A Novel Psychosocial Virtual Reality Intervention (BMT-VR) for Patients Undergoing Hematopoietic Stem Cell Transplantation: Pilot Randomized Clinical Trial Design and Methods

Abstract

Background:

Although patients undergoing hematopoietic stem cell transplantation (HSCT) must cope with psychological distress and isolation during an extended transplant hospitalization, psychosocial interventions to address these unmet needs are lacking. Virtual reality offers an innovative modality to deliver a patient-centered psychosocial intervention to address psychosocial needs of patients undergoing HSCT. However, there are currently no supportive care interventions leveraging virtual reality in patients undergoing HSCT.

Objective:

To describe the methods of a randomized clinical trial (RCT) to assess the feasibility and preliminary efficacy of a self-administered, virtual reality-delivered psychosocial intervention (BMT-VR) to improve psychological distress and quality of life (QOL) for patients hospitalized for HSCT.

Methods:

This study entails a single-center RCT of BMT-VR compared to usual transplant care in 80 patients hospitalized for HSCT. Adult patients with hematologic malignancies hospitalized for autologous or allogeneic HSCT are eligible. BMT-VR includes psychoeducation about the HSCT process, psychosocial skill building to promote effective coping and acceptance, and self-care and positive psychology skills to promote post-HSCT recovery. The primary aim is to assess the feasibility defined a priori as ≥ 60% of eligible patients enrolling in the study, and of those enrolled and randomized to the BMT-VR, ≥ 60% completing 4/6 BMT-VR modules. Secondary objectives include assessing the preliminary effects on psychological distress and QOL.

Discussion:

This is the first RCT of a virtual reality-delivered psychosocial intervention for the HSCT population. If deemed feasible, a future larger multi-site clinical trial can evaluate the efficacy of BMT-VR on outcomes for patients hospitalized for HSCT.

Introduction

Hematopoietic stem cell transplantation (HSCT) is a common and potentially curative therapy for many patients with malignant and non-malignant hematologic diseases.^1–4 However, patients undergoing HSCT must endure immense physical (e.g., mucositis, pain) and psychological (e.g., depression) symptoms during an extensive and socially-isolating 3–4-week hospitalization due to toxicities from pre-HSCT conditioning chemotherapy. ^4–6 Moreover, patients struggle with dramatic quality of life (QOL) deterioration and increased psychological distress during their HSCT hospitalization as up to 40% of patients report clinically significant depression and anxiety symptoms.^7–9 Unfortunately, 30% of patients also report acute traumatic stress symptoms due to the trauma related to their HSCT hospitalization.^10,11 Notably, the physical and psychological distress patients experience during HSCT can result in long-term morbidity including QOL impairments, post-traumatic stress symptoms, as well as post-HSCT complications (e.g., financial challenges).¹²

Despite the tremendous burden patients experience during their HSCT hospitalization, existing psychosocial interventions to promote effective coping during the HSCT hospitalization are limited, costly, difficult to replicate and scale, and personnel (e.g., social workers) are in scarce supply. Our prior work has shown that integrating specialty palliative care clinicians during the HSCT hospitalization is effective for improving psychological distress and QOL during the HSCT hospitalization and up to six months post-HSCT.¹³ However, given the limited availability of trained specialty palliative care and mental health clinicians, the majority of patients undergoing HSCT cannot access existing efficacious supportive interventions.^13–15 Consequently, innovative and accessible models of psychosocial care are needed to address the unmet supportive care needs of patients undergoing HSCT.

Digital health interventions, including virtual reality-delivered interventions, are novel and show promise in addressing the psychosocial needs of patients with cancer.^16–18 Compared to digital applications that have shown promising efficacy for reducing psychological distress in patients with psychiatric disorders^19–25 and cancer,²⁶ the three-dimensional and multi-sensory immersive capabilities of virtual reality may enhance user engagement and mitigate the isolation and loneliness patients experience during their prolonged HSCT hospitalization. Interestingly, virtual reality has been successfully used to deliver interventions (e.g., exposure-based treatments) for psychiatric conditions including anxiety disorders.^27–31 However, to date, virtual reality has not been leveraged to deliver psychosocial interventions for patients undergoing HSCT. Hence, we developed a novel, self-administered, psychosocial virtual reality-delivered intervention (BMT-VR) for hospitalized patients undergoing HSCT and use a randomized clinical trial to assess its feasibility and preliminary efficacy for improving patient’s QOL, psychological distress, symptom burden, coping, and self-efficacy.

Methods

Conceptual Framework

Our conceptual framework (Figure 1) for BMT-VR development is informed by 1) the empirical model of our integrated palliative care intervention for HSCT, which suggests that palliative care improves patients’ outcomes via leveraging psychoeducation, supportive psychotherapy, and psychosocial skill building to improve symptom burden, psychological distress, and coping ^13,32 and the 2) Technology Acceptance Model which emphasizes that patients will accept and use digital health technology based on their perception of its usefulness and ease-of-use.^33,34 Hence, during the HSCT hospitalization, BMT-VR may enhance symptom burden management and self-management strategies. More specifically, it may represent an opportunity to improve self-efficacy and supportive psychotherapy skills for coping and management of expectations and uncertainties that commonly accompany the HSCT hospitalization and recovery. Ultimately, BMT-VR could positively impact health-related outcomes like QOL and mood during the HSCT hospitalization and post-HSCT. BMT-VR also contains several features to foster patients’ engagement, including educational games, optional tailored content, resources to promote adaptive skill-building, and a record of progress through the intervention. Informed by the Technology Acceptance Model, these features improve usability and retention with rates ranging from 85–100% and represent critical components of behavior change theory.^35,36

Fig. 1.

Conceptual model.

Study Design

Our study design (Figure 2), a pilot randomized clinical trial (NCT05629676), was modeled after the National Institute of Health Stage Model for Behavioral Intervention Development (i.e., Stage 1b), which proposes a framework for intervention development where treatment development is ongoing until it reaches its maximum level of effectiveness and is implementable in the target population.³⁷ This pilot randomized trial may be the first to establish the feasibility and preliminary effects of a virtual reality-delivered intervention for improving psychological distress and QOL among patients hospitalized for HSCT or any hospitalized oncology population. Hence, this work will guide future efficacy (Stage 2 and 3) and dissemination (Stage 4) trials of BMT-VR.

Fig. 2.

Study design.

Participants

Our inclusion criteria entail adults (≥ 18 years old) with hematologic malignancies admitted for autologous or allogeneic HSCT at the Massachusetts General Hospital (MGH) who can speak, understand, read, and respond to questions in English, since BMT-VR is only available in English for this single-site feasibility study. We will consider how the content of BMT-VR can be culturally tailored and translated to other languages for future larger multi-site trials. We will exclude patients undergoing: 1) HSCT for benign hematologic conditions, 2) outpatient HSCT, or 3) patients with acute or unstable psychiatric or cognitive conditions which will interfere with their ability to complete informed consent or comply with the study procedures. We will exclude patients undergoing HSCT for benign conditions and those undergoing outpatient HSCT because their recovery trajectories are different than those undergoing inpatient HSCT.^38,39

Recruitment and Enrollment

A clinical research coordinator will review the scheduled HSCT census weekly to identify potentially eligible patients and inquire about concerns regarding study participation from their HSCT clinician. If the HSCT clinician has no objections, the clinical research coordinator will approach patients for study participation within 72 business hours of their HSCT hospitalization. Eligible patients who are interested will undergo written informed consent and completion of baseline assessments. Consented participants will be given 48 business hours from consent to complete baseline assessments prior to enrollment and randomization to either the BMT-VR or usual care control groups.

Randomization

Once a patient has completed baseline assessments, a member of the research team (independent from study staff) will perform 1:1 randomization using a computer-generated randomization schema, stratified by transplant type (autologous vs. allogeneic HSCT) since transplant type affects symptom burden and stress that may accompany the HSCT hospitalization,⁴⁰ with assignments concealed in numbered envelopes. Participants will not be blinded to the intervention or usual care control group, like prior technology delivered interventions in this population.⁴¹

The BMT-VR Intervention

Principles from the Technology Acceptance Model^33,34 and our understanding of the psychosocial needs of patients undergoing HSCT from prior supportive intervention studies in HSCT^13,15 undergirded the conceptual framework for BMT-VR development. Additionally, a thorough review of the literature,¹⁴ key components of our efficacious palliative care intervention for HSCT¹³ and expert input from our multidisciplinary team of HSCT clinicians and researchers from oncology, palliative care, psychiatry, and psychology also contributed to the development of the comprehensive script for BMT-VR’s content. Accordingly, the six modules of BMT-VR include: 1) psychoeducation and managing expectations and the stress of the HSCT process; 2) coping skills and strategies to promote adjustment, validation, and coping with the intense HSCT hospitalization; 3) psychosocial skill building to promote mindfulness, acceptance, and gratitude in the midst of uncertainties that commonly accompany the HSCT hospitalization and recovery; 4) psychoeducation to specifically manage patient’s expectations for HSCT and the recovery period as well as to mobilize social supports; 5) psychosocial skill building to foster effective coping strategies, optimism, and adaptive thinking for common transitions of care in HSCT; and 6) an overview of psychosocial skills grounded in cognitive behavior therapy, mindfulness, and positive psychology.^{10 25,26}

A comprehensive, iterative process for digital therapeutics development and numerous rounds of user testing and review by our research team was also foundational to intervention development.⁴² We designed the BMT-VR modules to be an educational and interactive experience for patients as they navigate through their HSCT hospitalization. Hence, BMT-VR is self-administered with several features to promote adaptive coping,^13,43 engagement and health behavior change including gamification strategies,⁴² videos of HSCT survivors, and optional content. To translate the BMT-VR script into the virtual reality technology frames (Figure 3), we collaborated with Novobeing, a leading virtual reality digital health company with the necessary technical expertise to construct the sophisticated virtual reality algorithm and platform for BMT-VR. We worked closely with Novobeing to ensure the intervention is user-friendly for the unique needs of the HSCT population.

Fig. 3.

Two-dimensional representation of BMT-VR intervention frames.

Intervention Procedures

Participants randomized to the intervention group will use BMT-VR during their HSCT hospitalization with a suggested timeline of reviewing two modules each week to complete the required modules by week 4 of their hospital stay. Once a patient is randomized to use BMT-VR, a study team member will provide the patient with the virtual reality headset device. The study team member will then orient the patient to the use of the device and the intervention set-up, and participants will also receive an information sheet which details how the device works with intervention set-up. Additionally, a study team member will do weekly check-ins to troubleshoot any issues with the device or intervention as the study is in progress. Once a patient completes the intervention and we retrieve the virtual reality headset, we will monitor adherence to BMT-VR by electronically tracking various metrics including the time spent with the device, time spent with each module, module environment most accessed, time of the day the device was used, and module completion.

Intervention Fidelity

The study team will ensure fidelity of the intervention delivery by monitoring and reviewing weekly data which the virtual reality software will collect, including modules accessed and completed, proportion of each module completed, and time spent on each module.

Usual Care Control Condition

We will use a usual care control condition, and participants assigned to the usual care group will not receive BMT-VR. We define usual care as the standard psychosocial care for all patients hospitalized to undergo HSCT which entails a 1:1 in-person meeting with an inpatient HSCT social worker. This routine psychosocial care with social workers entails supportive therapy which does not focus on psychosocial skill-building or cognitive strategies emphasized in the BMT-VR intervention. We chose a usual care control condition versus an attention-matched control offering a placebo intervention because of the ethical reasoning that we should not put patients hospitalized for HSCT through the unnecessary burden of a placebo intervention during the vulnerable hospitalization period instead of standard care.⁴⁴

Study Outcome Measures

Our conceptual framework will guide the specific patient-reported outcome questionnaires with strong psychometric properties and responsiveness to change which are commonly used to assess psychological distress and QOL in this population.¹³ We will administer questionnaires at week-2, week-4, week-12, and week-24. We carefully chose these time points to assess patients at the peak of symptom severity during the HSCT hospitalization (i.e., week-2) based on the type of transplant as well as their expected recovery post-HSCT. ¹³ Participants will be able to complete questionnaires on paper or electronically via a secured REDCap survey link, or over-the-phone.

Sociodemographic information

All participants will use a self-report questionnaire to provide their sociodemographic data including age, race, ethnicity, gender identity, marital status, religion, education, income, and perceived confidence with technology.

Primary Endpoint

Feasibility:

The primary endpoint for this study is feasibility defined as at least 60% of eligible patients enrolling in the study, and of those enrolled and randomized to BMT-VR, at least 60% completing four out of the six BMT-VR modules.^32,45,46

Secondary Endpoints

Anxiety and Depression:

We will use the 14-item, Hospital Anxiety and Depression Scale (HADS), with two separate 7-item subscales to evaluate symptoms of anxiety and depression during the past week.²⁷ Each subscale score ranges from 0–21 and higher scores imply more symptoms of anxiety and depression. Participants will also complete the 9-item Patient Health Questionnaire-9 (PHQ-9) to assesses major depressive disorder symptoms according to the Diagnostic and Statistical Manual of Mental Disorders, fifth edition.⁴⁷ Higher scores (range, 0–27) on the PHQ-9 also suggests worse depression.

Post-traumatic Stress Disorder Symptoms:

We will use the 17-item Post-traumatic Stress Disorder Checklist-Civilian Version (PCL) to evaluate severity of PTSD symptoms.²⁰ Higher scores (range, 17–85) suggests worse PTSD symptoms.

Quality of Life:

We will use the 47-item, Functional Assessment of Cancer Therapy-Bone Marrow Transplant (FACT-BMT) to assesses QOL in five domains (i.e., physical, social, emotional, and functional wellbeing, as well as bone marrow transplant-specific symptoms).²⁸ Higher scores (range, 0–164) suggests better QOL.

Symptom Burden:

We will use the 10-item, revised Edmonton Symptom Assessment Scale (ESAS-R) to assess various symptoms relevant to patients undergoing HSCT.^48,49 As in previous work, we will define scores 4–10, as moderate-to-severe symptoms.^50,51

Coping:

We will use the 13-item Measure of Current Status Part A (MOCS-A) to measure self-perceived status on several coping skills such as relaxation, cognitive restructuring, and assertive communication.⁵² Higher scores (range, 0–52) suggests greater coping skills.

Self-efficacy:

We will use the 17-item Cancer Self-efficacy Scale-Transplant (CASE-T) to assess patients’ confidence in managing the impact of their illness.⁵³ Higher scores (range, 0–170) indicate greater self-efficacy.

Usability of BMT-VR (only those randomized to BMT-VR):

We will use the 10-item System Usability Scale (SUS) at week-4 post-intervention to assess the usability of BMT-VR.⁵⁴ Higher scores (range, 0–100) suggests higher perceived usability of BMT-VR.

Qualitative Exit Interviews

We will use semi-structured phone exit interviews with 20 participants randomized to BMT-VR to collect additional feedback about the intervention. We will purposefully sample these participants to ensure representation among patients with high and low level of engagement with BMT-VR as well as transplant type. We will audio record all exit interviews.

Statistical Methods

We will describe participant characteristics with descriptive statistics (e.g., mean, standard deviation) for continuous variables and proportions for categorical variables. We define a p-value <0.25 a priori as promising for preliminary efficacy in this pilot study. We will also consider a p-value of 0.05 to be statistically significant. We will use an intention-to-treat approach for treatment group comparisons. We will adhere to the consolidated standards of reporting clinical trials (CONSORT) standards.⁵⁵

Feasibility

Feasibility is the primary outcome for the study. We will compute the feasibility using the proportion of eligible patients who enroll in the study and the proportion of enrolled and randomized participants to the BMT-VR group who complete at least 4/6 BMT-VR modules. The 60% feasibility benchmark is commonly used in behavioral intervention studies.^15,42

Secondary Endpoints

We will first assess the internal consistency of our patient-reported outcome measures using Cronbach’s alphas.⁵⁶ We will compare patient’s QOL and psychological distress symptoms (i.e., depression, anxiety, PTSD symptoms) at week-2 and week-4 between the study groups, controlling for baseline values and demographic and clinical factors (as necessary for any imbalances in baseline variables) using Analysis of Covariance (ANCOVA). We will also use longitudinal mixed effect models with Maximum Likelihood to account for missing data when examining the effect of BMT-VR on QOL, depression, anxiety, and PTSD symptoms longitudinally across all timepoints. Since symptom burden, coping, and self-efficacy are key mediators of supportive care intervention effects on QOL and psychological distress symptoms,⁵⁷ we will compare patients’ coping, symptom burden, and self-efficacy between the study groups at week-2 and week-4 using ANCOVA. Similarly, we will use mixed linear effect models with Maximum Likelihood for missing data to examine the effect of BMT-VR on these outcomes longitudinally across all timepoints.

Missing Data

For our initial analyses, we will focus on study completers to estimate the effect of BMT-VR on patients who completed the intervention as intended, without assumptions about missing data. Our use of the intention-to-treat principle with all randomized subjects and sensitivity analyses will also allow us to explore how various assumptions about missing data and differences between completers and non-completers affect the estimated outcomes. However, if data seem missing at random, we will use multiple imputation methods,⁵⁸ maximum likelihood estimate approach with EM algorithm,⁵⁹ and mixed-effects modeling⁶⁰ to adequately account for data missing at random. Alternatively, if we discover that participants do not complete the study because of disease worsening, suggesting missing data are not random, we will use pattern mixture modeling or joint modeling approaches⁶¹ to analyze incomplete data, and perform sensitivity analysis⁶² to assess the impact of missing data.

Power Consideration

We chose a sample size of 80 participants based on the feasibility of completing the project during the proposed timeframe and ability to assess the preliminary efficacy of the intervention. Consistent with the behavioral intervention literature for this population, at least 30 participants are needed in each group in a pilot study to estimate a parameter.^63,64 Therefore, the proposed sample size will provide us with preliminary data that can be utilized to determine the effect size and adequately power future randomized trials.

Qualitative Data Analysis

We will transcribe verbatim, code, and thematically analyze all recorded exit interviews. Two team members will independently review the transcriptions and utilize a rapid analysis framework developed from the interview guide and refined through transcript review.⁶⁵ The two coders will compare their findings and discuss disagreements to achieve consensus.

Discussion

Although patients hospitalized for HSCT have persistent psychological distress, QOL deficits, and experience immense isolation during their HSCT course, novel, accessible, and engaging psychosocial interventions to address these needs are lacking. Hence, we provide a comprehensive overview of the first pilot randomized clinical trial to assess the feasibility and preliminary effects of a novel virtual reality-delivered psychosocial intervention, BMT-VR, tailored to address the psychosocial needs of patients undergoing HSCT. This work will provide foundational data and a framework for leveraging virtual reality to deliver novel supportive care interventions for HSCT recipients and other patients experiencing prolonged hospitalizations. The use of a virtual reality psychosocial intervention is particularly novel in oncology and HSCT. The patient-centered and self-administered nature of this intervention has the potential to address the unmet distress symptoms and isolation commonly experienced by patients as they manage a prolonged hospitalization and recovery in isolation.^14,66–68 Compared to other digital health interventions (e.g., those using text-messaging and mobile application) with promising efficacy for improving psychological distress in various populations,^19–25 the three-dimensional features of virtual reality allow patients to be transported to more desirable virtual environments (e.g., gardens, tropical beaches), which may promote engagement and mitigate the isolation and loneliness often experienced by this population. To our knowledge, virtual reality psychosocial interventions for HSCT and other oncology populations have not been tested. While a few studies have shown feasibility of leveraging virtual reality among hospitalization patients, ^69–76 this study will be the first to show whether patients undergoing HSCT can really engage with the virtual reality platform, and it will provide data regarding the preliminary effects of such an intervention for improving the QOL and care for this vulnerable population.

Compared to other virtual reality interventions among hospitalized patients which have focused on specific symptoms like anxiety or pain,^69–76 we adapted the core multicomponent features (e.g., psychoeducation about common physical and psychological stressors that may accompany HSCT and psychotherapy strategies) for the BMT-VR intervention from our team’s palliative care intervention for HSCT, which was efficacious for improving patients’ psychological distress symptoms and QOL.¹³ Despite the benefits of enhancing specialty palliative care and other existing psychosocial care models for patients undergoing HSCT, severe shortages of trained palliative care and specialty mental health clinicians undermine access to current models of supportive care for HSCT. ^77–79 Hence, digital health interventions like BMT-VR have the potential to enhance access and broaden dissemination of supportive care interventions for not only the HSCT population but also potentially all hospitalized patients with cancer.

There are several limitations for this study worth considering. First, our study methods and feasibility data may not be generalizable as our study will be performed at a major tertiary academic HSCT center. Second, while our sample size of 80 participants is sufficient for feasibility and exploring preliminary effects of the intervention on study outcomes, studies with larger sample sizes are needed to demonstrate the efficacy of BMT-VR on patient reported outcomes including psychological distress symptoms and QOL. Third, for this pilot study, the content of BMT-VR was only in English. However, for future larger trials of BMT-VR, we will translate and culturally adapt the content in other commonly used languages in the US including Spanish to reach patients from diverse and ethnic minority backgrounds. Overall, this randomized clinical trial of the first VR-delivered psychosocial intervention for hospitalized HSCT patients offers important groundwork information for addressing the psychosocial needs of vulnerable populations, including and beyond the HSCT population. If the BMT-VR intervention is feasible, future large randomized controlled efficacy trials in diverse populations will help to establish its efficacy on patient outcomes.

Table 1.

BMT-VR Intervention Modules and Sample Content

Module	Theme	Sample Content
One	Expectations for common HSCT related stress	Roadmap of the HSCT experience from pre-HSCT to the acute recovery for patients undergoing autologous and allogeneic HSCT emphasizing common physical and psychological symptoms.
Two	Supportive psychotherapy to promote adjustment, validation, and coping during hospitalization	Provides psychoeducation on problem and emotion-focused coping strategies, for patients during their transplant hospitalization. An interactive game is included, allowing patients to practice categorizing coping strategies as problem-focused or emotion-focused.
Three	Psychosocial skills for mindfulness, acceptance, and positive psychological well-being like gratitude	Several mindfulness and breathing exercises which patients can self-select as part of the intervention.
Four	Mobilizing social support for recovery	Provides psychoeducation on assertive communication and how to utilize assertive communication skills to mobilize support outside the hospital and prepare for discharge from the transplant hospitalization.
Five	Adaptive thinking for transitions of care in HSCT	A review of thought distortions and cognitive reframing around common stressors pertaining to transitioning from care inside the hospital to outside the hospital.
Six	Psychosocial skills grounded in Cognitive Behavior Therapy, Mindfulness, and Positive Psychology	Exercises designed to introduce cognitive behavior therapy, mindfulness, and positive psychology through gamification.