Adapting and Assessing the Feasibility of a Telehealth Diabetes Prevention Program for Hispanic Adolescents

Overview

Hispanic adolescents are disproportionately burdened by type 2 diabetes (T2D). Social determinants of health (SDoH) serve as barriers to behavior change and participation in disease prevention efforts, especially among vulnerable adolescents. Telehealth is a potentially effective approach for delivering disease prevention programs as it addresses some SDoH like transportation, childcare needs, and parent work schedules. Unfortunately, there are no theory- or evidence-based telehealth diabetes prevention program for Hispanic adolescents. Therefore the purpose of this study is to adapt an evidence-based diabetes prevention program for delivery via telehealth and to test the feasibility of this study among Hispanic adolescents (12-16 years) with obesity.

Hispanic adolescents (N=24) and parents (N=24) will be invited to participate in at least three study activities across Phases 1-3 and will be compensated. In Phase 1, adolescents and parents will participate in separate focus groups. As an alternative to the focus group, youth will have the opportunity to answer the same questions via an interactive survey. The purpose of these focus groups is to elicit perceptions of Self-Determination Theory (SDT), text messages, desired program features, and gather quantitative data (survey) on technology use. For this study, the SDT constructs will include autonomy, relatedness, and competence. We will elicit feedback from adolescents and parents on strategies for feasibly engaging parents in the intervention. Focus group questions will be open-ended, allowing participants to fully describe their thoughts and preferences. In Phase 2, an expert panel will meet once during Phase 2 and once during Phase 3. In the first meeting, we will present initial program adaptations and findings from Phase 1 to explain the adaptations made. The panel will provide feedback and suggestions on additional modifications. In the second meeting, an example web-based nutrition education session will be reviewed, followed by a discussion on session content and delivery format, and a 10-item feedback survey. In Phase 3, a field usability testing with 12 of the original sample of 24 to provide feedback on a sample nutrition education session and the functionality of the activity tracker and text message components. Participants will be asked to use their own screen device to watch a sample session and answer questions regarding the acceptability, usability, and age and cultural appropriateness of the session. We will use a brief phone interview to clarify survey responses. To conduct field testing on the activity tracker and text messages, six participants will be given a Fitbit device and will self-select 1-2 peers who will provide support as they receive daily text messages that promote physical activity and sleep for four weeks through a groupchat platform, Remind. No data will be collected on the peers, they will simply provide support given that this was requested by youth in Phase 1. The other six participants will also be given a Fitbit device and will receive daily text messages that assist youth in intention planning to promote physical activity and sleep for four weeks. A chatbot will be developed by Mobile Coach and will be used to assist youth in setting behavioral goals and planning opportunities to be active or to improve sleep behaviors. This is a HIPAA secure chatbot that has no natural language processing or conversational capabilities and can only use text provided from our text message bank. A brief phone interview will be used to elicit perceptions on using the Fitbit, text message content, format, and frequency. We will also ask participants to identify barriers to using the Fitbit, suggestions for sustaining engagement, and the feasibility of tips and strategies shared via text. Following these 3 phases, we will test the feasibility and acceptability of the intervention in a new sample of Hispanic youth (N=40). We will use a two-armed, pre-post study design. Following baseline assessments, adolescents will be randomized (1:1), stratified by sex, to the intervention or a control group. A priori feasibility criteria, process evaluation data, and exit interviews will be used to assess feasibility and acceptability. Participants will complete pre-and post-intervention assessments to examine preliminary efficacy and will be compensated for their time and participation. Intervention The intervention promotes changes in diet and PA by: 1) fostering social support and 2) enhancing self-efficacy for healthy behavior changes. There are 12 pre-recorded nutrition and well-being sessions in Spanish and English lasting 20-40 minutes and hosted on an elearning platform, Learnworlds. This is a secure platform and each participant and their family members will have a secure, unique, password protected login. Only study participants will have access to Learnworlds via a website browser or its accompanying smartphone application. The platform includes a discussion board to share challenges and successes in behavior change, a recipes tab, and message system to ask instructors questions or report technical issues. An initial orientation will be used to introduce participants to the platform, give them study materials including intervention instructions, a Fitbit, a workbook, and food items for the food tasting experience. At orientation, the rules and privacy regulations for use of the activity tracker, text messaging, and expectations for program participation. Participants and their family will be asked to watch one session per week. Sessions reinforce healthy diet and physical activity (PA) habits, assist in setting weekly diet goals, problem-solving challenges in meeting goals, and an outline of session objectives. Participants will be incentivized through a point system for attendance, out of class assignments, participating in group discussions, and progress towards their individual health goals. Families will be encouraged to work together to prepare four recipes throughout the intervention to expose them to healthy recipes and teamwork. These recipes will include apples and peanut butter, oatmeal pancakes, tostadas, and lentils. The research team will provide families with oatmeal, peanut butter, black beans, and corn to make the recipes feasible to prepare. There will also be three live sessions held separately for parents and youth via Learnworlds to address questions regarding content and to foster social support among participants. PA Component: Participants will receive a Fitbit Charge, which has shown high validity and reliability for assessing steps/day in adolescents. A chatbot developed by Mobile Coach will be used to assist youth in setting behavioral goals and planning opportunities to be active or to improve sleep behaviors. This is a HIPAA secure chatbot that has no natural language processing or conversational capabilities and can only use text provided from our text message bank. Starting week 2, the bot will assist youth in setting goals for steps/day based on their average daily step count from week 1. In successive weeks, participants will be guided to gradually increase their step count goals to meet or maintain current recommendations of at least 12,000 steps/day. Participants that average \<5,000 steps/day will be guided to increase their average by 600 steps/day each week, participants that average 5,000-11,999 steps/day will be guided to increase their steps/day goal by 10% of their week 1 average in successive weeks, and participants with ≥12,000 steps/day or greater will be instructed to maintain this average in successive weeks. Step count data from the Fitbit Charge will be continuously collected and wirelessly transmitted to Fitabase, a web-based platform. Fitabase aggregates daily step data and provides customized data reports. Implementers will provide feedback on goal attainment. Daily text messages will reinforce session content and promote autonomous motivation for PA. At the start of each week, texts will promote autonomy by encouraging adolescents to develop their own solutions for achieving their weekly PA goal. Mid-week texts will emphasize competence by acknowledging progression towards goals, providing evidence-based strategies for increasing PA, and reinforcing content from nutrition sessions. End of week texts will provide feedback on goal attainment and encourage adolescents to connect goal achievement to a personal value to build relatedness and a sense of connection. At the completion of the intervention, adolescents will return the Fitbit Charge to the study team. Assessments Primary Outcomes: Feasibility is the primary outcome and will be evaluated using a priori criteria. Recruitment and Data Collection criteria will assess the feasibility of recruiting and retaining participants from community clinics. This will provide information on the availability of high-risk adolescents in community clinics and the feasibility of recruiting and engaging adolescents from this setting. The integrity of Study Protocol criteria will be used to identify the feasible and acceptable intervention doses. Technical Issues criteria will assess the feasibility of using Learnworlds, a Fitbit, and text messaging. Fidelity will be assessed as a team member will observe 75% of nutrition sessions and will complete fidelity checklists. Satisfaction will assess program acceptability. If feasibility criteria are satisfied it will be used to develop a future R01. Secondary outcomes: Type 2 diabetes risk will be assessed using a point of care multi-assay analyzer, the Afinion 2 analyzer. Height and weight will be measured to the nearest 0.1cm and 0.1 kg, and used to calculate BMI% and BMI will be calculated as the percentage over the 95th percentile. Physical Activity (PA) will be assessed using a hip-worn accelerometer (wGT3X-BT, Actigraph). Participants will wear the accelerometer during wake time hours for 7 days. Participants will receive daily text messages to promote compliance in device wear. A valid day of wear will be defined as ≥10 hours of wear on ≥4 days, one of which has to be a weekend day. Dr. Soltero has extensive experience using accelerometers to assess PA and has achieved an 82% compliance rate in similar protocols. Data will be analyzed using Evenson cutpoints in the ActiLife software to report average minutes of PA/day. Assessments will be collected following the orientation at a convenient, private community center including Baker Ripley or the YMCA. Tertiary Outcomes: Autonomous motivation for physical activity will be assessed using the Behavioral Regulation in Exercise Questionnaire. Both instruments have been validated among adolescents. We will also use the digital health equity survey at baseline to ensure the equitable implementation of a digital intervention among a high-risk, low socioeconomic population. Last, youth will complete the physical activity identity survey to understand the intervention's ability to improve one's belief that they can live an active life. At baseline, parents will be asked to complete a demographic survey. Process Evaluation data will evaluate feasibility and identify contextual factors that impact implementation and outcomes. We will notate recruitment (e.g. staff time, materials, equipment), implementation, and retention efforts (e.g.participant calls, texts). We will also assess implementation fidelity, dose of intervention delivered and received, and potential contamination.Implementation fidelity, dose delivered, and dose received, will be defined by device wear and technical issues. Control-youth in the control will receive a handout on diet and PA guidelines and setting SMART goals. Control youth will be given access to the nutrition education website after their participation as a control for 12-weeks. No additional data will be collected on control youth beyond the 12 week intervention period.