All participants engage in a full course of PT for MSK pain that is delivered by a trained and licensed physiotherapist. PT sessions are individually tailored based on the Guide to Physical Therapist Practice 3.0, consisting of (1) therapeutic exercise (strengthening and endurance exercises), (2) neuromuscular re-education (balance and proprioception exercises), (3) therapeutic activities (functional performance and activities of daily living), and (4) use of modalities (heat or cold packs). All adolescents receive a home exercise program (HEP) as part of the standard PT treatment. The full course of PT includes approximately 6 to 8 PT sessions delivered over 6 to 12 weeks. An adequate dose of treatment is considered to be 75% completion of the prescribed sessions. The number of treatment sessions varies based on patient presentation and individualized treatment goals as determined by the clinician and research team. Patients can continue any treatments they are currently involved in. If they choose to initiate a new treatment after PT treatment begins, they are asked to notify the research team immediately as it may affect their involvement in the study.

All participants receive a VR headset to use for the duration of treatment. The research team provides participants with an Oculus Quest 2 (Oculus) VR headset and orients the adolescents to its functionality.

For standard PT participants, the VR headset is preloaded with distraction-based games for recreational use at home until treatment is completed (Table 1). Clinicians do not discuss VR headset use with the standard PT participants. The research team provides an orientation session regarding the VR headset functions and preloaded games.

Physiorehabilitation with VR participants bring their VR headset to each PT appointment, and a portion of the session can be delivered in VR. Participating clinicians are told that the goal for VR use in a session is at least 8 minutes, but they are also given clinical decision-making power to use VR more or less as they determine clinically feasible. Physiorehabilitation with VR engages participants in a series of immersive games customized and chosen to allow for a progressive increase in standing endurance and support individual PT goals (Table 2). Importantly, this trial does not test a specific VR game or program but the broad implementation of VR in clinical practice. Games implemented for this purpose may include Fruity Feet (Stanford Chariot Program), Alien Defense (Stanford Chariot Program), Beat Saber (Beat Games), and Tilt Brush (Google), among others. Physiotherapists discuss VR headset use with physiorehabilitation with VR participants and actively incorporate VR activities into the HEP. The pediatric pain rehabilitation team at Stanford Children’s Health worked collaboratively with the Stanford Chariot Program to develop physiorehabilitation with VR content [24]. Fruity Feet was developed using a user-centered approach with patient and clinician end-user feedback across four phases: (1) needs assessment, (2) prototyping, (3) iteration and refinement, and (4) feasibility and acceptability. It was designed to be developmentally appropriate for youth by focusing on fun while leaning on stylized graphics and encouraging in-game feedback. Gameplay mechanics were built around PT movement goals, for example, multiplanar stepping (ie, forward, side, and back), stomping, marching, kicking, raising legs to different heights, and active ankle range-of-motion tasks for lower extremities. Importantly, it was also built to scale to a patient’s mobility, ensuring that patients of all abilities could play the game and benefit from the VR intervention. Consistent with the recommendations for VR clinical trial methodology, this user-centered iterative design process yielded a program that, in addition to gamification, provides back-end mechanics, giving PT clinicians the capability to control intensity, affected side or extremity emphasis, mirroring, and movement exaggeration to leverage the potential neuromodulatory effects of VR coupled with targeted pain PT.

Adolescents and caregivers complete baseline, discharge, and 3-month follow-up assessments. The adolescent completes daily diaries, and the caregiver completes weekly health cost diaries from the date of consent until the end of treatment at discharge. An additional 7 daily diaries by the adolescent and 1 additional health cost diary by the caregiver are completed at the start of the 3-month follow-up. All adolescent and caregiver surveys are completed on the web through the secure web-based app REDCap (Research Electronic Data Capture; Vanderbilt University), and the diaries are completed through the mobile phone–based app LifeData (LifeData, LLC) [42]. Baseline and discharge assessments are completed in person or on the internet, with the 3-month follow-up completed on the internet. Table 3 details the outcomes, measure names, respondents, and time of assessment.

Once eligibility is confirmed, the baseline assessment is completed in 2 parts. First, adolescents and caregivers complete consent and baseline self-report questionnaires. Adolescents and caregivers are oriented to the mobile diaries, which adolescents complete daily and caregivers complete weekly while enrolled in treatment. Following the baseline study visit, adolescents and caregivers undergo a pretreatment data collection period for the number of days between consent and their subsequent PT appointment. During this time, adolescents complete daily diaries, and caregivers complete weekly health cost diaries. The second part of the baseline assessment takes place at the adolescent’s next PT session, during which a research coordinator completes a baseline physical assessment via ViFive (ViFive, Inc) [43], a motion capture app, consisting of the 6-minute Walk Test, Single Leg Balance Test, and Closed Kinetic Chain Upper Extremity test, and the adolescent receives an ActiGraph watch (ActiGraph, LLC) that monitors sleep and activity level throughout the trial.

The discharge assessment occurs after all PT treatment sessions are completed. The discharge assessment consists of a physical assessment at the final session as well as self-report questionnaires. The diaries conclude, and the adolescents return the ActiGraph watch as well as the VR headset. Following completion of the study, adolescents and caregivers in the physiorehabilitation with VR group complete Zoom (Zoom Video Communications)-based or in-person semistructured interviews to provide feedback regarding their treatment experience. Finally, clinicians complete an acceptability measure assessing their experiences integrating VR into patient sessions for each participant in the physiorehabilitation with VR arm and, subsequent to discharge of their final trial patient, will complete Zoom-based semistructured interviews to further assess PT perceptions of feasibility in integrating VR.

The follow-up assessment occurs at 3 months after discharge, at which time adolescents complete a set of self-reported questionnaires as well as 7 additional daily diaries. Caregivers complete 1 additional health cost diary. Adolescents receive the battery of self-report questionnaires via REDCap and respond to the daily diaries via LifeData. Caregivers receive their single health cost diary via REDCap.

Following each PT session, the adolescent’s clinician completes a brief series of questions regarding the adolescent’s engagement in PT and exercise exertion as well as any adverse events that occurred during the session. Clinicians document and describe the HEP they prescribed to the adolescent to be completed between sessions.

The LEFS is a 20-item self-report survey that asks participants to report on their ability to perform everyday tasks using their lower extremities [44]. Participants rate the level of difficulty associated with a range of everyday activities on a 5-point Likert scale (0=“extreme difficulty/unable to perform activity” to 4=“no difficulty”). Summed scores indicate the level of function, with higher scores indicating more functionality.

The UEFI is a 20-item self-report survey that asks participants to report on their ability to perform everyday tasks using their upper extremities [45]. Participants rate the level of difficulty associated with a range of everyday activities on a 5-point Likert scale (0=“extreme difficulty/unable to perform activity” to 4=“no difficulty”). For both the LEFS and UEFI, scores are summed, with a maximum score of 80 and where higher scores indicate better functioning. A minimum level of detectable change (confidence level=90%) is defined as a ≥9-point change based on the existing literature on patients with pain [46-48].

The FOPQ-SF consists of 10 items, with each item rated on a 5-point Likert scale (0=“strongly disagree” to 4=“strongly agree”) [49]. The FOPQ-SF contains questions assessing both fear of pain and avoidance of activities in the context of pain. The total score is derived by summing the items, with higher scores indicating greater pain-related fear and avoidance of activities.

The PHODA-Youth is a 50-item measure assessing worry associated with activities of daily living (13 items), sports or exercise activities (15 items), school or social activities (13 items), and upper extremity activities (9 items) [50]. To complete each item, the patient is exposed to a photograph and label of the activity and asked to rate their worry “that this activity would be harmful to your pain” by dragging each photograph along a “worry thermometer” ranging from 0 to 10. Each photograph is given a rating according to its position on the thermometer. Patients then rate their anticipated pain if they engaged in the activity. The mean perceived harm and anticipated pain scores (ranging from 0 to 10) are calculated as the sum of each rating divided by the total number of pictures.

The TSK-17 is a 17-item self-report measure assessing fear of movement that has been implemented in a variety of pain conditions [51], including youth with chronic pain [52,53]. The 2 subscales assess activity avoidance and somatic focus, with higher total and subscale scores indicating a greater fear of movement, activity avoidance, and somatic focus.

Treatment satisfaction is assessed using an adapted version of the Pain Service Satisfaction Test. The Pain Service Satisfaction Test is a 22-item measure that asks patients about their experiences in pain treatment, for instance, perceptions of the effectiveness of the intervention, the treatment team, and the impact on their own outcomes [54]. Scores are summed based on responses, with higher scores indicating greater satisfaction with treatment.

To evaluate the acceptability of the treatment intervention, the VR acceptability questionnaire is an 11-item measure completed by patients and clinicians at discharge. For patients, it asks about their enjoyment of VR, satisfaction with the VR treatment, perceived reduction of pain, and barriers experienced during the physiorehabilitation with VR intervention. For clinicians, it assesses their perceived difficulty associated with VR use, assessment of the games, ability of VR to facilitate engagement, overall satisfaction levels, and willingness to implement VR in the future for those in the physiorehabilitation with VR arm.

To assess patient engagement in the physiorehabilitation with VR intervention, clinicians complete a postsession survey following every session, which includes the Pittsburgh Rehabilitation Participation Scale [55]. Clinicians rate the perceived patient’s motivation. Engagement is rated on a 6-point Likert scale (1=“none” to 6=“excellent”). Clinicians do not fill out a survey if the patient did not attend their session. Clinicians are instructed to select a lower rating when in doubt, for instance, “good” rather than “very good.” In addition, VR use is tracked via the ManageXR software (Mighty Immersion, Inc) preloaded onto each VR headset. Analytics regarding which VR games are played, duration of play per game, total play duration, and the number of application launches are displayed in the Oculus dashboard (Figure 2). The Oculus dashboard will be used to assess engagement in VR and fidelity to VR use for physiotherapy and HEPs.

Treatment expectations are measured using the child Treatment Expectancy and Credibility measure (TEC-C) [56]. The TEC-C comprises 6 items assessing expectations related to the effectiveness of the current treatment. The TEC-C is completed by the patient after the first treatment session.

Patient and clinician feedback regarding the VR intervention is collected via semistructured interviews. The goal of the interviews is to better understand the experience of using VR, strengths of the VR intervention, and barriers or considerations for future use.

Treatment fidelity is assessed by examining VR use analytics displayed on the ManageXR dashboard (Figure 2). For the physiorehabilitation with VR arm, clinicians aim for at least 8 minutes of the session to be dedicated to VR exercises and, therefore, the number of sessions reaching this benchmark can be tracked. In addition, at least 15 minutes of the HEP include VR in the physiorehabilitation with VR arm. HEPs across both treatment arms are requested in the clinician postsession survey and, for the physiorehabilitation with VR arm, we can compare with VR use tracked in the ManageXR software.

Adherence and retention are assessed by examining patient adherence to daily diaries, the percentage of patients who drop out before treatment completion, and the percentage of sessions completed on schedule.

The clinician tracks adverse events related to VR use. Specifically, in the postsession survey, clinicians can indicate any adverse events of dizziness, nausea, or disorientation that occurred. As part of clinician orientation to the VR, important safety precautions are discussed to reduce the potential for accidents to occur.

Functional disability is assessed using the FDI, a 15-item self-report measure of perceived difficulty in performing activities in the school, home, physical, and social contexts. Items are rated on a 5-point Likert scale (0=“no trouble” to 4=“impossible”) [57]. Items are summed to obtain a total score, with higher scores indicating greater disability. The FDI is widely used in pediatric pain research and is recommended as the gold-standard measure of physical functioning in school-age children and adolescents for clinical trials in pediatric chronic pain.

Mobility is assessed using the Patient-Reported Outcomes Measurement Information System (PROMIS) Mobility measure, a subset of the PROMIS physical health function self-report outcomes [58]. This test is typically used in adult and pediatric populations with chronic conditions. The Mobility subscale measures perceived capabilities related to mobility tasks such as getting up from a chair or running. The PROMIS Mobility items are written in the past tense (eg, “I could...”), all use a standard recall of “in the past 7 days,” and have a 5-point Likert scale (0=“not able to do” to 4=“with little to no trouble”). Scores are summed, with higher scores indicating greater mobility.

The PROMIS Pain Interference instrument assesses patient perception of the impact of pain on their ability to engage across several life domains—namely, social, emotional, and recreational activities [58]. The PROMIS Pain Interference items use a standard recall of “in the past 7 days” with items such as “it was hard to have fun when I had pain.” Responses are provided on a 5-point Likert scale (0=“almost never” to 4=“almost always”). PROMIS Pain Interference items are summed, with higher scores indicating more pain-related interference in a patient’s life.

The Pain Catastrophizing Scale-Children assesses negative cognitions associated with pain [59]. The Pain Catastrophizing Scale-Children comprises 13 items rated on a 5-point Likert scale (0=“not at all true” to 4=“very true”). A total score is obtained by summing the scores for all items. Higher scores indicate higher levels of catastrophic thinking.

Patients provide their pain rating when they complete their daily diary (Textbox 1) at the same scheduled time using a standard 11-point visual analog scale (0=“no pain” to 10=“most pain possible”) [60]. Average pain intensity ratings are calculated for 7 days before the first treatment session (baseline average pain) and 7 days before the discharge assessment (discharge average pain).

The Pain Self-Efficacy Scale-Children is a 7-item self-report questionnaire that measures patient beliefs about their ability to complete daily activities despite being in pain [61]. The Pain Self-Efficacy Scale-Children is scored on a 5-point Likert scale whereby patients rate their certainty about their ability to complete an activity (1=“very sure” to 5=“very unsure”), with higher scores indicating less self-efficacy in the context of pain.

The self-report measure Patients’ Global Impression of Change reflects a patient’s belief about the efficacy of treatment [62]. The Patients’ Global Impression of Change is a 7-point scale depicting a patient’s rating of overall improvement. Patients rate their change as “very much improved,” “much improved,” “minimally improved,” “no change,” “minimally worse,” “much worse,” or “very much worse.”

The child presence measure assesses the patients’ perceived involvement or immersion, realism, and transportation while using the VR headset [63,64]. The measure comprises 12 items rated on a 3-point Likert scale (0=“no” to 2=“a lot”). A total score is obtained by summing the scores for all items. Higher scores indicate higher levels of immersion and engagement.

The Modified Borg Dyspnea Scale is completed by the patient after each appointment [65]. The Modified Borg Dyspnea Scale measures patient rate of perceived exertion to monitor and guide exercise intensity. The scale item states the following—“How much difficulty is your breathing causing you right now?”—and was slightly modified given the deliverance to state “How much difficulty did the patient’s breathing cause them during today’s therapy session?” Responses are provided on a 12-point scale (0=“nothing at all,” 0.5=“very, very slight (just noticeable),” 5=“severe,” and 10=“maximal”).

To assess changes in physical ability and function, the 6-minute Walk Test [66], Single Leg Balance Test [67], and Closed Kinetic Chain Upper Extremity Stability Test [68] are performed at baseline and discharge. For the 6-minute Walk Test, patients walk back and forth in a straight line for 6 minutes, and the total distance covered is calculated. The Single Leg Balance Test consists of balancing on one leg at a time, and participants’ scores reflect the amount of time they are able to balance within a 60-second time frame. The Closed Kinetic Chain Upper Extremity Stability Test assesses participants’ upper body and core strength by asking them to start on their hands and knees or in a plank position and, while remaining in good form, continually tap the opposite hand that remains planted on the ground. The total score is the number of hand touches completed within the allotted time of 15 seconds. The Walk Test is recorded by a physiotherapist and research assistant. For the remaining 2 exercises, ViFive is used, a motion capture app that reads the patient’s body and allows for automatized counting and timing (Figure 3). The ViFive technology can track additional informatics such as range of motion, balance, flexibility, and endurance metrics as well as real-time pose correction to the user. For example, ViFive captures the patient’s body position during their Closed Kinetic Chain Upper Extremity Stability Test. Changes in performance from baseline to discharge across all 3 tests are assessed.

To examine daily physical activity, the participants wear an ActiGraph watch for the duration of their study participation (baseline phase and across treatment). The ActiLife software (ActiGraph, LLC) will be used to extract data and calculate the mean and peak daily activity. Physical activity (mean and peak) is modeled at 2 time points, and the rate of change in physical activity from baseline to discharge will also be examined.

Diaries on health care service use; personal costs; and support provided by family, friends, and professional carers are completed by parents once at baseline, on a weekly basis from pretreatment baseline to end of treatment, and once at the 3-month follow-up. Parents report on youths’ health care service use—general and specialist medical practitioners, physiotherapists, alternative health care practitioners, medications, hospital admissions, and out-of-pocket costs—and other impacts on youth and parental activity—athletic extracurricular activities and parental days off work and sick leave.

Variables related to chronic pain, including pain onset, duration, and intensity of pain symptoms as well as course and medications, are collected.

Demographic variables—namely, age, gender, sex, school grade, and ethnicity—are assessed via adolescent and parent reports at baseline.

Linear mixed effects models will be used to compare physiorehabilitation with VR with standard PT across all non-SCED outcomes. We will model our outcomes at 3 time points using a mixed effects linear model with fixed effects for treatment assignment, period, interaction between treatment and period, and baseline covariates, and a random effect for individual. The random effect will allow us to account for the correlation in the outcome within an individual over time.

To examine biomechanical data, physical assessment metrics will be extracted. We will model physical assessment metrics of walk distance, single leg balance duration, and hand taps using mixed 2 (time)×2 (group) ANOVAs. If physical assessment metrics differ by pain site (upper, trunk, lower, or diffuse), this will be included as a covariate. To examine actigraphy data, the ActiLife software will be used to extract data and calculate the mean and peak daily activity. Published data reduction methods will be used [69]. We will model physical activity (mean and peak) at 2 time points using mixed 2 (time)×2 (group) ANOVAs. The rate of change in physical activity from baseline to discharge will also be examined using the randomization tests used for SCED outcomes described in the following sections. To examine health cost diary data, we will model health care cost variables using t tests and linear and mixed regression models.

Mean satisfaction and acceptability scores will be examined for both patients and clinicians. To assess patient engagement, the mean adolescent daily diary completion, percentage of patient dropouts before treatment completion, and percentage of sessions with benchmark VR met will also be examined. To assess patient and clinician feedback regarding the feasibility and acceptability of the VR treatment, thematic analysis will be conducted of semistructured interviews to identify barriers, catalysts, and perceptions of integrating VR into clinical care. To evaluate engagement in the VR treatment, the mean clinician ratings on the Pittsburgh Rehabilitation Participation Scale will be examined. Acceptability of the VR treatment will be assessed using patient-reported treatment expectancy mean scores. Finally, adherence to the suggested benchmark VR engagement will be assessed by examining the mean time engaged in VR across participations (standard PT and physiorehabilitation with VR) as well as across VR game type (eg, Fruity Feet vs Vacation Simulator).

The data obtained from the randomized SCED used in this study have a hierarchical 2-level structure with observations (level 1) nested within patients (level 2). This nested structure induces dependency within the data—observations vary not only because of random sampling within a patient but also between different patients. For data analysis, we will use a hierarchical linear model, allowing us to combine all patients’ data into a single multilevel model while also considering both the within- and between-patient dependencies. The within- and between-patient variability, as well as the overall effects of the treatment across patients, will be modeled. For conducting the multilevel analysis and obtaining inference results in R (R Foundation for Statistical Computing), MultiSCED will be used [70]. These daily individual data also allow for the use of randomization tests to assess the difference in daily diaries between baseline and discharge and between baseline and 3-month follow-up.