

“For me, it is just like keeping patients occupied, rather than sometimes lying down there for so many hours doing nothing. Sometimes they get so bored.”

(OCH 317, Therapy assistant, >20 years of experience)

Some participants believed that selected patients living with mild-to-moderate cognitive impairment might find familiar or calming content engaging. However, staff did not regard all immersive content as inherently therapeutic. They repeatedly distinguished between an enjoyable heritage tour and an application that could directly support rehabilitation. The current prototype was therefore seen as illustrative rather than ready for routine clinical use.

“It is quite limited (in utility). If they are going to buy and it is going to be the same pictures or videos, then I think it does not add any value.”

(OCH 304, Occupational therapist, >20 years of experience)

3.4. Theme 2: Anticipated Patient Acceptability Depended on Person–Content Fit and Social Context

Participants anticipated substantial variation in patients’ willingness to try VR. They attributed possible reluctance to fear of unfamiliar technology, concerns about eye strain or safety, and limited prior exposure. These were staff expectations about patients rather than findings from patients themselves.

“It depends on how the patient embraces new technology. Some patients could be very fearful of new technology, especially these goggles. They think it could hurt their eyes, so they will not be willing to use it. But some might be willing to try.”

(OCH 313, Therapy assistant, 1–3 years of experience)

Age alone was not viewed consistently. Some participants expected younger patients to be more comfortable with technology, whereas others believed the heritage content would be more meaningful for older patients. This tension suggested that the relevance of the content and the patient’s previous experiences might matter more than chronological age in isolation.

“The slightly elderly age group (might be more engaged by the goggles), because of the content of the kampung. The younger ones may look at it but may get bored after a while. The older ones, it is for their memory.”

(OCH 301, Physiotherapist, 1–3 years of experience)

The social environment of the ward was also seen as influential. Participants described how seeing another patient use the headset could prompt curiosity and willingness to try it. Caregiver support was considered important when patients required reassurance, assistance, or consent-related discussion.

“It is more of how much we educate the next-of-kin and how much they understand about it. That would perhaps form their perception of whether they will be keen for the patient to use it.”

(OCH 303, Physiotherapist, 1–3 years of experience)

3.5. Theme 3: Simple Navigation Did Not Remove Physical and Sensory Barriers

Most participants found the gaze-based navigation understandable after instruction. The visual cursor and directional prompts reduced the need to learn complex controller movements.

“It was generally quite easy. There were some parts where I felt very high up and had to bend down to see the footprints, which was a bit harder. But overall, it was quite understandable—you see the white dot, move to that cursor, and then it enters.”

(OCH 311, Physiotherapist, <1 year of experience)

Nevertheless, straightforward navigation did not eliminate physical discomfort. Participants reported or anticipated giddiness, nausea, a heavy-headed sensation, and neck or upper-limb effort from supporting the headset. They considered these issues potentially more consequential for older or frailer patients.

“I need to look up and down, here and there, to find the step or signage. I do not have a motion-sickness issue, but I did feel a bit giddy and heavy-headed.”

(OCH 308, Therapy assistant, 3–5 years of experience)

Visual clarity differed across users. Although some participants found the visuals and signs clear, spectacle wearers described blurring and difficulty reading text. Realistic environmental audio was viewed positively and contributed to immersion, but sensory quality did not offset concerns about physical comfort and accessibility.

“Because I am wearing glasses, everything seemed a bit blurred. When reading the words, I had to squeeze my eyes.”

(OCH 303, Physiotherapist, 1–3 years of experience)

3.6. Theme 4: Clinical Integration Required a Supporting Service Model

Participants’ suggestions extended beyond changes to the headset. They wanted selectable languages, a wider range of culturally and personally relevant environments, and interactive content linked to rehabilitation goals. Entertainment and novelty were considered insufficient unless the application also offered clear clinical or functional value.

“If it is programmed to do some exercises, for example total knee replacement bed exercises, it could help them practise when they are free and follow the exercises being taught in the goggles.”

(OCH 313, Therapy assistant, 3–5 years of experience)

Content variety was also viewed as necessary to maintain interest in using the device. Participants suggested games and movement-based activities but emphasised that content should be developed with occupational therapists, physiotherapists, and other relevant professionals rather than added solely for novelty.

“Maybe games—how they are going to manoeuvre their body and do the movements. It would also be good if they were able to play games.”

(OCH 314, Therapy assistant, 3–5 years of experience)

Routine use was thought to require staff who could set up the device, screen users, respond to discomfort, troubleshoot technical problems, and know where to obtain further support. These requirements implied that VR would function as a service requiring ownership and workflow integration, rather than as a device that could simply be placed on a ward.

“They will need to know how to troubleshoot and be confident enough to troubleshoot it. You do not want something to happen, and they do not know how to fix it, and then it is just lying there.”

(OCH 303, Physiotherapist, 1–3 years of experience)

Participants discussed purchase or post-discharge use only conditionally. They believed patients or caregivers might consider paying for a device if it contained credible exercise or cognitive-training applications. Because no actual price, purchasing decision, or economic outcome was evaluated, these findings represent an anticipated value proposition rather than evidence of price value or cost-effectiveness.

“If the device teaches you how to exercise or engage in cognitive training activities, then there is a possibility that the caregiver would want to purchase it.”

(OCH 316, Dietitian, 10–20 years of experience)

4. Discussion

This study explored how healthcare professionals perceived patient-facing immersive VR after exposure to a reminiscence-oriented prototype in a Singapore community hospital. Participants perceived possible value for reminiscence, culturally meaningful engagement, and inpatient stimulation. However, their support was conditional: the prototype would need relevant therapeutic content, appropriate patient selection, accessible hardware, and a supporting operational model before clinical use could be justified. The findings therefore concern perceived acceptability and anticipated implementation requirements, not actual adoption, patient benefit, or cost-effectiveness.

The first important finding was the distinction between immersion and clinical usefulness. Participants appreciated the capacity of VR to transport users beyond the ward environment and believed that familiar places might support reminiscence or conversation. Similar potential has been described in studies of VR experiences in stroke rehabilitation and in the development of VR applications for people living with dementia [24,25]. While participants did not view the Kampung Lorong Buangkok tour as a rehabilitation intervention in itself, they wanted applications that supported explicit functional, cognitive, or therapeutic goals. This finding cautions against treating engagement, novelty, or enjoyment as substitutes for clinical benefit. It also distinguishes the present findings from studies in which purpose-built, goal-directed VR content has achieved measurable therapeutic gains; for example, a structured therapeutic role-playing VR game targeting concentration and fine motor skills produced significant improvements in attention among children with intellectual disabilities [26]. Read together, these findings suggest that the limiting factor for clinical VR in post-acute settings is not immersion itself but the availability of content engineered around explicit therapeutic goals—something the passive, video-based touring content examined here did not provide.

Second, participants viewed acceptability as a matter of fit between the person, the content, and the social setting. Previous qualitative studies have described apprehension among people with limited VR experience and more positive engagement after supported exposure [27]. Research on older adults’ interactions with digital devices likewise suggests that perceived usefulness, confidence, prior experience, and support can be as important as age [28]. In the present study, staff made differing assumptions about whether younger or older patients would be more receptive. These accounts should not be interpreted as evidence that age or education predicts or correlates with acceptance; instead, they show why patient and caregiver perspectives must be examined directly. The culturally specific prototype used in this study may have been more personally meaningful to some older adults while remaining less engaging for others.

Third, the findings demonstrate that apparent ease of navigation does not establish usability for the intended patient population. Participants were younger than the community hospital patients for whom the technology was being considered, and people with severe motion sickness or relevant head-injury histories were excluded. Even within this selected staff sample, some reported giddiness, weight-related discomfort, and difficulty seeing text while wearing spectacles. Cybersickness and physical discomfort are well-recognised limitations of immersive VR [9,10]. Before use in the patient population, it is therefore important to conduct feasibility testing with older adults and individuals with motion-sickness or related histories; such testing should include tolerability, visual and hearing accessibility, cognitive load, postural requirements, fall risk, duration of use, and explicit stopping criteria rather than relying on staff impressions alone.

Fourth, participants framed implementation as a sociotechnical and service-design problem. Their suggestions encompassed content curation, language accessibility, patient screening, staff training, device setup, troubleshooting, workflow ownership, and post-discharge support. This aligns with the implementation literature showing that healthcare technologies are unlikely to be sustained when technical functionality is considered separately from staff roles, local workflows, and organisational support [8,15]. Claims that VR will reduce manpower should therefore be tested rather than assumed. Depending on the clinical workflow, headset preparation, supervision, cleaning, troubleshooting, and monitoring for adverse effects may create additional work before any efficiency gains are realised. Although individual barriers such as cybersickness, headset weight, and maintenance demands are well documented [9,10,14], the contribution of the present study lies less in identifying these barriers anew than in showing how they configure together, in a post-acute community hospital, into a distinctly conditional pattern of acceptance: staff support was contingent on therapeutic relevance, person–content fit, accessibility, and a resourced service model, rather than on any single barrier or enabler. This conditional configuration, and its explicit mapping against UTAUT2 constructs, differentiates the present findings from prior usability- and attitude-focused reports conducted in acute and specialist settings [13,14,24,25,27].

UTAUT2 offered a useful vocabulary for discussing expected benefit, ease of use, social influence, facilitating conditions, enjoyment, and value [16]. However, the analysis also exposed limitations in applying the model to this setting. The interviewed staff were sometimes describing their own use of the headset, sometimes anticipating patient or caregiver responses, and sometimes judging organisational investment. These actors do not share a single behavioural intention or use context. For this reason, UTAUT2 was most useful as a sensitising framework rather than a complete explanatory model. Three specific tensions were apparent when applying UTAUT2 in this setting, and these indicate both the limits and the incremental value of the framework. First, performance expectancy in its original formulation concerns gains in the user’s own performance, whereas participants evaluated the technology chiefly in terms of anticipated therapeutic benefit for patients; perceived value was thus judged vicariously, on behalf of a third party, a dynamic that the model does not explicitly accommodate. Second, facilitating conditions in UTAUT2 refer to an individual’s perception that support exists, whereas participants described facilitating conditions as an organisational service model—staffing, training, troubleshooting, infection control, and workflow ownership—that must be collectively designed and resourced rather than merely perceived. Third, several inductively derived subthemes, including person–content fit, the cultural relevance of content, and hygiene requirements, extended beyond the original constructs, suggesting that in institutional healthcare settings UTAUT2 is best treated as a starting scaffold requiring supplementation with content-related and organisational constructs. These observations are consistent with recent methodological reviews of UTAUT2 in qualitative health research [17] and clarify the incremental value of the framework in this study: it enabled systematic comparison across accounts while making explicit where technology acceptance in a community hospital diverges from the individual consumer adoption context for which the model was developed. Habit could not be assessed after a single session, and price value was only hypothetical because participants were not asked to make an actual purchasing decision using a specified price. Future implementation studies may benefit from combining technology-acceptance constructs with an implementation framework that explicitly addresses intervention characteristics, workflow, organisational context, and implementation outcomes.

4.1. Implications for Practice and Research

The findings do not support broad deployment of the tested prototype. They instead support a staged development and evaluation pathway. Initial work should identify specific use cases, such as facilitated reminiscence, relaxation, meaningful activity, or selected rehabilitation exercises, and co-design the content with patients, caregivers, therapists, nursing staff, and operational or technical teams. Each use case should specify the target population, therapeutic rationale, contraindications, supervision requirements, session duration, adverse-event procedures, cleaning and storage arrangements, and accountable staff role.

For post-acute community hospitals specifically, these findings translate into several actionable steps. First, VR sessions should be scheduled as structured adjuncts within existing therapy timetables—for example, as facilitated reminiscence or relaxation blocks between formal therapy sessions—rather than offered as unsupervised ward-based entertainment, with therapy assistants trained and rostered as session facilitators. Second, a simple two-stage screening protocol should be embedded into the admission or therapy-planning workflow, combining exclusion screening (for example, a history of severe motion sickness, vestibular or seizure disorders, or wounds and precautions affecting headset use) with suitability triage (cognition, vision and hearing, sitting balance, and preferred language). Third, content procurement should prioritise multilingual, locally and culturally relevant environments co-designed with occupational therapists and physiotherapists, with rehabilitation-oriented modules, such as guided bed exercises or functional task practice, specified before purchase. Fourth, each ward or department deploying VR should designate an operational owner responsible for device charging, cleaning and infection control between users, fault reporting, and escalation pathways. Finally, adoption should proceed through a time-limited pilot with predefined stopping criteria, structured adverse-event recording, and measurement of workflow burden before wider procurement decisions are made.

Subsequent feasibility studies should involve patients and caregivers directly and assess acceptability, demand, usability, tolerability, fidelity, workflow burden, and implementation cost. Comparative studies of different content types would help determine whether culturally familiar environments, interactive rehabilitation exercises, or passive immersive experiences offer sufficient added value over simpler and less costly alternatives. Only after feasibility and workflow integration have been established should studies examine sustained use, functional or psychological outcomes, healthcare utilisation, and cost-effectiveness.

4.2. Study Limitations

Nevertheless, several limitations should be acknowledged. First, participants experienced a single culturally specific VR prototype, which was intended to prompt reflection on the potential use of immersive VR in clinical practice rather than to evaluate the programme itself. Their views may therefore have been influenced by the specific content presented. Second, no clinical implementation occurred, and the study did not assess patient outcomes, sustained use, workflow effects, resource utilisation, or cost-effectiveness. Third, findings relating to patients and caregivers were based on healthcare professionals’ perceptions rather than direct accounts and may therefore reflect professional assumptions. Direct evaluation with patients and caregivers is required before any conclusions can be drawn about patient acceptability, usage-related barriers, or clinical effects. Fourth, the sample comprised predominantly rehabilitation and therapy staff, limiting representation from other clinical, technical, and operational groups within the community hospital. Fifth, participants were aged 21–50 years, and those with severe motion sickness or relevant head-injury histories were excluded, which may have resulted in a more favourable assessment of usability than might be observed in older or frailer patients. Sixth, each participant’s exposure to the prototype was brief (approximately 10–15 min) and occurred on a single occasion; perceptions of usability, comfort, and potential adoption formed after such a short exposure may not capture the experiences that would emerge with repeated or prolonged use, and no conclusions about sustained acceptability or habit formation can be drawn. Finally, using UTAUT2 to inform both data collection and the initial analytical framework may have foregrounded expected technology-acceptance constructs, although inductive coding was used to capture findings beyond the framework.

5. Conclusions

Healthcare professionals identified potential roles for immersive VR in reminiscence, meaningful inpatient engagement, and selected rehabilitation activities in post-acute care. However, its perceived clinical value depended on four interrelated considerations: therapeutic relevance, an engaging and accessible user experience, person–content fit, and adequate implementation support. Participants valued the immersive and reminiscence potential of VR while also identifying physical and sensory barriers that could limit its accessibility and use. These findings provide a basis for the co-design and feasibility testing of clinically relevant VR applications. For future work, evaluation with patients and caregivers is needed to determine whether these attributes translate into sustained use, clinical benefit, and successful integration into post-acute care.