Supporting Secondary School Student Wellbeing Through a Mobile App and Wearable Biofeedback

Overview

This study aims to investigate the effects of an app and wearable device on the wellbeing of secondary school students. The app will offer support for stress management, sleep improvement, and time management. Additionally, it will provide access to educational resources, text-based coaching, and biofeedback monitoring through the wearable device. The device, worn around the wrist, will measure heart rate variability as an indicator of stress response. Throughout the study, students will complete bi-monthly surveys on sleep, stress, and overall wellbeing. To assess the specific impact of each feature, the app features will be released incrementally over the course of a year. The intervention group, comprising students in their last two years of secondary school, will have access to the app and wearable device, while a control group within the same class will not use the system but will complete the same wellness surveys. Coaches on this platform will come from the Centre for Positive Health Sciences MSc in Positive Health Coaching as part of their continued training in coaching. Coaching will be asynchronous so that students can post questions at any time during the day, but coaches will respond within 24 hours, during office hours. It will be made clear that this is not a therapeutic service and will be directed to other resources for these services. Coaches will file an incident report if a risk arises from a message with participants. Participants will be given guidelines for the type of questions they can ask such as asking for advice for setting goals related to their digital habits and sleeping better. The study\'s findings will provide valuable insights into the effectiveness of the app and wearable device in enhancing student wellbeing leading up to their leaving certification exams.

App-based wellness resources, coaching support, and biofeedback monitoring for sustained health and wellbeing of secondary school students Justin Laiti, Dr. Elaine Byrne, Dr. Pádraic Dunne Background Secondary school students across the world face ongoing academic demands that can negatively impact their mental, physical and social health. Short-term impacts can lead to poor academic achievement, decreased motivation and increased school dropout, while long-term impacts can result in mental health issues (anxiety and depression), sleep disturbances, substance abuse and ultimately, unemployment. The Economic and Social Research Institute (ESRI) conducted a longitudinal study (2011) of 900 Irish Post-primary school students and found that exam-related stress was a significant issue, especially related to the Leaving Certificate exam. Some of this exam-related stress was mitigated by a positive relationship between the school teachers and the students. While there have been reports of decreasing wellbeing among secondar school students1, little evidence exists regarding the biopsychosocial impacts of school-related stress on Irish students. However, the investigators might assume that Irish students experience similar issues to their international counterparts. Biofeedback Biofeedback involves the use of wearable devices that can measure physiological parameters such as heart rate, pulse, brain waves, heart rate variability (HRV) and skin galvanic response. Users can monitor these parameters (wirelessly or by direct connection) in real-time through the wearable device itself or via a smart phone application (app). Users can monitor how behaviour, as well as external or internal stimuli can impact on these physiological parameters. In essence, biofeedback can help users to improve behaviours related to greater health and wellbeing. For example, HRV is widely regarding as an excellent objective measurement of stress when measured over time (Kim, Cheon, Bai, Lee, \& Koo, 2018; Lehrer \& Gevirtz, 2014). HRV as a physiological parameter, is a measure of the time intervals (seconds) between heartbeats, as well as the variability, duration and strength (measured by power density spectral analysis) of the same interval. Coaching An increasing evidence base is emerging from the scientific literature that describes the positive impact that coaching can have on the overall wellbeing of senior secondary school students in particular (Dulagil, Green, \& Ahern, 2016; Gibbs \& Larcus, 2015; Green, Grant, \& Rynsaardt, 2020). Telemedicine (Pearl \& Wayling, 2022) and online coaching (Passmore \& Evans-Krimme, 2021) have become more palatable to the public and impactful on health outcomes since the global COVID-19 pandemic in March 2020. The investigators have shown from our preliminary research (RCSI Coach Connect) that online coaching for the health and wellbeing of Irish Healthcare professionals working in the RCSI Hospital Group, can have positive impacts on wellbeing, overall happiness, positive emotions and burnout (pre-publication). Online, text-only coaching with accredited coaches, trained in online communication can provide consenting users with an accessible and economically viable form of frequent support. Project aim This project aims to build and implement an application with text-based coaching, educational resources, and biofeedback monitoring of fluctuations in HRV for secondary school students. The investigators are interested to learn if a combined approach whereby students are supported via online coaching plus continuous biofeedback can have a positive impact on their health and wellbeing. Outcomes/objectives Measure uptake and engagement by secondary school students of the app during the feasibility study and during the students' final year. Observe changes in perceived stress as measured by the Perceived Stress Score, Pittsburgh Sleep Quality Index-Adolescent (PSQI-A) assessment, and the EPOCH Measure of Adolescent Well-being at 3-month intervals. Perceived stress, sleep quality, and the wellbeing will be compared between the study cohorts from both schools, compared with representative counterparts who will not have access to the support platform. Observe continuous changes in HRV of study participants, as measured by a bespoke, Bluetooth-enabled wearable device. Observe correlations between HRV and routine check-ins delivered three times a week through the app which will include mood, stress level, and drowsiness Conduct a qualitative assessment (6-month intervals) of representative users regarding platform engagement, and the coaching process, as well as needs related to health and wellbeing. Deliverables Report on psychological, social and physical health baselines of on-boarded participants Three-monthly reports on perceived stress (Perceived Stress Scale), Pittsburgh Sleep Quality Index-Adolescent, and EPOCH Measure of Adolescent Well-being Six-monthly reports on quality improvement cycles to inform better functioning, uptake and engagement Summary analysis report for students, educators, and policy makers Summary aggregate data report for public consumption on appropriate FAIR platforms Project Timeline Needs Assessment Survey In the first phase of this project (February 2023), a needs assessment survey was sent to the students, teachers, and parents at the schools involved in this study. The purpose of this survey was to gather information about the wellbeing needs of these groups so that the investigators can tailor the rest of the project towards. This survey was sent during students\' transition year (TY) and the data recorded was anonymous. It asked questions related to sleep, stress management, social connections, and the use of apps and wearables to support student wellbeing. Co-design In the second phase of this project (April-November 2023), co-design sessions were run with TY students at each school to facilitate discussions and workshop activities about students\' wellbeing needs and their opinions about the design and features of the wearable device and mobile app that will be used in this project. These sessions were completed over the course of four weeks in four, one-hour-long sessions. The anonymous notes and feedback from students were compiled from this phase to inform the development of the following phase. Wearable System Feasibility Test In the third phase of this project (March 2024 - May 2024), students will test the functionality of the app and wearable. They will initially onboard onto the app in March when they will also give feedback on the personalised features for their wearable including the case and strap colour and an inscription on the face of the case. Students will then receive their wearable during the first week of April. They will use the app and wearable until the end of the school year, completing routine check-ins and taking their HRV baseline at the start and end of the study. In the final week of the study, students will complete outboarding surveys and a small group at each school will participate in focus group feedback. Wearable System Implementation Pilot Study In the fourth phase of this project (September 2024 - May 2025), students will use the app and wearable throughout their final year at school. At this point they will be given access to a wearable heart rate variability monitor and a mobile application which will connect them with online coaching, educational resources, and information streamed from the wearable. Students will be able to use this system to support them throughout their 6th year leading up to their leaving certification exams. The following project description outlines the components of this fourth stage. Study design, approach and methodology Study design: The investigators will use a participatory mixed method design (Cresswell et al, 2011), incorporating a prospective cohort study design and qualitative research assessment with participation of the teachers and students involved in the study. The prospective cohort study will have two study groups based at two schools. The human coach-led chat, messaging smart phone application system, will allow accredited coaches to support secondary school student cohorts (44 students in each cohort) based at two Irish secondary schools between transition year (year 4) and the final two years of secondary school (years 5 and 6), culminating in the Leaving Certificate examination. The investigators also propose to support guidance counsellors and the year head teachers (based at the same schools) over the same period, since teachers are also exposed to significant stress (Scheuch, Haufe, \& Seibt, 2015). There will be assessment of bio-psycho-social parameters, including continuous HRV measurements, perceived stress, and wellbeing between baseline and 6-monthly assessments up to 30 months. Coaches on this platform will come from the Centre for positive Psychology and Health MSc in Positive Health Coaching as part of their continued training in coaching. Each coach will be supervised by a fully accredited Master Coach, provided by the Centre. Oversite of the coaching process will be managed by the PI (Pádraic Dunne) and Prof Christian Von Nieuwerburgh (co-director of the MSc in Positive Health coaching and international coaching expert). Platform Development An android and apple compliant smartphone app will be created that allows participants to communicate with coaches via text messages. Coaches will use the application to share resources while communicating with participants. The app will also display HRV data for each participant and their respective coach. A collaborative and interactive programme (Figma) will be used to create wireframes for the app with the required functions. These wireframes generate clickable mock-ups of the app flow, layout and graphic design. Students and teachers will have the opportunity to create their own designs or give feedback on already existing designs for the app. Once the layout is confirmed, all information will be coded to develop a functional smartphone app. Swift and Android Studio will be used to code for apps that will be functional on Apple iOS and Android devices, respectively. Once the app is developed, the user interface and functionality will be tested with the teachers and students involved in the study. This will ensure the functionality of the app in terms of being able to (1) connect participants to coaches, (2) stream user HRV data, (3) provide a seamless interface between user and the system and ensure data security. Wearable Device Development A relatively low-cost, wearable device (estimated cost per assembled unit is €70) will be assembled that measures HRV. The device will likely fit around the wrist and be enclosed in a skin compatible casing. The device will include a power button, a light indicator, and a charging/data transfer port. Components within the device will include a small circuit, a processing board, an SD card, and a power supply. Either a light emitting diode or an electrode will be placed along the bottom of the device to be in contact with the skin and measure heart rate using either photoplethysmography (PPG) or electroencephalography (EEG). Both methods are non-invasive and commonly used to measure heart rate and HRV. This will be a co-design process, so the students and teachers involved in the study will assist with the design and assembly decisions throughout the construction of the device. Once the device is assembled, it will go through a series of testing procedures before being included in the study. The investigators will verify the accuracy of the HRV sensor by comparing it to two commercial heart rate sensors (Inner Balance device developed by HeartMath; Muse 2 device developed by Muse Inc.) owned by the Centre for Positive Psychology and Health. The investigators will collect data on both devices simultaneously to ensure that there is not a significant difference in the variation data being captured by the devices. A risk matrix will also be created to identify any aspects of the device that could cause harm to the participants. Adjustments will be made to the device to ensure minimal risk to the participants. Once the device is tested and any risks to participants are minimal, the investigators will order the additional supplies necessary to construct all of the sensors and develop a standard testing technique to ensure the accuracy and safety of each one. The components for the device will be purchased from Digi-key Electronics and Amazon.co.uk. Individual devices will subsequently be co-assembled with students at each school. Digital support repository Secure video content on health and wellbeing, created by RCSI, HSE, Healthy Ireland, Sláintecare will be accessible in the form of secure uniform resource locator (url) addresses from government websites or secure RCSI Vimeo and YouTube accounts. Content will include a complete 8-week Attention-based Training (ABT) programme for health and the online RCSI Science of Happiness for Young People (8 and 10 week courses, respectively) provided by the RCSI Centre for Positive Psychology and Health. Data security A Data protection impact assessment (DPIA) will be completed to mitigate risks associated with data security. The data from both the wearable device and the messaging platform will be secured within the device itself and on an external drive. The wearable device data will be stored on an SD card until uploading to the password protected, mobile application via Bluetooth. The data will be available to the participant and the coach through this application and will be uploaded to an RCSI One Drive account which the research team will have access to. In addition to authentication required to use the application, data will be encrypted as it is transferred from the wearable device to the application and then from the application to the One Drive account. The investigators will obtain an SSL/TLS Certificate to securely transfer data using HTTP as well as encrypt the data while it is stored. The application will also be able to be updated to that security vulnerabilities can be addressed during the testing phase. The data from the wearable device will include HRV values and timestamps which will be uploaded to the app via Bluetooth. Application data (messages and HRV values) will be kept on the app for 30 days. This project will be conducted in line with current GDPR guidelines: the lawful basis of processing of personal health data is based on Legitimate Interests, under Article 6(1)(f), and Consent, Article 6(1)(a). All data collected will be reviewed and determined to be of high necessity and commensurate in proportionality. Data collected will be determined to be the lowest needed data processing to achieve the desired outcome. The data processing will be purpose built to help individuals most effectively monitor and improve their wellbeing. To prevent function creep, all data collected and processed will be reviewed by the PI and approved by the data controllers (PI Dr Pádraic Dunne and PhD Scholar Justin Laiti). For all data, nothing more than what is needed will be collected, and the scope of the processing activity will be held only to its defined purpose(s). The data controller and PI will ensure that privacy policies are built with the purpose of guaranteeing limitation and data minimisation. At consent, individuals (and their legal guardians) will be provided with the data collected and purpose of the data. If changes occur, and are approved by the data controller, individuals will be notified of the specific change(s). An updated consent form will be generated. All researchers must document, demonstrate, and provide proof that they meet GDPR requirements for data protection/security, privacy, and processing. Data will be hosted, stored, and backed-up in a virtual private cloud (RCSI) physically located in the EU (Dublin). Recruitment and participants. Forty-four secondary school students (beginning in transition year and culminating in final year of school) will be recruited from Athy College Post Primary DEIS School in Co. Kildare. Twenty secondary school students (beginning in transition year and culminating in final year of school) will be recruited from Cross and Passion secondary School in Kilcullen, Co. Kildare. Two year-head teachers and two guidance counsellors will be recruited from each school. Twenty representative control students (5th year and 6th year students) and six general schoolteachers, who will not have access to the platform, will be recruited from the same schools. The Centre for Positive Psychology and Health has run the RCSI Winter School at Athy College since 2020 and therefore has a good relationship with teachers and students at the school. The investigators have learned that an online information morning provided by RCSI researchers, and facilitated by head teachers and guidance counsellors with transition year students present, is the best way to provide information on the study and to answer any questions. The investigators will adopt the same approach both schools. The investigators will also hold an information evening for students and their guardians who may have questions about the project. Students, teachers and guardians/parents will receive hard copies of participant information leaflets (PIL) and consent forms. There will be an option for students who reach the age of 18 years during the life of the project to complete and sign a replacement consent form. The school guidance counsellor for each school will act as the gatekeeper for communication between students, their guardians and RCSI researchers. Recruitment of control participants and the Principle of Equipoise Although the investigators hypothesise that online text-based coaching will prove beneficial for the health and wellbeing of users (students and teachers), the benefits remain uncertain. At the end of the project and providing the coach-led platform is beneficial to wellbeing, the investigators will consult with teachers and students how the investigators might scale this offering to the remaining students and teachers at each school (resources permitting). Garda vetting for RCSI researchers The schools will facilitate the completion and processing of GARDA vetting forms for researchers in contact with students (virtually or in the classroom), as with previous iterations of the RCSI Winter School. Researchers will never be alone with any student; either the school guidance counsellor or year-head will be present in the classroom for in-person sessions. Digital support platform on-boarding process On-boarding will occur through smart phone applications, whereby consenting participants will: Receive a unique study code identifier and chose avatar name from a provided list of male and female first names. The code is made up by answering the following questions: Q1 Please type in the first letter of your mother\'s first name (e.g. Maria: M) Q2 Please type in the first letter of your father\'s first name (e.g. Andrew: A) Q3 Please type in the your house number (e.g. Number 42 = 42); if the participants don't have a house number, type "oo" Q4 Please type in the first letter of your place of birth (e.g. London: L) Q5 Please tick the box that contains the last number of your year of birth (e.g. if the participants were born in 2009 = 9): 0=12; 1=23; 2=24; 3=25; 4=26; 5=27; 6=28; 7=29; 8=30; 9=31. A full example might look like this: BP16D25 Identify preferred communication style and readiness to change level Identify personal goals relating to sleep, eating well, physical activity, alcohol, tobacco and other substance use, relationships and stress management Complete paper surveys in the classroom at six-month intervals assessing wellbeing and perceived stress Introduced to their coach via text through the bespoke project smartphone app Questionnaires and assessments The investigators will ask participants to complete two questionnaires in paper format in the classroom setting that are designed to measure current state of health and wellbeing and perceived stress. Perceived Stress score and the EPOCH measure of wellbeing. In his 2011 book Flourish, Dr. Martin Seligman, founder of the field of positive psychology, defined 5 pillars of wellbeing, PERMA (positive emotion, engagement, relationships, meaning, accomplishment). Seligman and colleagues adjusted the model to be developmentally appropriate for adolescents and referred to it as the EPOCH measure of wellbeing. The resulting model consists of five different positive characteristics that together support higher levels of well-being: engagement, perseverance, optimism, connectedness, and happiness (EPOCH). This is instrument has been validated in a number of different international research projects1. Quality improvement study. The investigators will use the IHI Model for Improvement (MFI) developed by Langley et al (1992) to create and model changes, test proposed changes, measure the results and then accept or modify the proposed changes. The MFI promotes rapid testing for rapid learning and each phase of QI will be fed back into the design and implementation of the coaching intervention. The approach involves three questions and a fourth section which is a tool for testing changes. The questions are: What are the investigators trying to accomplish? How will I know if a change leads to an improvement? What changes could the investigators make that will result in improvement? With the fourth section of Plan, Do, Study, Act (PDSA) allowing for rapid cycles of changes, measurement and learning. At the outset tools such as process maps, Pareto charts, Ishikawa (fish-bone) diagrams, and key driver diagrams will assist in the structuring and visual representation of the processes and setting. Besides the quantitative measures outlined above that will be included at all phases of the process semi-structured interviews (with teachers n=4) and Focus Group Discussions (n=2 with 7-8 students) will be conducted before and after each cycle. Students (in the form of focus group discussions) and teachers (semi-structured one-to-one interviews with researchers) will be interviewed for a maximum of 40 minutes after quality improvement cycles on each school campus. Researchers will not be permitted to interview students under 18 years, alone; a teacher will be present at all times. These interviews will be recorded by researchers using digital recorders. Once transcribed, all audio files will be permanently deleted. Transcripts will be stored in a secure location on RCSI main campus for a maximum of 12 months. Interviews will be assessed using NVivo software and thematic analysis. A participatory design approach will underpin our study design- from conceptualisation, implementation and improvement. Participant involvement is research 'with' or 'by' members of the public rather than 'to', 'about' or 'for' them (INVOLVE, 2016)2. Teachers and students will initially be recruited to assist in the design of the programme as well as the technological support, such as type of material that they potentially would find useful, timing of activities, type of wearable device and app they would find appealing etc. Thematic Analysis. Thematic analysis will be used to analyse all qualitative data. Thematic analysis is a theoretically flexible approach that allows the combination of inductive and deductive methods. This facilitates participatory research, where participants are seen as collaborators, and produces analysis that is accessible and can inform policy development. During the co-design stage of the project, analysis will focus on identifying themes that relate to the needs and experiences of students and teachers. In the evaluation stage, analysis will explore whether the online coaching support platform can help cultivate and sustain mental health and wellbeing among participants and any barriers to engagement. Thematic analysis will also allow researchers to focus on identifying patterns related to participants lived experience of using the digital health platform. Risks mitigation related to data management Risk 1 - User data is collected that isn\'t listed in the consent: a full review of consent, before each new release will be conducted to ensure that all data collected is included in the consent agreement. Existing users with receive a new consent with a description of the additional items added. Risk 2 - User is added as a user in the application prior to initial consent: users will have a clear process that is adhered to for gaining consent prior to adding new users to the application. Risk 3 - In cases of unauthorised exposure of data. To prevent data breach, the following aspects will be built into the online platform and data capture system: encryption keys; anonymised database; two-factor authentication; Host Intrusion Detection System (HIDS) Intrusion detection; malware protection; active system monitoring; distributed denial-of-service (DDoS) protection; a deny-all firewall; hypervisor detection; network \& host vulnerability scanning; logical and physical segregation; test data security systems; implementation of Auth0 software; implementation of GDPR training for all researchers. In the case of a data breach, we will inform the user and their legal guardian of this breach and subsequently conduct a comprehensive policy \& procedure review and implement remediation. Text coaching risk - "What if something goes wrong when I'm taking part in this study?" It is important to note that this coaching platform is not designed to take the place of psychotherapy and is not a therapeutic service. However, the psychological health of participants will be monitored by the system and assigned coach using the two survey instruments described in the methods (EPOCH wellbeing measure and Perceived Stress Score). High or low scores (reported by the bespoke system in terms of heart rate variability or via survey results) will be reported to the participant and their legal guardian by the PI (Dr Padraic Dunne), using their pre-defined preferred communication method. Users (as well as their legal guardians in the case of those under 18 years) will be advised to contact their personal GP and psychotherapeutic support. Coaches will complete an incident report for the Principal Investigator (Dr Padraic Dunne) who is a qualified, accredited psychotherapist. Physical metrics measured throughout the study will involve those related to heart rate variability (HRV) as an objective measure of stress. Participants (as well as their legal guardians in the case of those under 18 years) experiencing HRV measurements that are indicative of ongoing stress will be asked to contact their personal GP and/or psychology support. An incident report will be made to the Principal Investigator (Dr Padraic Dunne). Individuals who are in need of psychological support are referred to the list of supports provided by the Health Service Executive of Ireland (HSE) (https://www2.hse.ie/services/mental-health-supports-and-services-during-coronavirus/). This list includes contact details for Turn2Me, Shine, MyMind, Suicide-Or-Survive, Samaritans, PIETA house, Stress Control, Clanwilliam Institute, Helplink Mental Help, Aware, and Alone. Note - the RCSI sponsorship officer (Maurice Dowling) will be informed of this study and will receive a complete clinical trial registration form. Statistical analysis Descriptive statistics and longitudinal modelling of outcomes will employed throughout; the RCSI Data Science centre have and will advise the research group on appropriate application of relevant statistics during the project. Based on this advice, the investigators will analyse the data as follows: Internal consistency of the validated instruments will be measured using Cronbach's alpha test. Single imputation will be used to replace missing values. Mean and standard deviation values will be denoted as follows: (x= x\[y\]). Paired and unpaired t-tests (denoted as P values) will be used to examine intra- and inter-group differences. One- and two-way mixed analyses of variance will be used to compare multiple datasets with independent variables of group and time. Linear regression analysis will be used to examine changes in either group over time. Cohen's d and associated effect size calculations (U3) will be applied to mean and standard deviation results. Probability of superiority will be subsequently calculated using effect size values. Cluster analysis will be conducted on both schools as advised by the RCSI Centre for Data Management who have been consulted about this project. Qualitative interviews will be thematically analysed iteratively to inform quality improvements. TRE-AIM framework as an evaluative tool six-monthly quality improvement cycles focused on RE-AIM headings: Reach (uptake), Effectiveness (trajectories of distress), Adoption (applicability to other cohorts in future), Implementation (accounting for adaptations/changes, barriers), Maintenance. Other quality improvement tools such as fishbone analysis, SWOT analysis, and stakeholder maps will also be employed. Power calculations will be determined at the end of year one when changes in wellbeing and perceived stress will be compared between controls and intervention group students and teachers.