Abstract
Pediatric inpatient caregivers often experience significant anxiety. Although improving caregiver anxiety is critical for improving patient outcomes, inpatient interventions targeting patients’ families are rarely implemented. This study assessed the effectiveness of a language-concordant, virtual reality (VR)-guided meditation on inpatient caregiver anxiety. Using a prospective, pragmatic, randomized study, the primary aim assessed caregiver anxiety using the Visual Analog Scale for Anxiety (VAS-A). Secondary aims explored state and trait anxiety using the State-Trait Anxiety Inventory (STAI), differences in anxiety reduction between English- and Spanish-speaking participants, and satisfaction. With 200 participants included in the final analysis, VAS-A scores in the VR group were lower compared to the standard of care (SOC) group (p = 0.0005), with a greater reduction for Spanish-speaking participants (p = 0.005). After controlling for baseline trait anxiety, the STAI found a greater reduction in state anxiety in the VR group (p ≤ 0.0001). VR-guided meditation effectively reduced caregiver anxiety in pediatric inpatient settings. VR's immersive and versatile nature offers diverse therapeutic options to improve caregiver and patient health, including marginalized populations with previously limited interventions.
Keywords
Key Points
VR-guided meditation effectively reduced inpatient pediatric caregiver anxiety across multiple established anxiety measures.
Given the need of inclusivity of diverse populations in mental health research, this study investigated the effects of VR-guided meditation on Spanish-speaking populations, where results indicated a greater reduction in anxiety for Spanish-speaking participants compared to English-speaking participants.
Satisfaction ratings were high for participants in the VR group, which has implications for acceptability and usability in clinical settings.
Introduction
Caregivers, defined as the parents or primary caretakers of pediatric patients, often experience stress and anxiety throughout their children's hospital stay, even when the hospitalization is brief. 1 Nearly two-thirds of caregivers report moderate anxiety, and over 30% experience severe anxiety.2,3 Following discharge, more than 25% of parents experience post-traumatic stress symptoms. 4 Caregivers are the primary support for their hospitalized children and their behaviors directly impact their children's well-being. 5 Untreated caregiver distress exacerbates their children's anxiety and pain perception.5–7 Although most hospital care teams primarily focus on patient healing, recognizing and alleviating caregiver emotional distress directly benefits their child's health.8,9
Various interventions target caregiver anxiety, including music therapy, counseling, play therapy, and audiovisual (AV) technologies.7,10–12 However, most strategies specifically focus on preoperative caregiver anxiety, with limited attention towards postoperative and inpatient caregivers. Despite the effectiveness of these interventions, widespread adoption of routine caregiver anxiety treatment is low, and caregiver anxiety remains prevalent. 2
It is not surprising that caregivers’ distress is commonly under-recognized in the hospital since mental health ailments in the community are also undertreated compared to somatic conditions. 13 Under-recognition is even more pronounced in Spanish-speaking communities. 14 This community disparity translates into the hospital, where Spanish-speaking caregivers are at a greater risk of untreated anxiety due to language barriers and cultural differences. 15 Given the inconsistent attention to caregivers, their distress remains prevalent in hospital settings, necessitating targeted strategies. 16
Mindfulness and meditation techniques promote overall well-being and physiologic relaxation.17,18 Despite their benefits, there is a lack of understanding of Hispanic individuals’ responses to mindfulness-based approaches. 19 Furthermore, in-person, guided meditation, while effective, is resource-intensive and time-consuming, reducing its practicality in hospital settings. Feasible alternatives rely on technological advancements in digital software, which enhance accessibility, customization, schedule flexibility, and efficacy.20,21 Virtual reality (VR) is a promising anxiolytic technology with increasing clinical adoption due to its ability to immerse patients in completely novel, peaceful environments. Compared to tablets or computers, VR is portable, increasingly affordable, and provides fully immersive AV meditation. 22 Caregivers facing stress during inpatient hospitalizations may benefit from VR-guided meditation. A critical aspect of pediatric care involves implementing family-centered care, where families are engaged in the healing process. 23 Addressing caregiver anxiety through an inclusive, family-centered approach improves healthcare quality and patient outcomes. 24
This study aimed to reduce caregiver anxiety in the pediatric inpatient setting by utilizing VR combined with language-tailored mindfulness approaches. As a non-pharmacologic alternative, we hypothesized that VR-guided meditation would be associated with decreased caregiver anxiety. The primary aim evaluated the effectiveness of VR-guided meditation in reducing self-reported caregiver anxiety. The secondary aims explored the influence of caregiver personality traits on anxiety, treatment differences between English-speaking and Spanish-speaking caregivers, and intervention satisfaction.
Methods
Setting and Context
We conducted a pragmatic, prospective, randomized study (NCT05729347, 2/6/23) from June 2023 to February 2024 that was approved by the Stanford Institutional Review Board (IRB #69068). Interventions occurred at Lucile Packard Children's Hospital Stanford (Palo Alto, CA), a quaternary care, academic pediatric hospital with 365 inpatient beds. Participants 18 years and older serving as caregivers for hospitalized pediatric patients were recruited. Caregivers with a history of seizures, motion sickness, significant cognitive impairment, were pregnant, using corrective glasses, experiencing nausea, or had children who were clinically unstable or required urgent interventions were excluded.
Intervention
Eligible participants provided written informed consent, and after enrollment, were randomly assigned to either the standard of care (SOC) or VR group prior to starting the intervention. Randomization occurred via a random number generator provided by Research electronic data capture (REDCap). 25 Both groups completed demographic and preintervention surveys and during both SOC and VR interventions, participants were observed by a study research assistant to ensure both safety and fidelity. SOC participants were asked to engage in their usual anxiety-alleviating activities that they typically perform during their visits, such as reading, using their phone, talking with someone, or listening to music. VR group participants were fitted with an adjustable VR headset (Oculus Go, Meta, Inc.) that displayed a meditation application called Aurora (Stanford Chariot Program), a guided 6-minute meditation experience. The application provided AV cues for controlled breathing exercises within a simulated relaxing scenery of waterfall landscapes and northern lights. The guided meditation voice was presented in either English or Spanish depending on the participant's language preference, with the Spanish version interpreted by the hospital interpreter staff to ensure accuracy.
Outcomes
The primary outcome investigated the anxiolytic effect of the VR-guided meditation on caregiver anxiety. The first secondary outcome explored the association of caregiver anxiety traits with their state of anxiety while engaged in the intervention. The next secondary outcome evaluated the differences in the intervention's efficacy between English-speaking and Spanish-speaking caregivers. The final secondary outcome explored caregiver satisfaction with VR.
Measures
Caregiver anxiety was measured using the Visual Analog Scale for Anxiety (VAS-A) assessment, a self-reported measure for overall anxiety (Appendix A). 26 This scale ranges from 0 (no anxiety) to 100 (worst possible anxiety) and has been validated as a measure of anxiety in postoperative and inpatient hospital settings.27,28
To measure caregiver personality traits and their effects on anxiety, caregivers completed the standardized State-Trait Anxiety Inventory (STAI). 29 This scale consists of a 40-item, self-reported measure of anxiety using a 4-point Likert scale (Appendix B). The items are divided into two 20-item subscales assessing state and trait anxiety, resulting in a maximum score of 80. Classification of state and trait anxiety scores consists of low (20-39), moderate (40-59), and high (60-80). 29 The STAI compared caregiver states while controlling for baseline trait anxiety.22,30 Completion of the study required participants to complete VAS-A and STAI before and after the intervention.
To evaluate the differences between English- and Spanish-speaking caregiver anxiety, demographic data were collected including age, sex, ethnicity, and race. VAS-A and STAI results were evaluated for English- and Spanish-speakers and group differences were compared.
Caregiver satisfaction was evaluated using a satisfaction survey consisting of a 5-item questionnaire using a 5-point Likert scale (Appendix C). The VR group also completed an additional VR questionnaire to assess VR satisfaction. This questionnaire is composed of 4 items assessing VR satisfaction using a 5-point Likert scale and an additional 2 items assessing prior VR use and adverse side effects (Appendix D). 22 All surveys were provided in the participant's preferred language.
Sample Size
An anxiety score of 40 or less is considered mild anxiety on the VAS-A scale. 26 Given baseline caregiver anxiety of 70.55 ± 26, we powered the study to show a clinically relevant reduction to a score of 40, indicating mild anxiety. 22 Assuming an alpha of 0.05 and a power of 80, sample size calculations indicated 84 participants, with a target of 100 for each group to account for incomplete data and dropouts.
Analysis
Linear regression models of VAS-A and STAI scores were performed for the SOC and VR groups and for English-speaking and Spanish-speaking caregivers. STAI trait scores were controlled in the secondary outcome evaluating anxiety. Within-group mean differences were tested for normality and analyzed using a one-sample t-test or Wilcoxon test. Satisfaction between groups was assessed with a Wilcoxon rank-sum test and the VR group satisfaction survey was reported as descriptive statistics.
Results
Demographics
A total of 212 participants were recruited and 200 caregivers were included in the final analysis (Table 1). The majority of caregivers were between 31 and 50 years old (74.5%) (Table 1). Half of participants were English-speaking and half were Spanish-speaking. Twelve participants were excluded due to not completing the study interventions (Figure 1). SOC group (n = 99) self-soothing methods included using their cell phone (45.5%), talking with others (62.6%), reading (9.1%), listening to music (5.1%), and other activities such as watching television (33.3%) (multiple methods were allowed) (Table 1).

Participant flow diagram. Study flow diagram illustrating participant recruitment, screening, enrollment, allocation, drop out, and study conditions. For all participants, recruitment, enrollment, and study conditions were performed on the same day.
Participant Demographics.
Multiple answers allowed.
Primary Outcome
For the SOC group (n = 99), participants had a preintervention VAS-A mean of 51.9 ± 27.9 (95% confidence interval (CI): 46.4, 57.5) and a postintervention mean of 43.84 ± 28.6 (CI: 38.2, 49.5). For the VR group (n = 101), participants had a preintervention VAS-A group mean of 53.3 ± 33.1 (CI: 46.8, 59.7) and a postintervention mean of 31.42 ± 27.7 (CI: 26.0, 36.8). The postintervention VAS-A scores in the VR group were lower than those in the SOC group (p = 0.0005) (Figure 2A).

Standard of care (SOC) versus Virtual Reality (VR) mean preintervention and postintervention (A)Visual Analog Scale Anxiety and (B) State-Trait Anxiety Inventory (STAI) state anxiety (B). The Violin Chart depicts the group mean anxiety as colored regions with 95% confidence intervals, and individual scores with solid dots with frequency density depicted by violon plot width. Abbreviations: CI, confidence intervals. Significance: *** p ≤ 0.001, **** = p ≤ 0.0001.
Secondary Outcomes
For the preintervention trait scores between the VR and SOC groups, there were no differences in baseline trait anxiety (p = 0.63). The SOC group had a preintervention trait mean of 40.9 ± 10.5 (CI: 38.8, 43.0) and a postintervention mean of 39.5 ± 11.2 (CI: 38.8, 43.0) (p = 0.02). The VR group preintervention trait mean was 40.2 ± 11.5 (CI: 37.9, 42.5) and the postintervention mean was 33.8 ± 10.7 (CI: 31.7, 35.9) (p ≤ 0.0001). Postintervention trait scores were different between the two groups (p = 0.0002). A linear regression controlling for preintervention state and trait anxiety was performed. Normality checks on the residuals were performed and the assumptions were met. There was a reduction in postintervention state anxiety scores for the SOC group (p = 0.01) and the VR group (p ≤ 0.0001) (Figure 2B). For the SOC group, the mean preintervention state anxiety was 44.9 ± 12.9 (CI: 42.3, 47.5) and 41.6 ± 13.9 (CI: 38.9, 44.4) after the intervention (p ≤ 0.0001). For the VR group, the preintervention state anxiety was 44.5 ± 13.0 (CI: 41.9, 47.1), which decreased to 33.4 ± 11.3 (CI: 31.2, 35.7) after using VR (p ≤ 0.0001). The reduction in state anxiety was greater in the VR group (p ≤ 0.0001).
For the English-speaking SOC group (n = 49), participants had a preintervention VAS-A mean of 52.9 ± 27.4 (CI: 45.2, 60.6) and a postintervention mean of 46.5 ± 26.7 (CI: 39.0, 54.0) (p = 0.06). For the English-speaking VR group (n = 51), participants had a preintervention VAS-A group mean of 55.4 ± 28.8 (CI: 47.5, 63.3) and a postintervention mean of 41.7 ± 27.0 (CI: 34.3, 49.1) (p ≤ 0.001). The change in English-speaking postintervention VAS-A scores in the VR group was not different compared to those in the SOC group (p = 0.13). For the Spanish-speaking SOC group (n = 50), participants had a preintervention VAS-A mean of 51.0 ± 28.7 (CI: 42.9, 59.2) and a postintervention mean of 41.2 ± 30.3 (CI: 32.6, 49.8) (p ≤ 0.01). For the Spanish-speaking VR group (n = 50), participants had a preintervention VAS-A group mean of 51.1 ± 37.2 (CI: 40.5, 61.6) and a postintervention mean of 20.9 ± 24.4 (CI: 14.0, 27.9) (p ≤ 0.0001). The postintervention Spanish-speaking VAS-A scores in the VR group were lower than those in the SOC group (p ≤ 0.0001). When comparing the postintervention anxiety VAS-A scores between the English-speaking and Spanish-speaking VR groups, the Spanish-speaking VR group had lower postintervention anxiety scores (p = 0.005) (Figure 3A).

English-speaking and Spanish-speaking VR group preintervention and postintervention comparison of mean (A)Visual Analog Scale Anxiety and (B) State-Trait Anxiety Inventory (STAI) state anxiety (B). Colored bars indicate group means with their respective CI indicated by the error bars. Significance: ** p ≤ 0.01, *** = p ≤ 0.001, **** = p ≤ 0.0001.
When comparing the preintervention anxiety traits between the English-speaking and Spanish-speaking VR groups, there were no differences (p = 0.14). The English-speaking VR group had a preintervention trait mean of 41.8 ± 11.8 (CI: 38.5, 45.2) and a postintervention mean of 36.4 ± 10.5 (CI: 33.4, 39.4) (p ≤ 0.0001). The Spanish-speaking VR group had a preintervention trait mean of 38.5 ± 11.1 (CI: 35.3, 41.6) and a postintervention mean of 31.2 ± 10.4 (CI: 28.2, 34.12) (p ≤ 0.0001). Postintervention trait scores were different between the two groups (p = 0.01), with the Spanish-speaking group reporting a greater reduction in trait anxiety. A linear regression controlling for preintervention state and trait anxiety was performed. Normality checks on the residuals were performed with assumptions met. The postintervention state anxiety scores were lower for the Spanish-speaking VR group compared to the English-speaking VR group (p = 0.03). For the English-speaking VR group, the mean preintervention state anxiety was 45.8 ± 11.9 (CI: 42.5, 49.2), which reduced to 36.4 ± 10.8 (CI: 33.4, 39.5) after the intervention (p ≤ 0.0001). For the Spanish-speaking VR group, the preintervention state anxiety was 43.2 ± 14.0 (CI: 39.2, 47.1), which decreased to 30.3 ± 11.1 (CI: 27.2, 33.5) after using VR (p ≤ 0.0001) (Figure 3B).
A caregiver satisfaction survey was collected from all participants. Seventy-seven of 101 caregivers in the VR group and 49 out of 99 of those in the SOC group expressed satisfaction or strong satisfaction (4 or 5 out of a 5-point scale) across all 4 satisfaction items. For the SOC group, mean satisfaction was 3.4 ± 0.9 (CI: 3.2, 3.5) and 3.9 ± 1 (CI: 3.7, 4.1) for the VR group (p ≤ 0.0001). For the VR questionnaire, VR group participants reported high satisfaction, with 81 of 101 (80.2%) caregivers expressing satisfaction or strong satisfaction (4 or 5 on a 5-point scale) across all satisfaction metrics (Appendix E).
Potential adverse events were monitored throughout the duration of the study. One participant reported dizziness and 3 reported eye strain, which all quickly resolved without further interventions.
Discussion
The primary aim assessed the effect of VR-guided meditation on inpatient caregiver anxiety. We observed a greater reduction in caregiver anxiety in the VR-guided meditation group compared to SOC. The first secondary outcome evaluated the association of trait and state anxiety between groups. In this STAI analysis, we confirmed that even though participants were a heterogeneous group, there were no differences in baseline anxiety traits between groups. After controlling for baseline trait and state anxiety, the caregivers in the VR group had a greater postintervention decrease in anxiety compared to SOC. The next secondary outcome explored the differences between English-speaking and Spanish-speaking caregivers, revealing that while both groups experienced anxiety reduction with VR, the effects were more pronounced in the Spanish-speaking group, suggesting that the VR-guided meditation was more effective at reducing anxiety in Spanish-speaking participants. The final secondary outcome revealed high satisfaction, with most participants indicating a strong desire to use VR-guided meditation in the future. With lowering hardware costs and improved user interfaces, practitioners should consider VR-guided meditations as an efficient method to target caregiver anxiety.
Given the high levels of caregiver anxiety in pediatric hospital settings, various interventions targeting anxiety have been developed, including music, educational programs, play therapy, and AV devices. 12 In this study, SOC participants engaged in similar activities such as reading or watching television, which resulted in some anxiety reduction. Although these standard methods were helpful, VR mindfulness resulted in even further reductions in anxiety, improving the mood in caregivers of pediatric inpatients. 31 In-person mindfulness approaches also reduce anxiety, but personal instruction and time prohibit wide-scale adoption, contrary to VR which offers self-guided meditation. 32 In addition, non-immersive, screen-based interventions do not block out surroundings, allowing visual and auditory distractions and difficulty maintaining attention to the mindful activity. 33 VR, however, places patients in a head-mounted display with complete immersion, heightening a sense of presence with reduced mind-wandering.10,34 VR-guided mindfulness effectively reduces caregiver anxiety, which has important implications for addressing caregiver mental health while caring for their hospitalized children.
Addressing caregivers’ needs enables healthcare practitioners to engage in family-centered care, which decreases caregiver distress while promoting psychological well-being. 35 Key principles of family-centered care involve providing families with emotional support, responding to their needs, and delivering health care in a culturally competent manner. The VR application used in this study effectively reduced caregiver burden by decreasing anxiety to improve mental health. Since caregivers play a vital role in the health and recovery of pediatric patients, prioritizing caregiver mental health leads to better health outcomes for pediatric patients and a more supportive environment for families. 36 Hospital providers should consider offering VR with mindfulness-based applications to address inpatient caregivers’ unmet needs. Additionally, VR is an effective analgesic and anxiolytic for pediatric inpatients. 37 Considering these benefits combined with those for caregivers, VR devices expand effective therapeutic options to both patients and their caregivers by providing a library of interventions within a single device.
In addition to family-centered care, this study also addressed inclusivity of Hispanic populations in mindfulness-based research and their mental health concerns. 19 Hispanic populations are disproportionately affected by unfavorable social determinants of health (SDOH). Those with limited English proficiency have higher mental health illnesses compared to the English-speaking population.38,39 Given the impact of SDOH on mental health, this underscores the importance of developing inclusive research and clinical programs that specifically target this growing populace.40,41 With a 59% decrease in the Spanish-speaking VR group compared to a 24.7% decrease in the English-speaking VR group, this study demonstrated greater anxiety reductions for Spanish-speaking participants utilizing a feasible and affordable intervention. Nonprofit companies, such as Invincikids, Inc., license software free of charge to hospital settings domestically and abroad, lowering the barrier for healthcare providers seeking to implement VR-based mindfulness.
This study had several limitations. Conducting the study at a single academic institution may limit generalizability. However, the sample size was large and included a heterogeneous group of participants. Despite this, future studies should include multicenter trials to enhance external validity. Second, caregivers’ past medical histories were not collected where unmeasured anxiety conditions could potentially bias results; although randomization should have mitigated this potential bias and the analysis indicated no difference in group baseline trait anxiety. Third, this study was conducted during a single session, where future research should investigate the long-term anxiolytic effects of repeated VR mindfulness interventions. Last, anxiolytic effects were collected on caregivers without investigating patient effects. Future investigations should measure pediatric patient anxiety and relevant outcomes to provide a holistic understanding of family-centered care.
Conclusion
VR-guided meditation was an effective, safe, and adaptable tool for reducing caregiver anxiety during pediatric inpatient hospitalizations. While both English- and Spanish-speaking caregivers experienced a reduction in anxiety, the effect was more pronounced in the Spanish-speaking group, highlighting the impact VR-guided mindfulness can have on populations with historically limited access to interventions.
Supporting caregiver mental health is crucial for the well-being of both caregivers and patients, enhancing the quality of care, meeting children's emotional and psychological needs, and aligning with family-centered care models. By leveraging its immersive, engaging, and versatile nature, VR lends itself to numerous therapeutic options, fostering better health outcomes for pediatric patients, and a more supportive environment for families. Future studies should explore factors influencing caregiver anxiety, the effects of mindfulness VR on other non-English speakers, and the long-term benefits of repeated mindfulness VR interventions during prolonged hospital stays.
Supplemental Material
sj-docx-1-jpx-10.1177_23743735251326671 - Supplemental material for Culturally Sensitive Treatment of Caregiver Anxiety With Virtual Reality: A Prospective, Pragmatic, Randomized Study
Supplemental material, sj-docx-1-jpx-10.1177_23743735251326671 for Culturally Sensitive Treatment of Caregiver Anxiety With Virtual Reality: A Prospective, Pragmatic, Randomized Study by Ricardo T Jimenez, Michelle Zuniga-Hernandez and Man Yee Suen, Ruth Feng, Kassandra Pinedo, Faaizah Arshad, Faith Collins, Avani Ganesan, Christian Jackson, Ellen Y Wang, Samuel T Rodriguez, Thomas J Caruso in Journal of Patient Experience
Footnotes
Author Contributions
Clinical Trial Registration
NCT05729347, February 6, 2023.
Declaration of Conflicting Interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: SR, EW, and TC are on the board of a nonprofit (Invincikids, Inc.) that seeks to distribute immersive technology to hospitalized children at no cost. They receive no financial compensation for their roles. The Stanford Chariot Program has received philanthropic donations from Meta Inc., and Magic Leap, Inc. SR received consultation fees from Apple, Inc. All other authors declare no competing interests or financial disclosures.
Ethical Approval and Registration
This clinical trial was registered under NCT 05729347 on February 6th, 2023. The study design was reviewed and received ethical approval by Stanford's Institutional Review Board (IRB 69068).
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by the Stanford Chariot Program.
IRB Registration
Stanford IRB #69068.
Statement of Human and Animal Rights
All procedures in this study were conducted in accordance with the Stanford's Institutional Review Board (IRB 69068) approved protocols.
Statement of Informed Consent
Written informed consent was obtained from the participants for their anonymized information to be published in this article.
ORCID iDs
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
