Randomised controlled trial of VR-based observation meditation with AI coaching (‘Otti’) for stress reduction in university students in the United States: study protocol

Abstract

Introduction

Stress is a major health issue in contemporary society, and mindfulness-based approaches reduce stress and anxiety but face practical barriers to consistent practice; this protocol evaluates a Virtual Reality (VR)-based observation meditation programme with an artificial intelligence (AI) coach (‘Otti’) that delivers real-time empathic, tailored prompts to support present-focused attention and emotion regulation in university students in the United States. A single-centre randomised controlled trial in Pennsylvania will assess immediate psychophysiological effects and user acceptability after a single 15 min session following a standardised Stroop stressor in a university laboratory setting.

Methods and analysis

An a priori power analysis (f=0.25, α=0.05, power=0.80) supports recruitment of 34 students (n=17 per group) in a single-centre randomised controlled design comparing AI-coached VR observation meditation to a no-treatment leisure control within a 30 min visit. Participants complete pre-intervention surveys Perceived Stress Scale-10 (PSS-10), Depression Anxiety Stress Scales (DASS-21), State–Trait Anxiety Inventory (STAI-State, STAI-Trait) and baseline heart rate/Heart Rate Variability (HRV) via smartwatch, undergo the 15 min intervention or control, then complete postintervention surveys and repeated heart rate/HRV recording; effects will be tested using repeated-measures analysis of variance, with heart-rate data exported and preprocessed per the prespecified plan. Primary outcomes include perceived stress (PSS-10), emotional state (DASS-21, STAI-State, STAI-Trait), physiological stress response (heart rate/HRV) and participant satisfaction via a structured postintervention survey (usability, perceived effectiveness, comfort).

Ethics and dissemination

The study received IRB approval from The Pennsylvania State University Institutional Review Board (PSU CATS IRB: STUDY00025978; ClinicalTrials.gov: NCT06704282), and all participants will provide written informed consent prior to procedures. Findings will be disseminated via open access publication, conference presentations and stakeholder-focused briefs, with an anonymised primary-outcome dataset available on reasonable request in line with BMJ Open policies and Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT)/International Committee of Medical Journal Editors (ICMJE) guidance.

Trial registration number

NCT06704282.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• A randomised, single-centre design compares VR+AI coaching with a no-treatment control, using standardised stress induction via the Stroop Colour–Word Task and conducting pre–post assessments within a single 30 min session.

• Multimodal measurement using validated scales (DASS-21, PSS-10, STAI-State and STAI-Trait), participant satisfaction/acceptability and physiological indices (heart rate and HRV).

• Prespecified analysis (repeated-measures analysis of variance) with complementary usability assessment.

• Convenience sampling of one-site university students and small sample size limit generalisability.

Introduction

Stress is a pervasive health issue in modern society, adversely affecting both mental and physical health.¹,³ Chronic and sustained stress contributes to a myriad of health complications, including depression, anxiety and cardiovascular diseases, significantly impairing individuals' quality of life.⁴,⁶ Stress negatively influences cognitive, emotional and social functioning, diminishing work performance, interpersonal relationships and, over time, escalating into severe psychological and physiological problems.^{7 8} Therefore, effective stress management is crucial for maintaining individual well-being and quality of life.⁹ In addition, numerous studies underscore that effective stress management can help reduce societal healthcare costs and enhance public health outcomes.¹⁰,¹²

Mindfulness has been established as an effective approach to stress management, with extensive evidence supporting its efficacy.¹³,¹⁵ Mindfulness involves an individual cultivating a non-judgemental awareness of the present moment.¹⁶ Engagement in mindfulness can play a critical role in mitigating stress and anxiety as well as improving emotional well-being. For example, research on the Mindfulness-Based Stress Reduction (MBSR) programme, developed by Kabat-Zinn, clearly demonstrates the programmes ability to significantly reduce stress and anxiety and promote emotional well-being.¹⁷ MBSR fosters individuals' awareness of their circumstances, allowing them to respond adaptively to stressors while also offering physiological benefits, such as improved immune function and reduced blood pressure.¹⁸,²⁰ It has been shown that even short courses of mindfulness training can produce measurable shifts in brain activity toward more approach-related, positive affective states and exert beneficial biological effects.²¹ However, despite these advantages, maintaining consistent mindfulness practice remains challenging for many, especially given the demands of daily life and high levels of stress. Consequently, challenges like these can limit the broader impact of mindfulness-based interventions.

Virtual reality (VR) technology offers promising solutions to the challenges of traditional mindfulness practices.²² Even if a conducive environment for meditation is not available or there are distracting elements in the real world, VR can create an immersive environment that allows for focused meditation anytime and anywhere.²³ VR-based meditation employs visual and auditory stimuli to generate a fully immersive experience, thereby enhancing users' ability to maintain concentration. This VR environment contributes to maximising the effectiveness of psychological interventions.²⁴,²⁶ Prior studies have shown that VR-based psychological therapies can effectively address various mental health issues, such as Post-Traumatic Stress Disorder (PTSD), anxiety disorders and phobias, with more research needed to better understand the potential impact of using immersive and personalised material within VR settings to enhance psychological treatment efficacy.²⁷

Contemporary society is facing a shortage of professional mental health therapists, leading to increasingly long wait times for treatment and causing many to miss the golden window for emotional stabilisation.²⁸,³⁰ Additionally, social stigma, economic challenges and the limited availability of mental health centres prevent many individuals from prioritising their emotional well-being.^{31 32} This highlights the need for new forms of emotional support that are easily accessible and pose minimal barriers for users.

This study leverages VR technology to investigate the effects of a novel observation meditation programme, that combines VR with AI-driven, personalised coaching, on stress reduction. The AI coach ‘Otti’ assesses participants' emotional states in real-time and provides individualised questions and empathetic responses, thereby tailoring the meditation experience. By offering real-time feedback based on each participant’s specific emotional states, the AI coaching system helps participants immerse themselves more fully in the meditation process, explore and accept their emotions, and overcome obstacles that may arise, with the stated aim of leading to more effective and sustained mindfulness practice.³³ This programme allows individuals to receive appropriate meditation guidance and emotional empathy in real-time, even without face-to-face counselling with a therapist. It serves as an effective psychological intervention that reduces spatial and temporal constraints and minimises the burden on users, laying the groundwork for providing psychological support to a wider audience.^{34 35} Furthermore, by offering a structured and brief observation meditation tailored to contemporary society, it explores new possibilities for the practice of mindfulness in a simple and efficient manner, while aligning with evidence that mindfulness training—over both short and extended periods—can attenuate amygdala reactivity to emotional stimuli, suggesting a neural pathway through which such practice may rapidly and cumulatively reduce stress reactivity.³⁶

At its core, this study asks whether a single 15 min VR-based observation meditation session augmented with AI coaching (Otti) produces greater immediate reductions in state stress and physiological arousal than a control condition, and how acceptable and usable this intervention is in a university student sample. To operationalise this enquiry, we examine four questions: (1) does AI-coached VR observation meditation meaningfully enhance students’ capacity for stress regulation? (2) following Stroop-induced stress, does a single session yield immediate pre-to-post improvements in state stress (STAI) and cardiac indices (heart rate/HRV)? (3) do short-term improvements emerge in depressive and anxiety symptoms as measured by the Depression Anxiety Stress Scales (DASS-21) Depression/Anxiety subscales and STAI (STAI-State, STAI-Trait)? and (4) do participants report high satisfaction and intention to continue use, and are these acceptability indicators associated with the magnitude of stress/anxiety improvement? In line with best practice for protocol introductions, the aim is to evaluate the intervention’s immediate psychophysiological effects and user acceptability; the primary hypothesis is that the session will significantly reduce state stress and anxiety by the end of the visit, with secondary hypotheses that it will reduce depressive and anxiety symptoms, decrease perceived stress, stabilise heart rate under an induced stressor and elicit participant preference for the AI-coached VR programme.

A randomised controlled trial will be utilised to quantitatively and qualitatively evaluate the impact of combining a VR-based meditation programme with AI coaching on stress management. Key evaluation metrics include physiological indices (heart rate and HRV) measured via smartwatch; participant satisfaction/acceptability with the intervention and validated instruments including the DASS-21,^{37 38} the PSS-10^{39 40} and the State–Trait Anxiety Inventory (STAI-State and STAI-Trait).^{41 42 43} The findings of this study will help to illuminate the clinical utility of using a VR and AI-based observation meditation programme as a tool for promoting mental well-being and advancing strategies for stress management and mental health improvement.

Methods and analysis

Trial design and flowchart

This study is a randomised controlled trial designed to evaluate the impact of the VR meditation programme combined with AI coaching on stress reduction. The efficacy of the VR-based meditation programme will be assessed by comparing two groups: The experimental group will participate in the VR meditation coaching programme, while the control group will not be asked to engage in or be provided with specific activities for stress management. The study environment is designed to remain unbiased, allowing control group participants to have free and relaxing time. The study is conducted as a single-centre trial, with all procedures completed within 30 min. This study will be conducted from 10 December 2024 to 31 December 2025.

Participants

This study will recruit US university students through voluntary participation, and all participants will provide written informed consent before the study begins. A G-Power analysis indicated that a total sample of 34 is required under the assumptions of effect size f=0.25, significance level (α)=0.05 and statistical power=0.80; accordingly, 17 participants will be assigned to each group to ensure adequate power.⁴⁴ The medium effect size (f=0.25) was selected conservatively in light of meta-analytic findings from similar digital mindfulness interventions and the novel nature of the AI-coached VR programme, with the aim of detecting a clinically meaningful effect.

Inclusion criteria are as follows: adults aged 18 years or older who are currently enrolled university students recruitable via the Behrend Psychology Research Participation SONA Pool, who can understand the study description, provide written informed consent and complete a single 30 min laboratory session. Exclusion criteria are as follows: individuals who self-report a history of severe motion sickness or prior adverse reactions to VR that could compromise safe participation; individuals with current health conditions that, in the investigator’s judgement, would substantially interfere with safe VR use or adherence to study procedures (eg, acute vestibular disorders) and individuals who are unable to provide written consent or comply with study procedures. An example consent form is provided in online supplemental material.

Procedures

This study will follow the procedures detailed below and has received IRB approval from Pennsylvania State University, where the first author’s research is being conducted. The entire experimental procedure is presented in figure 1.

Figure 1. Experiment flowchart flow diagram outlining the randomised controlled trial process: participant recruitment, informed consent, random assignment, stress induction using the Stroop Test, pre-intervention surveys and baseline heart rate measurement, intervention phase (VR-based observation meditation with AI coaching for the experimental group, leisure time for the control group), and post-intervention assessments including surveys, heart rate monitoring and programme satisfaction evaluation. DASS-21, Depression Anxiety Stress Scales; PSS-10, Perceived Stress Scale-10; STAI, State–Trait Anxiety Inventory; VR, virtual reality.

1. Consent and explanation: Participants will receive comprehensive information regarding the study and provide written informed consent. Participants will then be randomly assigned to the intervention condition or no-treatment control condition. Randomisation will be conducted using a computer-generated random number sequence to create a 1:1 allocation. A research assistant, independent of the intervention process, will prepare sealed, opaque envelopes containing the group assignment. These envelopes will be opened sequentially by the primary researcher only after a participant has provided consent and completed all baseline assessments.

2. Induction of stressful situation: Participants will complete the Stroop Colour–Word Test to induce a moderate level of stress.

3. Pre-intervention surveys and heart rate measurement: Participants will complete stress-related surveys (PSS-10, DASS-21, STAI-State and STAI-Trait) before the intervention, and baseline heart rate and HRV will be measured via smartwatch.

4. Intervention (15 min): The experimental group will engage in the AI-coached VR observation mindfulness programme, while the control group will have leisure time (eg, quiet seated computer use) under comparable environmental conditions.

5. Postintervention surveys and heart rate measurement: Following the intervention, participants will complete postintervention surveys (PSS-10, DASS-21, STAI-State and STAI-Trait) and a participant satisfaction/acceptability survey specific to the VR programme; heart rate and HRV will be recorded again via smartwatch.

Outcomes and measures

The primary outcomes of the study are stress reduction and improvement in emotional well-being, evaluated using DASS-21, PSS-10, STAI (STAI-State, STAI-Trait), and physiological data collected via smartwatch (Fitbit Charge 3), including heart rate and HRV. The DASS-21 is a 21-item self-report instrument that measures depression, anxiety and stress, with seven items dedicated to each subscale.^{38 45} This tool is essential for comprehensively assessing stress and emotional well-being as a primary outcome in this trial, providing quantitative insights into participants’ psychological states at both pre- and postintervention timepoints. The PSS-10 is a 10-item questionnaire that measures perceived stress levels and will be administered pre- and postintervention to capture subjective stress experiences and short-term change as part of the primary outcome battery.⁴⁵ The STAI (STAI-State, STAI-Trait) is a self-report questionnaire designed to measure an individual’s state and trait anxiety. By using these psychological assessment tools alongside physiological data (heart rate and HRV) from smartwatches, the study uses multiple modalities to evaluate the programme’s effectiveness.⁴⁶,⁴⁸ Participant satisfaction is a primary outcome assessed immediately postintervention via a structured survey (including usability, perceived effectiveness and comfort).

AI coaching: the role of Otti

The AI coach ‘Otti’ serves as a pivotal component of the study, functioning as a virtual coach within the VR-based intervention. Otti guides participants by providing personalised questions, reflecting on their emotional and cognitive states, and fostering self-insight. During the mindfulness training, Otti helps participants remain focused on the present moment while recognising and addressing automatic, stress-inducing thoughts. Furthermore, Otti offers positive feedback and motivational support to help participants overcome difficulties they may encounter during the training.

As an AI-driven chatbot, Otti employs several techniques to offer a personalised experience. First, Otti uses emotion analysis to assess participants' emotional states in real-time. Using the twitter-roberta-base-sentiment model, Otti classifies participant responses as ‘positive’, ‘neutral’ or ‘negative’ and provides tailored feedback.⁴⁹ This interaction helps participants feel that their emotions are acknowledged and understood, thereby enhancing their sense of immersion.

Second, Otti presents a range of questions based on designed prompts, which aim to encourage participants to delve deeper into their experiences during meditation. These prompts encompass various elements of observation, including colour, light, shape, texture, composition, space, distance, movement and sound, enabling participants to explore the subject from visual, tactile and emotional perspectives.⁵⁰ These questions facilitate participants' efforts to break away from automatic negative thinking and approach observation with greater openness.

Third, Otti incorporates Motivational Interviewing (MI) techniques to encourage participants' intrinsic motivation for change. MI is a counselling approach that helps individuals identify their goals and values and fosters behaviour change.⁵¹ Otti uses open-ended questions and reflective listening to support participants in exploring their experiences in depth. For instance, when a participant expresses negative emotions, Otti reflects those emotions and encourages the participant to consider the underlying reasons, fostering acceptance and understanding as a basis for positive change.⁵²

Fourth, Otti implements a question duplication prevention algorithm to ensure participants continually receive new prompts, thus maintaining a natural flow of non-repetitive interactions. Questions are selected by randomly mixing categories, ensuring that previously used questions are not repeated, thereby offering participants new opportunities for reflection and contributing to the programme’s effectiveness.

Lastly, Otti provides emotionally empathetic responses based on emotion analysis results. If negative emotions are detected, Otti expresses emotional support with statements such as, ‘I understand that it feels challenging right now. I'm here with you’. Conversely, if positive emotions are identified, Otti reinforces the participant’s positive state, with sayings, such as, ‘That’s wonderful! I can sense your positive energy’. This type of empathetic feedback fosters participants' sense of emotional security and encourages deeper engagement in the meditation process.^{53 54}

Otti supports participants in moving away from negative thinking patterns and developing positive or neutral alternative thoughts. Through reflective questioning, Otti assists participants in better understanding their cognitive responses to stress and developing adaptive coping strategies. Additionally, through group interaction prompts, Otti facilitates cognitive restructuring, enabling participants to adopt more positive and flexible attitudes when confronted with stress. Otti and the overall system structure are illustrated in figure 2.

Figure 2. ‘Otti’ programme structure schematic representation of the VR-based observation meditation programme guided by the AI coach ‘Otti’. The figure illustrates the programme’s main components, including emotion recognition, tailored questioning, motivational interviewing, reflective feedback, empathy-based responses and observation prompts within a Snow Globe VR environment. ACT, Acceptance and Commitment Therapy; VR, virtual reality.

Mindfulness observation meditation in the VR programme

The mindfulness observation meditation VR programme uses a ‘Snow Globe’ tool to facilitate observation meditation. The Snow Globe contains natural environments and specific objects, creating a calming and visually appealing setting. The Otti system was developed with Unreal Engine 5 and will be delivered to participants via Meta Quest 3 VR headsets. Participants are instructed to observe the natural scenery and objects within the Snow Globe, paying attention to colours, textures and interactions among elements to remain in the present moment. This process trains participants to observe and be present-focused, without judgement, helping them detach from automatic, stress-inducing thoughts.⁵⁵

During the observation meditation process, Otti engages with participants by asking them to describe what they see and feel. This practice helps participants develop mindfulness skills and strengthens their capacity to separate emotional responses from cognitive thoughts. By learning to disengage with worries about the future or ruminations about the past and instead focusing on the present moment, participants learn to associate the present-focused mindset with less negative emotions which can lead to a reduction of stress levels.⁵⁶

This meditation practice aligns with the concept of ‘defusion’ as described in Acceptance and Commitment Therapy, focusing on fostering participants' ability to distance themselves from and accept negative, stress-triggering thoughts.⁵⁷ By cultivating the habit of mindful observation, participants can disrupt negative thought patterns and enhance overall emotional well-being.⁵⁸ Moreover, to help establish this, time is intentionally not explicitly displayed, within the VR programme. Instead, participants experience the natural passage of time through the rising and setting of the sun within the VR environment. This design allows participants to complete their meditation session without the pressure of specific time constraints, ensuring they remain comfortably immersed in the training.

Data analysis

Data analysis will involve comparing quantitative differences between the two groups using pre- and postintervention measurements. Repeated measures analysis of variance and other statistical techniques will be employed to assess the significance of the results and validate differences between the groups regarding primary outcomes.⁵⁹

Quantitative analyses will be conducted using IBM SPSS (eg, pre–post comparisons, between-group tests and Analysis of Covariance(ANCOVA) as appropriate).

In addition to the primary analyses, a structured postintervention satisfaction/usability survey will be administered to the intervention group. The survey will include 7-point Likert-scale items (1=strongly disagree, 7=strongly agree). The survey categories and example items are summarised in table 1.

Table 1. Satisfaction/usability survey categories and example items.

Category	Example item
Overall satisfaction	‘’Overall, I am satisfied with this VR meditation programme’.
Perceived usefulness	‘I feel that my stress decreased after the session’.
Ease of use/learnability	‘It was easy to understand how to use the programme’.
Immersion/presence	‘The VR environment felt realistic (sense of presence)’.
Comfort/cybersickness	‘I felt generally comfortable during the session’.
AI coach (Otti) empathy/helpfulness	‘Otti’s feedback made me feel understood’.
Content quality (observation prompts)	‘The prompts helped my observation practice’.
Emotional impact/mindfulness skills	‘The session contributed to my emotional stability’.
Intention to reuse/recommend	‘I would like to use this programme regularly in the future’.

An open-ended question will also elicit qualitative feedback: ‘Please describe what was most helpful, what could be improved and your impressions of the interaction’. This item is designed to capture unstructured user experience and contextual nuance that may not be reflected in scale-based responses, enabling richer interpretation of acceptability and usability outcomes.

The brief analysis plan for satisfaction/usability proceeds in two complementary streams. Quantitatively, analyses will report item- and subscale-level descriptive statistics (mean, SD, 95% CI), evaluate internal consistency for any multi-item subscales using Cronbach’s alpha,⁶⁰ examine correlations between satisfaction/usability scores and pre–post changes in stress-related outcomes (for example, ΔPSS-10, ΔDASS-21 Stress) and run exploratory regression models that adjust for baseline covariates to assess whether satisfaction predicts outcome change. Qualitatively, open-ended responses will undergo inductive coding and thematic analysis by two independent coders with consensus-based resolution of discrepancies; inter-rater agreement will be reported (eg, Cohen’s kappa), and qualitative themes will be triangulated with quantitative satisfaction scores to enrich interpretation and strengthen trustworthiness of inferences.⁶¹

Heart-rate data will be extracted by exporting Comma-Separated Values (CSV) files from the Fitbit web dashboard, with additional automated exports through the Fitbit web Application Programming Interface (API) when necessary; the exported files will then be preprocessed and standardised using Python scripts to generate analysis-ready datasets. Following completion of data collection, the authors will make an anonymised dataset available on reasonable request. Data sharing will occur in accordance with ethical approval and data-protection guidelines.^{62 63}

Patient and public involvement

No patients or members of the public were involved in the design, conduct, reporting or dissemination plans of this single-session feasibility protocol. This decision reflects the study’s acute laboratory setting, brief exposure and student volunteer population, which limited opportunities for co-design within the current timeline and resources. Although PPI was not feasible for the present protocol, the team plans to incorporate patient and public input in a subsequent multi-arm trial (eg, advisory input on control comparators, acceptability outcomes and dissemination materials).

Discussion

This protocol is designed to assess immediate changes in state stress (STAI-State, STAI-Trait) and cardiac indices (heart rate/HRV) following a standardised Stroop stressor after a single 15 min VR-based observation meditation session augmented with AI coaching (Otti), while concurrently evaluating acceptability and usability in a structured manner.⁶⁴ The single-visit pre–post design is well suited to identify early signals of efficacy and feasibility within a low-burden, 30 min assessment frame, thereby providing a pragmatic on-ramp to subsequent multisite and multisession clinical trials. By concurrently measuring subjective indices (STAI, DASS-21, PSS-10) and physiological indices (HR/HRV), the study enhances convergent validity for short-term responses and probes whether the presence and attentional benefits of VR, together with the personalised interaction afforded by AI coaching (emotion analysis, empathic feedback, MI), can mitigate traditional barriers to mindfulness practice (time, space, reliance on an in-person coach).⁶⁵ In addition, by prespecifying the linkage between acceptability indicators (satisfaction, intention to reuse) and clinical change (eg, ΔSTAI, ΔDASS-21 Stress), the study provides hypothesis-generating evidence on the relationship between user experience and short-term clinical effects.

Interpretation should consider two points. First, single-centre convenience sampling of university students may constrain sample representativeness and thus limit external validity, that is, the generalisability of findings to groups with different age, cultural or clinical characteristics. Second, because the study targets immediate, single-session effects, conclusions regarding durability or translation to everyday use should be reserved; as commonly noted in pilot and feasibility research, these questions are best addressed in follow-on work using multisession intervention schedules and longer-term follow-up (eg, ≥3–6 months).⁶⁶ Rather than detracting from the contribution, these considerations help prioritise design features for subsequent confirmatory studies. In practical terms, adopting multisite, multiregion sampling with stratified randomisation can strengthen external validity by accommodating population and context heterogeneity, while direct comparisons with alternative formats (traditional mindfulness, mobile applications, non-VR digital interventions) can more precisely gauge relative effectiveness.⁶⁷ Further, characterising dose–response relationships by session number and exposure time, incorporating HRV subcomponents with additional biomarkers and digital interaction logs to test mechanistic pathways (eg, emotion regulation, attentional shifting, defusion), and conducting implementation and cost–effectiveness evaluations will inform real-world scale-up strategies.⁶⁸

In sum, this protocol is positioned to detect initial signals that a concise VR-plus-AI observation meditation session can elicit immediate reductions in state stress and physiological arousal while achieving acceptable user experience.⁶⁹ Within the acknowledged premises of single-centre convenience sampling and a single-session scope, the planned assessments provide a methodological and practical foundation to justify and shape subsequent multisite, multisession trials and mechanistic investigations.

Ethics and dissemination

Ethics approval and consent to participate: The protocol was approved by The Pennsylvania State University Institutional Review Board (PSU CATS IRB: STUDY00025978, ClinicalTrials.gov code: NCT06704282). All participants will provide written informed consent prior to enrolment and any data collection procedures.

The study findings will be disseminated through multiple complementary channels to maximise reach and utility. Primary dissemination will occur via peer-reviewed open access publication accompanied by supplementary materials consistent with journal policy. Secondary dissemination will include presentations at national and international conferences and seminars in digital mental health, VR/AI and clinical psychology. In accordance with Standard Protocol Items: Recommendations for Interventional Trials/ICMJE guidance, any changes to the dissemination or data-sharing plan will be reflected in the ClinicalTrials.gov registry as needed.

An anonymised dataset for key outcomes (eg, pre-/post-state stress and heart rate/HRV indices) will be made available on reasonable request within ethical and data-protection constraints, following a review process by the investigative team with data custodian oversight. The article will include a Data Availability Statement aligned with BMJ Open’s Tier 2 data policy and ICMJE guidance.