Figure 1 illustrates the study timeline. We will employ a single-blind 2-group randomized control design (treatment and control) with randomization occurring after baseline in the spring/end of the school year (Time 0) and 3 postintervention evaluation periods: immediately after intervention (end of summer; Time 1), 9 months after intervention (spring; Time 2), and 12 months after intervention (summer; Time 3). Because the primary objective of the i♥rhythm project is to prevent children from beginning a trajectory toward overweight/obesity in elementary school, we will explore the impact of the intervention on change in BMI during the following summer (Times 2-3), identifying whether a maintenance intervention is needed. Following the intervention (Time 1) and final data assessment (Time 3), qualitative interviews will explore the acceptability of the intervention, barriers, facilitators, difficulties with study procedures, maintenance of improvements, and self-efficacy to maintain improvements.

The i♥rhythm project is intended for children between the ages of 5 and 8 and their parents. This age group was selected because by using group-based trajectory modeling, we identified that about 19% of children began a trajectory toward overweight/obesity beginning the summer after kindergarten and the summer after second grade [28]. To be eligible for participation parent-child dyads must meet the following criteria: (1) index child must be between 5 and 8 years old and enrolled in kindergarten, first, or second grades; (2) index child must have a BMI >50th percentile (BMI>50th percentile in elementary school is associated with greater risk of transitioning to an unhealthy weight status or obesity at age 12 [29]); (3) parent identifies as a daily user of social media and willing to engage in a social media intervention on Facebook; (4) parent owns a smartphone; (5) parent is comfortable participating in the intervention and responding to questionnaires in English; (6) families must live within the Greater Houston area and be willing to attend 4 in-person assessment visits at the Children’s Nutrition Research Center (CNRC). Exclusion criteria include parent report that the index child (1) has been diagnosed with a chronic medical condition that influences sleep, eating behaviors, weight status, or circadian rhythms; (2) is being treated with a medication or supplement known to affect sleep, such as melatonin or stimulants; (3) has been retained 2 or more grades for academic reasons or has intellectual difficulties that would influence their ability to complete questionnaires or participate in interviews; (4) has participated in an obesity prevention or obesity treatment program in the last 6 months.

Sample size recommendations for pilot studies vary, with some recommending 30 per group and others suggesting a few as 12 per group [30-33]. Our intention at the outset of the trial was to recruit 30 parent-child dyads per group. To account for an attrition rate of up to 20% [34,35] we aimed to recruit 36 parent-child dyads per group. The trial was intended to begin in the Spring of 2020, but due to the COVID-19 pandemic, the start was delayed by 1 year. Because of a delay in the trial start date, our recruitment goals were amended. Given the shortened study timeline, the new recruitment goal was to recruit a sample size of 16 parent-child dyads per group. Accounting for an attrition rate of up to 20%, this established a new recruitment target of 20 parent-child dyads per group (40 total parent-child dyads).

Process evaluation documents intervention delivery and provides insight into the mechanisms by which programs work [44]. Following the framework of Baranowski and Jago [45], we will assess recruitment of participants, maintenance of participation, the context of implementation, resources necessary for implementation, implementation (completion of sessions, daily diaries, and engagement in social media), reach, barriers to implementation, and exposure to the program. Participant completion of sessions will be monitored via logins, video views, and completion of interactive content and diary data. Engagement in social media will be collected via manual abstraction from the newsfeed of the private group [46]. Other issues such as acceptability will be assessed by questionnaires after intervention. Postassessment interviews will examine (1) acceptability of the interventions, (2) perceived benefits of participation, (3) barriers to participation, (4) reasons for drop out. Following the intervention (Time 1) and final data assessment (Time 3), qualitative interviews will be conducted. Interviews will use a standardized script, open-ended questions, follow-up questions, prompts, and areas to probe. Interviews will be conducted until theoretical saturation is achieved [47,48]. Thematic data analysis will identify common themes and patterns [49]. Each statement will be assigned a broad category and assessed for meaning units, specific categories, and subcategories. These will be compared and contrasted by gender, overweight status, and SES level.

Participant completion of sessions will be monitored by an outcome assessor (HD) who will send SMS text message reminders to the family if they did not complete a session within 2 days after the release of data. Families who miss a session will be sent the next session the following week. Facebook post views and engagement will also be monitored using metrics provided by Facebook.

Two weeks before the scheduled assessments, actigraphs will be mailed to the child’s home. A link to an online instructional video will demonstrate proper wear and how to avoid covering the accelerometer with clothing [50]. Children will wear actigraphs (GT3X-BT; ActiGraph, LLC) on the wrist of their nondominant hand to assess their sleep and activity for 7 days and 8 nights during the school year. Parents will complete a daily sleep diary and monitor-wear logs to record the times the accelerometer is removed and the reason for removal [51-53]. A paper copy of these forms will also be provided for note-taking purposes. Parents will download an app called Centrepoint (ActiGraph, LLC) to their phone and will be asked to use this app to upload their child’s actigraph data to the cloud via a Bluetooth connection. Download instructions will be emailed to participants and research staff will follow-up to provide a PIN. During this follow-up, the parent will perform a preliminary upload to verify data are transferring. The Centrepoint app will allow research staff to monitor participants’ compliance with accelerometer wear and ensure sufficient wear time is achieved before coming into the sessions. Sufficient wear time will be defined as having at least five nights [51] and days of activity data. Valid days will be defined as at least ten hours of wear time in 24 hours [54]. As sleep will not be scored until after the actigraphs are returned, this determination will be made by research staff using visual inspection of the Centrepoint data. During the 8 days when the monitor is worn, parents will be asked to complete a web-based dietary food recall using the Automated Self-Administered 24-Hour Dietary Assessment Tool (ASA24) 3 times and complete a series of questionnaires (described below). Completion of the ASA24 will be monitored by the research staff. If the online diary status shows as incomplete, the parent is emailed the next day to complete another day of intake until 3 full days are complete. The aim is to obtain dairy intake data for 2 weekdays and 1 weekend day.

Approximately 1-2 days following the eighth day of accelerometer wear, assessment visits will be scheduled. Participants will be scheduled to arrive at the laboratory at least six hours before their typical bedtime. Typical bedtime will be estimated using visual inspection of the actigraphy data in Centrepoint along with parent sleep diaries. On the day of the laboratory visit, participants will be asked to avoid intake of caffeine, chocolate, nonsteroidal anti-inflammatory drugs, and cannabidiol products. Upon arrival at the laboratory, anthropometric assessments will be conducted and actigraph data will be downloaded. Children and their parents will be taken to a dimly lit (<5 lux) circadian phase assessment suite (a private room with a table and chairs and a futon chair with an adjoining bathroom). Following established procedures with children, saliva (~1 mL) samples will be collected using untreated Salivettes (Starstedt, Inc.) every hour beginning 5 hours before and ending 1 hour following typical bedtime. Before the collection of samples, children will be seated for 10 minutes to minimize postural effects on melatonin concentration. If participants eat or drink before the sample collection, they will gently brush their teeth with a soft-bristled toothbrush and water. Saliva samples will be centrifuged and frozen for later analysis of melatonin. Assessment visits will end after the final saliva sample collection.

To compensate participants for travel and the inconveniences associated with assessments requiring a minimum stay of 7 hours in the laboratory and keeping children awake past their bedtime at each assessment, participants will be offered US $100 at Times 0 and 1 and US $150 at Times 2 and 3 to encourage participation in assessments.

A manual of procedures, including protocols related to the collection of each measure and storage, was developed at the beginning of the study. In 2021, before the beginning of data collection, an in vivo training session was scheduled and all data collectors took turns practicing all assessment protocols, and feedback was provided by the PI (JPM) and senior coordinator (HD). The manual of procedures will be maintained and continually updated with input from all related study coordinators. Weekly meetings will be held with the PI and study coordinators to review recruitment, study progress, and data storage and management.

A statistical analysis plan will be written before the database is locked and breaking the blind. The feasibility of the intervention will be assessed from staff-collected and parent-reported variables according to the previously stated criteria. We will examine the internal consistency of all questionnaires using Cronbach alpha. Decisions regarding the need for changes before further testing of the intervention will be made based on the attainment of feasibility criteria and postassessment interviews. This pilot study is not powered to detect differences in outcome variables and therefore drawing conclusions about the efficacy of the intervention based on statistically insignificant differences in change in BMI, standardized BMI, or circadian phase is inappropriate [30,77-79]. Descriptive statistics and confidence intervals will be used to examine mean differences and the effect sizes of change in outcome variables between conditions. Exploratory analyses will be used to estimate intervention effect over time using a repeated-measures regression approach while controlling for sex and SES, and accounting for the within-subject correlation and the nesting of children within families. The model will include a fixed effect for the repeated measurements (Times 0, 1, 2, and 3), a random subject effect, and treatment group as a between-subjects factor (i♥rhythm and control). Children’s differences across the groups in change in BMI, standardized BMI, and circadian phase will be tested using a group × time interaction. Change in BMI was included as an outcome because it is considered the preferred measure of change in adiposity among children over periods of less than 1 year [80-82]. Separate models will examine other dependent variables of interest including sleep duration, the timing of sleep onset, and interdaily similarity. Because the study is not powered to detect group differences in outcomes, effect sizes will be examined based on the standardized mean difference (Cohen d) criteria for small (0.20), medium (0.50), and large (0.80) [83,84].

This study was approved by the institutional review board at Baylor College of Medicine (approval number H-47369).

Summer is a time during which children’s risk of obesity increases substantially [28,87]. Disrupted sleep patterns and behavioral rhythms, which may occur in the transition from the school year to summer vacation [6,15,16], have been implicated in the onset of obesity [7-9]. IPSRT is an evidence-based treatment for bipolar disorder that has demonstrated efficacy to prevent the reoccurrences of mood disturbances by preventing the misalignment of endogenous circadian rhythms through the promotion of stable behavioral rhythms (eg, sleep and meal patterns) [20]. The i♥rhythm project is an adaptation of IPSRT as an mHealth intervention to promote stable behavioral rhythms in children during summer vacation for the prevention of summer increases in BMI.

This study is a randomized controlled trial designed to evaluate the feasibility, acceptability, and preliminary efficacy of the i♥rhythm project to prevent accelerated summer weight gain. It is hypothesized that a priori feasibility criteria will be met and the intervention and assessment protocol will be found to be acceptable by participants. As this is a feasibility study, it is not powered to detect differences between groups; however, we anticipate that participants in the treatment condition will have circadian phases that trend toward being earlier and changes in BMI will tend to be smaller compared with participants in the control condition.

Strengths of this study include a rigorous design involving randomization, a comparator condition, and multiple follow-up assessments. Our hypotheses and outcomes have been stated a priori and criteria for evaluating the feasibility and acceptability of the study have been established. Additionally, procedures for process evaluation, intervention fidelity, and assessment of intervention adherence have been developed.

This is a feasibility study and is not powered to detect differences in circadian phase, sleep, or BMI outcomes. If deemed feasible in this study, a fully powered randomized controlled trial will be needed to determine the efficacy of this approach. The current research is being conducted during the COVID-19 pandemic. Due to the halting of new human subjects research in the spring of 2020, the trial onset was delayed by 1 year. The initial phase of recruitment (March-May 2021) began at the end of the second COVID-19 wave (2021) and Time 1 (postintervention assessments) ended just as the Delta wave was beginning in August 2021. For many children, the 2020-2021 school year may not have been typical. Many children may not have attended school in person, potentially weakening the effect of school-summer differences. The 2021-2022 school year has been more typical in the sense that most children returned to school in person. While there is evidence that the pandemic has exacerbated the obesity epidemic in children [88,89], the extent to which patterns of improvement in school-year weight outcomes and accelerated summer weight gain have persisted during the pandemic is unclear. Vaccines became available to 5-8-year olds in November of 2021. The next phase of recruitment will begin as the Omicron-fueled wave appears to be declining (February 2022). It is unclear the extent to which the pandemic has affected our ability to recruit and retain families in the study.

This study seeks to prevent child obesity by combining 2 distinct intervention concepts (enhanced sleep and stable behavioral rhythms) that have previously not been combined for the prevention of childhood obesity. This represents a departure from traditional obesity prevention approaches that have focused on the simple energy balance of diet and physical activity [5]. The proposed research aims to expand our understanding of the role of sleep and behavioral rhythms in the prevention of childhood obesity and adapting IPSRT for children should result in an innovative evidence-based approach to promoting stable behavioral rhythms for obesity prevention. In addition, this study focuses on summer, a time when young children experience significant increases in BMI [90,91]. To our knowledge, the proposed research is the first to (1) provide preliminary evidence regarding the impact of an intervention for the prevention of increases in BMI during summer and the longer-term impact on weight status, (2) combine 2 distinct theories regarding the obesogenic role of sleep and behavioral rhythms among children, and (3) examine a novel adaptation of an evidence-based treatment of bipolar disorder for the prevention of childhood obesity.