A Novel Obesity Prevention Program for High-Risk Infants in Primary Care (THRIVE)

A Novel Obesity Prevention Program for High-Risk Infants in Pediatric Primary Care: The THRIVE Randomized Controlled Trial

The goal of this clinical trial is to test a responsive parenting obesity prevention program with infants and caregivers of color (e.g., non-White; Hispanic/Latinx) and/or who are economically marginalized (i.e., publicly insured), delivered via Integrated Behavioral Health (IBH) in pediatric primary care. The main questions it aims to answer are: 1) Is the obesity prevention intervention delivered via IBH in pediatric primary care feasible and acceptable to families of color and/or families who are economically marginalized? 2) Will it prevent rapid weight gain during infancy? Participants will complete baseline (newborn), post-treatment (9 months), and follow-up assessments (12 months). Participants assigned to treatment will receive 4 prevention sessions as part of their typical well-child visit in pediatric primary care. Researchers hypothesize that infants in the obesity prevention intervention will have stable weight gain compared to infants in the control group (treatment as usual) will experience more rapid weight gain.

Unprecedented rates of obesity are occurring in childhood and disproportionally affect Black or African American youth, Hispanic or Latinx youth, and youth from economically marginalized backgrounds beginning as early as infancy. Obesity in infancy is not outgrown, tracks into adulthood, and places infants and children with obesity at a higher risk for significant medical comorbidities (e.g., cardiometabolic complications) in adulthood. The healthcare cost of obesity is ~$260 billion annually across the lifespan. Recent evidence suggests that infancy may be a critical period for the development of this high weight trajectory, as 10% of infants meet criteria for high weight-for-length; with the incidence being even higher among infants of color (16.3%), infants of Hispanic/Latinx ethnicity (12.1%), and economically marginalized infants (12.2%). Several modifiable predictors of obesity risk have been identified in infancy, including rapid weight gain in the first year, parental use of food to regulate infant distress, early introduction to solid foods, and insufficient infant total cumulative sleep. Given the greater risk for obesity, there is a critical need to identify and improve access to the most effective obesity prevention treatments for this population as a means of reducing the long-term healthcare burden. This pilot RCT will compare a responsive parenting intervention to treatment as usual in pediatric primary care with 144 infants who are from communities of color (e.g., non-White; Hispanic or Latinx) and/or economically marginalized backgrounds (i.e., publicly insured). Families will receive four prevention sessions with the integrated behavioral health specialist at their routine primary care well-child visits during the first six months of life. The primary outcome measured is conditional weight gain (an indicator of rapid weight gain) at 9 months of age. The responsive parenting approach, delivered by an integrated behavioral health expert, has been culturally adapted for infants and caregivers who are economically marginalized and/or from communities of color through focus groups. The delivery of this intervention via integrated behavioral health in a pediatric primary care setting has the potential to have a significant public health impact in terms of preventing pediatric obesity later in life; and thus, improving health outcomes and reducing health disparities.

obesity prevention, infancy, integrated primary care, integrated behavioral health, responsive parenting, responsive feeding, sleep, soothing

This is a Phase 2 pragmatic pilot and feasibility randomized controlled trial (RCT) of an obesity prevention intervention compared to usual pediatric care with 144 mother-infant dyads randomized n=72 per arm. Consistent with the definition of a pilot and feasibility clinical trial, our aims are to assess trial protocol feasibility based on THRIVE 2.0 and obtain effect sizes to inform the design of a Phase 3 multi-site efficacy trial.

The Teaching Healthy Responsive Parenting during Infancy to promote Vital growth and rEgulation (THRIVE 2.0) group will consist of 4 sessions delivered by the trained IBH provider, in conjunction with the 1, 2, 4, and 6 month well-child visit in primary care. The THRIVE curriculum teaches responsive parenting principles targeted to establish healthy eating, sleeping, and regulation habits for infants.

Control arm: Usual Pediatric Care. The control condition is usual care delivered by pediatricians at 1, 2, 4, and 6 month well child visits.

Overview of content and strategies: 1) Education on stage of development specific to infant feeding, sleep and infant regulation, attunement to cues; 2) Attunement to cues: reinforcing and modeling accurate discernment & responsiveness to cues; 3) Infant Soothing and Regulation: Coaching alternative strategies for soothing that feeding for non-hungry, fussy infants; 4) Routines & Strategies: Promoting developmental activities during wake windows and longer consolidated sleep at night as the infant grows.

Conditional Weight Gain scores will be calculated using length and weight of the infants. Length and weight will be measured in triplicate by masked research-reliable trained assessors and with standardized anthropometric procedures and averaged at each time point. Infants will be weighed unclothed, in a dry diaper using a Scale-Tronix 4802D Infant scale that is calibrated every day. Infant length will be measured using a Pediatric Stadiometer by O'Leary (PED LB35-07-X). Conditional weight gain is calculated as standardized residuals from the linear regression of anthropometric data (e.g., weight and length) on birthweight, with age and sex entered as covariates. Scores closer to 0 are ideal, with scores higher than 0 indicating more rapid weight gain and scores below 1 indicating slower weight gain. Primary outcomes will include conditional weight gain scores at the 9-month post-treatment time point.

Conditional Weight Gain scores will be calculated using length and weight of the infants. Length and weight will be measured in triplicate by masked research-reliable trained assessors and with standardized anthropometric procedures and averaged at each time point. Infants will be weighed unclothed, in a dry diaper using a Scale-Tronix 4802D Infant scale that is calibrated every day. Infant length will be measured using a Pediatric Stadiometer by O'Leary (PED LB35-07-X). Conditional weight gain is calculated as standardized residuals from the linear regression of anthropometric data (e.g., weight and length) on birthweight, with age and sex entered as covariates. Scores closer to 0 are ideal, with scores higher than 0 indicating more rapid weight gain and scores below 1 indicating slower weight gain.

Used to measure infant appetite and contains 5 subscales that assess enjoyment of food, food responsiveness, slowness in eating, and satiety responsiveness, and general appetite. Subscales of the BEBQ have good reliability and the measured appetite traits track throughout infancy and toddlerhood and are related to greater weight gain. Items are scored on a Likert scale ranging from 1 to 5 with higher average scores on each scale indicating more enjoyment of food, food responsiveness, slowness in eating, and satiety responsiveness, respectively.

An 83-item measure of parent feeding beliefs, behaviors, and style with the following subscales: laissez-faire, restrictive, pressuring, responsive and indulgent. Internal reliability measures for the subscales ranged from 0.75 to 0.95. Several sub-constructs, responsive to satiety cues, pressuring with cereal, indulgent pampering and indulgent soothing, were inversely related to infant weight-for-length z-score, meaning higher scores indicate greater risk for obesity.

A validated parent-report measure that assesses infant dietary intake within the previous 7 days. The FFQ measures duration of breast-feeding status, timing of introduction of solid foods, consumption of fruits and vegetables, consumption of proteins, consumption of energy-dense-nutrient-poor foods, and consumption of sugar sweetened beverages. The FFQ correlates well with weighted diaries in infancy. It measures intake across a diverse range of liquid and solid food groups. As infants are gradually introduced to solid foods and become less reliant on milk, the completion time of the measure may increase (requires 5-10 mins to complete).

Measures 3 consecutive days of infant feeding, sleeping, and fussy events. The BDD has been extensively validated and administered to families in infant obesity clinical trials. This measure has been adapted to capture frequency (i.e., mean number of feedings per day) and volume of feedings (i.e., mean number of ounces per feeding and solid food consumption) in addition to 24-h sleep frequency and duration across a three day period. Investigators are using the Daily Connect™ app to reduce burden, have the capability to remind parents to complete the logs, and download data more easily. For 3 consecutive days, parents will indicate all infant feeding, sleep, and fussy events. Modified Food Frequency Questionnaire during active treatment: For feeding events, parents will detail whether the infant was fed breast milk, formula, or solids, as well as the number of consumed ounces of formula or breast milk if offered by the bottle.

A 13-item measure assessing parent's use of food to soothe their infant comes from one factor with good internal consistency on The BBNQ asks parents to rate on two different 5-point Likert scales, how likely they would be to use food to soothe in different situations and the perceived effectiveness of using food to soothe in each situation. Higher food-to-soothe scores indicate greater use of food to soothe to quiet, or manage a distressed child in response to a variety of contexts, without regard for whether hunger is the source of infant distress.

A 30-item measure of parenting beliefs and practices reflecting structure (regularity of routines) and attunement (recognition of baby cues). Items ask about sleeping, eating, and crying. Items are rated on a 4-point Likert-type scale ranging from strongly disagree (1) to strongly agree (4). Total scores are derived for structure and attunement with higher scores indicating more structure and more attunement.

A 37-item measure of infant temperament that is scored using 3 subscales: positive affectivity/surgency, negative emotionality, and orienting capacity. Scores range from 0 to 7, with higher scores representing greater positive affect, negative emotionality and orienting and regulatory capacity.

A 20-item norm-referenced and validated caregiver-report measure of 4 subscales: infant sleep, parent perception of sleep, parent behavior, and total score. Scales on each subscale and total score are scaled from 0-100 using norm-based referencing. BISQ-R total score is an average of 3 subscales with higher scores denoting better sleep quality, more positive perception of infant sleep, and parent behaviors that promote healthy sleep.

A 10-item measure of feelings of efficacy in infant care, including soothing, feeding, understanding baby cues, and communicating with baby. Items are endorsed using a Likert scale and a total score is derived. The total score will be used and ranges from 10 to 40 with higher scores representing greater Maternal Self-Efficacy.

A 22-item measure of caregiver experiences of violence and adversity in childhood. It encompasses the original 10 ACEs and other experiences of trauma and violence associated with growing up in an urban community (e.g., discrimination, unsafe neighborhood). Item-level and total ACEs scores (original) and (Philadelphia ACEs) will be examined in analyses.

A 10-item questionnaire was developed to identify women who have postpartum depression and has subsequently been validated as a two-factor measure of postpartum anxiety and depression. The overall assessment is done by total score, which is determined by adding together the scores for each of the 10 items. Higher scores indicate more depressive symptoms.

A 9-item validated rapid assessment parent-report measure assessing movement behaviors in babies who have yet to reach their walking milestone. The MBQ measures tummy time or active play, restrained time, screen time, and sleep. Scores will be calculated for each individual item by determining the total number of minutes ((hours x 60) + minutes) per day that the baby engages in each type of behavior. Higher scores indicate more time in that activity.

At infant age 1 month (baseline, at infant age 9 months (post-treatment), at infant age 12 months (follow-up)

A 47-item validated parent-report measure assessing problematic child mealtime behaviors, use of food as a reward, parental concern about child's diet, and spousal stress related to child's mealtime behaviors. Higher scores indicate more problematic mealtime behaviors, more use of food as a reward, more parental concern about child's diet, and more spousal stress related to child's mealtime behaviors, respectively.

Inclusion Criteria: born 2500 grams or greater delivery occurring between 37 and 42 weeks gestation English speaking infant receiving care provided at our pediatric primary care setting from a racial / ethnic minority group (i.e., non-white, or Hispanic or Latinx) and/or economically marginalized background (i.e., household income at or below 138% of federal poverty level; qualifying for Medicaid) Exclusion Criteria: care in the Neonatal Intensive Care Unit (>7 days) infant congenital anomaly or neonatal condition that affects feeding (e.g., cleft lip/palate, metabolic disease) infant exposure to illicit drugs in utero [with the exception of tetrahydrocannabinol (THC)] diminished or impaired caregiver cognitive functioning family intent to move from the area within 1 year

Prior to sharing, all data will be de-identified in a HIPAA-compliant fashion. Data sets will be carefully reviewed to make sure that information such as age, gender, race, and ethnicity cannot be used to gather additional information that could potentially identify individual subjects. All modalities of data will be shared, including raw and aggregate data. Descriptors for all variables shared will be included to prevent misuse or confusion. Any analytical methods utilized to assess the data will be defined in shared formats. In addition, treatment manuals will also be shared after completion of the trial for future use.