Large Language ModelLLM-audiobookNCT07637617
Evaluating the Acceptability of AI-Personalized Virtual Reality Picture Books for Hospitalized Pediatric Patients
A study testing LLM-audiobook in people with Large Language Model, run by Stanford University.
Trial summary
Hospitalization strips pediatric patients of the environments, objects, and people that shape their daily lives. Hospitalized pediatric patients routinely experience painful procedures, psychological distress, boredom, and a disorienting loss of personal identity. Read more about this study
- Who is it for?
- Ages 5 to 17 years
- Conditions the study is for
- Large Language Model
- What is being tested?
- LLM-audiobook
- Type of treatment
- Lifestyle or therapy programme
- Could I receive a placebo?
- No, everyone receives an active treatment
- How many people
- 15 participants
- Where
- No sites listed yet
- Study dates
- 1 Oct 2026to 30 Sep 2027
- Phase
- Interventional
- Are there results?
- Not yet: the study is under way
Key requirements
Who can take part, as listed by the sponsor. Only the study team can confirm whether you are eligible, so take this list to your doctor.
- Ages 5 to 17 years
- All sexes
- No healthy volunteers
- Child
Who may be able to join
- Between age 5-17
- Current admission in inpatient unit
- Able to understand and interact with a virtual reality audiobook in English
Who cannot join
- Legal guardian not present to obtain consent
- Significant cognitive impairment and/or developmental delay
- History of seizures
- History of severe motion sickness
- Severe visual or motor impairment limiting picture book participation
- Child with active infection of the face or hand
- Acute medical instability
- Inability to understand English
Trial locations and contacts
No sites are listed yet. Contact the study team or a site near you to ask about taking part.
About this study
Hospitalization strips pediatric patients of the environments, objects, and people that shape their daily lives. Hospitalized pediatric patients routinely experience painful procedures, psychological distress, boredom, and a disorienting loss of personal identity. These experiences measurably worsen anxiety, reduce cooperation with care, and diminish the quality of the inpatient experience for both patients and families. Immersive digital interventions, including VR and tablet-based experiences, have emerged as a promising class of tools for addressing these challenges. Prior studies from The Stanford Chariot Program have demonstrated that digitally delivered, patient-centered experiences can meaningfully reduce procedural anxiety and improve engagement in hospitalized children.
Yet, an important limitation persists in these technologies - current digital interventions largely remain in one-size-fits-all formats. Every child receives the same content, regardless of who they are, what they love, or what makes them feel at home in the world. This design limits therapeutic relevance, constrains engagement, and represents a missed opportunity to engage children, reduce anxiety, and enhance their quality of life during hospital stays.
- Conditions
- Large Language Model
- Being tested
- LLM-audiobook
- Keywords
- feasibility, acceptability, usability
What participants receive
What participants receive. In a randomised study each person is assigned to one group by chance.
LLM-audiobookBehavioural
Develop and refine a two-stage personalization LLM pipeline: (1) a structured brief interview template and (2) an LLM prompt chain for content generation. The interview instrument will be a structured guide (~20 minutes) capturing details about the child's background, interests, and aspirations. All information collected during the interview will be deidentified prior to being passed into the LLM.
Content Generation: Deidentified interview responses will be input to a standardized LLM prompt pipeline. The pipeline generates a structured patient profile that interprets the participant's interests, personality, and preferences, then uses that profile to produce a personalized narrative, formatted as a picture book story. AI voice synthesis will generate an audio narration of the story. AI image generation tools will produce a set of 6-10 accompanying illustrations. The final product - a synchronized audio picture book - will be delivered via VR headset at bedside.
| Study group | Role | What they receive |
|---|---|---|
| Pediatric patient | Experimental | Not described Receives: Behavioral: LLM-audiobook |
What is being measured
The primary outcome is the main question the study is designed to answer.
Determine the acceptability of the Large Language Model generated audiobook
Semi-structured focus group interviews conducted in person immediately after intervention. Six core questions plus probing prompts assessing attitudes and opinions, and perceptions of the intervention experience
Measured: Immediately after intervention
Secondary outcomes (3)
To assess the degree to which participants identified with the protagonist of the personalized audio picture book.
Participants will complete the Cohen Identification Scale following the VR experience.
Items are answered using a 5-point Likert scale ranging from 1 = strongly disagree to 5 = strongly agree. Item responses will be averaged to generate a composite score, with higher scores indicating greater identification with the main character.
Measured: Immediately after intervention
To assess emotional state using Positive and Negative Affect Schedule - Child Form (PANAS-C)
The PANAS-C measures the degree to which children experience positive and negative affective states, yielding two subscale scores: Positive Affect and Negative Affect. Items are answered using a 5-point Likert scale ranging from 1 = very slightly or not at all to 5 = extremely. Higher Positive Affect (PA) Score indicates a higher level of positive emotions while higher Negative Affect (NA) Score indicates a higher level of negative emotions.
Measured: baseline, Immediately after intervention
To assess subjective wellbeing using the World Health Organization-Five Child Well-Being Index (WHO-5)
The WHO-5 is a brief, validated self-report measure assessing current mental wellbeing across five positively worded items covering mood, relaxation, energy, rest, and daily interest. The WHO-5 instructs respondents to rate how they have felt "over the last two weeks,"; the scale will be modified in this study to ask the participants to reflect on their current state. Items are answered using a 5-point Likert scale ranging from 0 = never to 5 = all of the time. To calculate the overall wellbeing score, the responses to the five items are summed and the total is multiplied by 5, yielding a final score between 0 and 100. Higher scores reflect higher levels of wellbeing.
Measured: baseline, Immediately after intervention
Timeline
From the start of the study to its completion; the mark is today. Estimated dates are the sponsor’s plan.
- First posted10 Jun 2026
- Study start1 Oct 2026 (estimated)
- Primary completion30 Sep 2027 (estimated)
- Study completion30 Sep 2027 (estimated)
Resources
- Official record on ClinicalTrials.govThe registry entry this page is built from, with its full change history.
- Other trials for Large Language ModelStudies of the same condition, open and under way.
- More trials run by Stanford UniversityThe sponsor's other registered studies.
- What the trial phases meanHow a treatment moves from first tests in people to approval.
More details
For doctors and researchers: how the study is designed, who runs it, and its registry IDs.
Study design
- Study type
- Interventional
- Phase
- Phase N/A
- Primary purpose
- Supportive care
- Allocation
- Not applicable
- Intervention model
- Single group
- Masking
- None (open label)
- Enrolment
- 15 participants (estimated)
Sponsor and investigators
- Lead sponsor
- Stanford University
- Sponsor type
- University, hospital or other
- Responsible party
- Principal investigator, Thomas Caruso, Stanford University
- Investigators
- Samuel Rodriguez, MD, Study director, Stanford UniversityThomas Caruso, MD,PhD, Principal investigator, Stanford University
IDs, dates and publications
- ClinicalTrials.gov ID
- NCT07637617
- Sponsor's protocol ID
- 87422
- Acronym
- LLM-audiobook
- Status last verified
- Sep 2026
- First submitted
- 3 Jun 2026
- Last updated
- 5 Oct 2026
- Results posted
- No
- Data monitoring committee
- No
- FDA-regulated drug
- No
- FDA-regulated device
- No
- Individual data sharing
- No
Study at a glance
From the registry record, last updated by the sponsor on 5 Oct 2026. Recruitment status is as the sponsor reported it: confirm with the study team.
- Official title
- Evaluating the Acceptability of AI-Personalized Virtual Reality Picture Books for Hospitalized Pediatric Patients
- ClinicalTrials.gov ID
- NCT07637617
- Acronym / sponsor ID
- LLM-audiobook / 87422
- Conditions
- Large Language Model
- Keywords
- feasibility, acceptability, usability
- Study type
- Interventional
- Recruitment status
- Not yet recruiting
- Enrolment
- 15 participants (estimated)
- Lead sponsor
- Stanford UniversityUniversity, hospital or other
- Study start
- 1 Oct 2026 (estimated)
- Primary completion
- 30 Sep 2027 (estimated)
- Study completion
- 30 Sep 2027 (estimated)
- First posted
- 10 Jun 2026 (actual)
- Results
- Not posted in the source record
Clinical summary
The research question as the sponsor registered it. This describes what the study plans to test, not what it found.
Hospitalization strips pediatric patients of the environments, objects, and people that shape their daily lives. Hospitalized pediatric patients routinely experience painful procedures, psychological distress, boredom, and a disorienting loss of personal identity. These experiences measurably worsen anxiety, reduce cooperation with care, and diminish the quality of the inpatient experience for both patients and families. Immersive digital interventions, including VR and tablet-based experiences, have emerged as a promising class of tools for addressing these challenges. Prior studies from The Stanford Chariot Program have demonstrated that digitally delivered, patient-centered experiences can meaningfully reduce procedural anxiety and improve engagement in hospitalized children.
Yet, an important limitation persists in these technologies - current digital interventions largely remain in one-size-fits-all formats. Every child receives the same content, regardless of who they are, what they love, or what makes them feel at home in the world. This design limits therapeutic relevance, constrains engagement, and represents a missed opportunity to engage children, reduce anxiety, and enhance their quality of life during hospital stays.
Study design
- Primary purpose
- Supportive care
- Allocation
- Not applicable
- Intervention model
- Single group
- Masking
- None (open label)
- Number of arms
- 1
| Arm | Type | Receives |
|---|---|---|
| Pediatric patient | Experimental | Behavioral: LLM-audiobook |
Who may qualify
- Ages
- 5 to 17 years
- Sex
- All
- Healthy volunteers
- Not accepted
- Age groups
- Child
Key inclusion criteria (3 listed)
- Between age 5-17
- Current admission in inpatient unit
- Able to understand and interact with a virtual reality audiobook in English
Key exclusion criteria (8 listed)
- Legal guardian not present to obtain consent
- Significant cognitive impairment and/or developmental delay
- History of seizures
- History of severe motion sickness
- Severe visual or motor impairment limiting picture book participation
Eligibility is decided by the study team after screening; this overview does not establish that a patient qualifies. Full inclusion and exclusion criteria.
Interventions
| Intervention | Type | Arms |
|---|---|---|
| LLM-audiobookDevelop and refine a two-stage personalization LLM pipeline: (1) a structured brief interview template and (2) an LLM prompt chain for content generation. The interview instrument will be a structured guide (~20 minutes) capturing details about the child's background, interests, and aspirations. All information collected during the interview will be deidentified prior to being passed into the LLM. Content Generation: Deidentified interview responses will be input to a standardized LLM prompt pipeline. The pipeline generates a structured patient profile that interprets the participant's interests, personality, and preferences, then uses that profile to produce a personalized narrative, formatted as a picture book story. AI voice synthesis will generate an audio narration of the story. AI image generation tools will produce a set of 6-10 accompanying illustrations. The final product - a synchronized audio picture book - will be delivered via VR headset at bedside. | Behavioural | Pediatric patient |
Primary outcome
Determine the acceptability of the Large Language Model generated audiobook
Time frame: Immediately after intervention
Semi-structured focus group interviews conducted in person immediately after intervention. Six core questions plus probing prompts assessing attitudes and opinions, and perceptions of the intervention experience
3 secondary outcomes
To assess the degree to which participants identified with the protagonist of the personalized audio picture book.
Time frame: Immediately after intervention
To assess emotional state using Positive and Negative Affect Schedule - Child Form (PANAS-C)
Time frame: baseline, Immediately after intervention
To assess subjective wellbeing using the World Health Organization-Five Child Well-Being Index (WHO-5)
Time frame: baseline, Immediately after intervention
Results and safety
This study is still under way. Results are not available in the source record yet.
Research links
- Official record on ClinicalTrials.govLast updated by the sponsor 5 Oct 2026.
- Individual participant dataSharing planned: No
Related trials for Large Language Model
RecruitingObservational study
Evaluation of Large Language Models in Parasternal Block Ultrasound Interpretation
Parasternal Block
Not yet recruitingObservational study
Automated Vetting and Protocolling of MRI Examinations
Analysis of Retrospectively Collected MRI Referral Data
RecruitingObservational study
AI-generated Treatment Recommendations in Precision Oncology
Cancer
Active, not recruiting
Conversational AI for Adult Flu Vaccination Decisions
Influenza, Human, Vaccine Hesitancy
Not yet recruiting
Tailored Brain Health Insights Through Nurturing Knowledge-based Change - Stakeholder-driven, Human-centered, AI-powered Refinement for Prevention
MCI, Dementia
Referral and contact
AzTrials does not enrol participants or take referrals. Contact the study team or a site listed below; they decide who can take part.
Samuel Rodriguez, MD study director
Stanford University
Thomas Caruso, MD,PhD principal investigator
Stanford University
Questions about this trial
What is this clinical trial studying?
A study testing LLM-audiobook in people with Large Language Model, run by Stanford University.
The full registered description is in the About this study section and the official record.
Who may be eligible to take part?
The sponsor lists 3 inclusion and 8 exclusion criteria. Eligibility is decided by the study team after screening, so read the full criteria and contact a listed site.
Is this clinical trial currently recruiting?
The record says: opening soon, as of the sponsor's last update on 5 Oct 2026. Status can change between updates, so confirm with the study contact before making plans.
What are the possible benefits and risks?
They depend on the study and the treatment, and no study can promise a benefit. The study team explains the known risks during informed consent, before anyone joins; the record itself does not list them.
Are results available for this study?
This study is still under way. Results are not available in the source record yet.
How can a doctor refer a patient or contact the research team?
Contact Samuel Rodriguez, MD (650-723-5728, sr1@stanford.edu), or a site near the patient. AzTrials does not enrol participants or take referrals.
What does the primary outcome measure?
Determine the acceptability of the Large Language Model generated audiobook (time frame: Immediately after intervention). It is what the study was designed to measure, not a finding.
Where can I read the original study information?
Does a listing mean the treatment is approved or proven to work?
No. A listing means the study is registered. It does not show that the treatment is safe, works, or is approved by a regulator.
How often is this information updated?
AzTrials copies changed records from ClinicalTrials.gov every night and reloads the whole registry every week. The sponsor last updated this record on 5 Oct 2026.