Family Matters Intervention

Reducing Childhood Obesity Using Ecological Momentary Intervention (EMI) and Video Feedback at Family Meals

The proposed study is a 12-month, individual randomized controlled trial (RCT). The main aim of the study is to decrease childhood obesity and improve child diet quality in children ages 5-10 years old by increasing family meal quality (i.e., dietary intake, interpersonal atmosphere) and quantity (i.e., frequency of family meals) via innovative technology (i.e., ecological momentary intervention (EMI), video feedback) and partnerships with primary care and Community Health Workers (CHWs).

The RCT has the following three arms: (1) Ecological Momentary Intervention (EMI); (2) EMI + HV + Video feedback (virtual); (3) EMI + HV + Video feedback (hybrid). Delivery of the intervention will last 6 months for each family, with a 6-month post-intervention evaluation visit. All arms will receive EMI family meal tips via smartphones for 16 weeks. Arms 2 and 3 will also receive 16 weeks of in-home training (arm 2 will be virtual, arm 3 will be hybrid virtual/in-home), with 8 weeks (every other week) in-home education visits with a CHW focused on family meal quality and quantity and a family meal preparation activity and 8 weeks "Try it Yourself" activities that reinforce the messages and skills taught by a CHW. Additionally, Arms 2 and 3 will receive video feedback from family's video-recorded family meals by a CHW focusing on family behaviors related to family meal quality and quantity. (EMI, in-home training, and video-feedback will occur during the same 16-week period.) After families have completed 16 weeks of the intervention, a 8-week maintenance phase will ensue. Having a maintenance phase is an evidence-based intervention component and will provide incrementally less support to families to build self-efficacy and increase sustainability of new behaviors. Over time, participants will receive less study support to evaluate if participants have internalized healthful behaviors. During the maintenance phase EMI meal tips will be reduced to only the days in which parents report their highest stress levels for all arms. The study will last 12 months, with three assessment time points including, baseline, 6 months (i.e., post-intervention) and 12 months. Children ages 5-10 and family members (i.e., parent/primary caregiver, siblings) from low-income and diverse households (i.e., African American, Hispanic, Native American, Asian, White) - who are at high risk for obesity - will be recruited for the study. Theory informs the intervention study design, research questions and related hypotheses, methods, measurement, and analysis. Family Systems Theory drives the decision to direct the intervention at the "family unit" to increase the likelihood of individual and family-level weight and weight-related behavior change and sustainability. Additionally, partnerships with existing community-based healthcare systems and CHWs will be utilized to reach participants in community settings where they have existing relationships and resources. This study utilizes innovative and research-informed intervention components (i.e., in-home visiting, EMI, video-feedback) to increase the likelihood of intervention effectiveness and sustainability. The "Family Matters" study will be carried out across two Phases, an R61 Phase and an R33 Phase. Specific aims for each Phase are described below: Specific Aims for the R61 Phase: Aim 1: Organize and prepare the study team, project materials, study protocols, and procedures. Aim 1 success will be measured by meeting established R61 Planning Phase transition milestones. Aim 2: Test recruitment, data collection, intervention protocols, and retention in the first 42 participants (4 per study arm and about equal numbers byrace/ethnicity). Aim 2 success will be measured by meeting transition milestones. Specific Aims for the R33 Phase: • Aim 1 (Primary Outcomes): Conduct a three-arm RCT comparing EMI, EMI+HV, and EMI+HV+Video Feedback in diverse children ages 5-10 and their families to test the hypotheses that: Hypothesis 1: BMI percentile will decrease and diet quality will increase in children in the EMI+HV+Video Feedback hybrid arm compared to children in the EMI or virtual-only arms. • Aim 2 (Secondary Outcomes): Examine intervention effects on familial, parental, and sibling factors. Hypothesis 1: Family meal quality and quantity will increase in households with children in the EMI+HV+ Video Feedback hybrid arm compared to children in the EMI or virtual arms. Hypothesis 2: Controlling feeding practices (e.g., restriction) will decrease and coping skills will increase in parents in the EMI+HV+Video Feedback hybrid arm compared to parents in the EMI or virtual arms. Hypothesis 3: BMI percentile will decrease in siblings in the EMI+HV+Video Feedback hybrid arm compared to siblings in the EMI or virtual arms. • Aim 3: Examine cost effectiveness and feasibility of intervention implementation in primary care. Hypothesis 1: The BMI z-score reduction resulting from the intervention will be cost-effective.

Like most behavioral interventions, it is not possible to double blind this RCT because interventionists will be delivering the in-home visiting components to participants. However, the proposed study incorporates participant, study staff, physician, and investigator blinding as much as possible. The intervention will be administered by trained staff assigned to the intervention team, who will not be involved with evaluation team responsibilities/meetings. Evaluation team members will carry out all measurement home visits and will be blinded to participant study arm assignment. They will not be involved with intervention team responsibilities/meetings. Specific co-investigators are responsible for training and supervising staff within the intervention team versus the evaluation team in to order to blind as many co-investigators as possible.

EMI (Arm 1): (1) tailored ecological momentary intervention (EMI) prompts sent to parents targeting momentary behaviors (e.g., stress) around family meal quality and quantity for 16 weeks; and (2) a 8-week maintenance phase with EMI tips delivered on high stress days.

EMI + HV + Video Feedback (Arm 2) education visits will all be delivered virtually, and the arm includes: (1) tailored ecological momentary intervention (EMI) prompts sent to parents targeting momentary behaviors (e.g., stress) around family meal quality and quantity for 16 weeks; (2) bi-weekly in-home educational visits (total of 8) with a community health worker (CHW) focusing on family meal quality (i.e., dietary quality, interpersonal quality) and quantity (i.e., frequency of family meals), and 8 weeks "Try it Yourself" activities that reinforce the messages and skills taught by a CHW (for a total of 16 weeks); (3) video feedback on a video-taped family meal delivered every other week during the in-home visit with the CHW; (4) a 8-week maintenance phase with EMI tips delivered on high stress days.

EMI + HV + Video Feedback (Arm 3) education visits will be delivered virtually and in-home, and the arm includes: (1) tailored ecological momentary intervention (EMI) prompts sent to parents targeting momentary behaviors (e.g., stress) around family meal quality and quantity for 16 weeks; (2) bi-weekly in-home educational visits (total of 8) with a community health worker (CHW) focusing on family meal quality (i.e., dietary quality, interpersonal quality) and quantity (i.e., frequency of family meals), and 8 weeks "Try it Yourself" activities that reinforce the messages and skills taught by a CHW (for a total of 16 weeks); (3) video feedback on a video-taped family meal delivered every other week during the in-home visit with the CHW; (4) a 8-week maintenance phase with EMI tips delivered on high stress days.

Families will participate in twelve in-home education activities. The visits will focus on family meal quality (dietary, interpersonal) and quantity (meal frequency) factors found in prior studies to be associated with child weight and weight-related behaviors. Eight of the visits (every other week) will be delivered by a CHW and will last 90 minutes. Additionally ,a family meal food preparation activity will occur to help support families in learning skills to increase family meal quality and quantity. On weeks opposite of the in-home education visits with a CHW, families will be given a "Try it Yourself" activity to reinforce the messages (e.g., use family meals as a family connection time) and skills (e.g., recipe that requires families to steam vegetables) taught by the CHW during the in-home education visits.

Parents will receive daily EMI messages regarding suggestions for increasing meal quality/quantity via a smartphone application, based on the stress level and source of stress they reported earlier in the day. Specifically, parents will receive two push notifications in the morning to report their stress level and type/source of stress (e.g., work, children, finances). Then, based on the reported stress level and source of stress, parents will be texted tips/ideas for carrying out a nutritionally and emotionally healthful family meal the same night, in the face of stress. If parents report no stress on their EMI measure earlier in the day, they will be provided with a menu of options regarding what type of meal tip they would prefer later in the day (e.g., recipe ideas, meal prep tips, mealtime conversation starters. Parents will be able to respond to the EMI messages indicating preferred tips so that EMI tips become more meaningful/relevant to the needs of each family.

Parents will video-record and upload via an iPad (study provided) one family meal every other week, which will be watched by intervention staff. Using Motivational Interviewing, CHWs will give specific feedback on how to adapt negative mealtime behaviors and provide reinforcement for positive behaviors seen in the videos, based on a validated interpersonal coding tool called the Iowa Family Interaction Rating Scales (IFIRS). Additionally, feedback will be given on the dietary healthfulness of the meal, based on a validated coding tool called the Healthfulness of Meal Index (HOM). After feedback, family members will be taught specific skills to improve their family meal processes and behaviors that vary from food preparation skills, to increasing healthful food options at meals, or role-playing family interactions to improve the emotional atmosphere at the meal. Families will also set SMART goals at each visit regarding family meal quality and quantity.

After 16 weeks, EMI family meal tips for all arms will be reduced to only the days in which parent's report their highest stress levels. Stress profiles will be created for each parent in order to tailor the EMI family meals tips to days in which they experience their highest levels of stress. For example, during the 16-weeks of EMI, a parent may report their highest stress levels on Tuesdays and Thursdays. The stress profile that would be created for the parent during the maintenance phase would include receiving text messages only on Tuesdays and Thursdays. The parent would then receive these EMI family meal tips only on Tuesdays and Thursdays throughout the 8-week maintenance phase..

Based on Centers for Disease Control and Prevention (CDC) guidelines, child age and sex will be used to calculate an adjusted BMI percentiles. Measured at three time points over 12 months by study staff.

Child three-day, 24-hour dietary recalls will be used to calculate a Healthy Eating Index (HEI) score, which measures overall dietary quality.

Family meal dietary quality as measured by the Healthfulness of Meal (HOM) index via video-recorded direct observation at three time points over 12 months.

Family meal emotional atmosphere quality as measured by the Iowa Family Interaction Rating Scales (IFIRS) via video-recorded direct observation at three time points over 12 months.

Based on Centers for Disease Control and Prevention (CDC) guidelines, sibling age and sex will be used to calculate an adjusted BMI percentiles. Measured at four time points over 12 months by study staff.

Parent feeding practices (e.g., restriction, pressure-to-eat) will be measured at three time points over a 12-month period using validated measures in EMA surveys.

Parent coping (i.e., ability to manage stress) will be measured at three time over a 12-month period using validated measures in EMA surveys.

STUDY INCLUSION CRITERIA The inclusionary criteria used to select participants for the proposed study includes: Boys and girls (n=525) ages 5-10 years old who attend a University of Minnesota Physicians (n=4) or Fairview (n=8) primary care clinic and their primary caregiver (e.g., mother, father, grandparent) and at least one sibling. A second primary caregiver and other family members can also participate. Must eat ≤3 family dinner meals per week. Research shows that youth who have more than three family meals per week are less likely to be obese ten years later. Thus, families who report three or fewer family meals per week will be recruited to optimize the potential for change in family meal frequency. Families will focus on the family dinner in the intervention to improve their family meal quality and quantity. This decision is based on: (1) Family Systems Theory, which indicates that change in one setting (e.g., family dinner) will generalize to other settings (e.g., breakfast, lunch, snacks). Child with age and sex adjusted BMI ≥75th percentile (no upper limit). One of the following race/ethnicities: African American, Hispanic/Latino, Native American, Asian American, or White. In order to examine racial/ethnic differences in study hypotheses, equal numbers of children per racial/ethnic group (total=525) will be recruited. Parent and family members who speak English or Spanish. Not expected to move within the next two years. STUDY EXCLUSION CRITERIA The exclusionary criteria used to exclude participants from the proposed study includes: Children with medically necessary dietary restrictions (reviewed by MD from primary care clinics). Non-custodial parents, who the child does not live with more than 50% of the time.

Data from the proposed study will be shared through the NHLBI data repository. As required by NHLBI, the data set will be submitted to the study NHLBI study Program Official no later than 3 years after the end of the clinical activity (final patient follow-up, etc.) or 2 years after the main paper of the trial has been published, whichever comes first. Following the NHLBI policy for data sharing from clinical trials and epidemiological studies and the guidelines for NHLBI data set preparation, we will submit our data set including the following measures: baseline, interim visit, ancillary data, procedural based data, and outcome data, along with laboratory measurements not otherwise summarized. Data will be formatted using standard statistical software and modifications and redactions will follow NHLBI guidelines.

No later than 3 years after the end of the clinical activity (final patient follow-up, etc.) or 2 years after the main paper of the trial has been published, whichever comes first (following NHLBI policy for data sharing).

As specified by NHLBI, we will submit our data set including the following measures: baseline, interim visit, ancillary data, procedural based data, and outcome data, along with laboratory measurements not otherwise summarized. Data will be formatted using standard statistical software and modifications and redactions will follow NHLBI guidelines.