Arkansas and Louisiana have among the highest rates of adult obesity in the United States at 38.7% and 38.6%, respectively. Prevention and intervention efforts are needed to reduce the number of children who will become obese adults and suffer the host of negative health consequences that accompany it. This proposal will develop and test strategies to stop the use of detrimental feeding practices by early childhood educators which promote unhealthy weight trajectories, inappropriate eating behaviors and poor dietary outcomes for children.

The long-term goal is to accelerate the implementation of evidence-based feeding practices in ECE settings that support healthy weight trajectories and diets for young children. The overall objective for this application, which is in response to PAR-22-105, is to determine the effectiveness and associated mechanisms of a package of deimplementation strategies to reduce the use of inappropriate feeding practices in ECE. The co-designed package of de-implementation strategies in this project will test includes six complimentary strategies. Three are foundational strategies that all educators will receive: training, a peer learning collaborative, and facilitation; three strategies are tailored to each educator's preferences for support and individual goals: audit and feedback, reminders, and educational resources. These six strategies will simultaneously support the removal of inappropriate practices and the uptake of alternative EBPs. To accomplish this goal, the strategies will target trust as a key mechanism for change, consistent with the Implementation Trust Building Theory of Change.

The rationale for the proposed project is that demonstrating the effectiveness of strategies for reducing inappropriate feeding practices in ECE and determining how the strategies work and for whom will provide new opportunities for their continued development and dissemination. In turn, these strategies will contribute to a strong foundation for lifelong healthy eating habits for children being served in ECE. This project includes a cluster-randomized Hybrid Type III trial with 88 ECE sites serving children ages 3 and 5. The trial will achieve two specific aims:

Aim 1: Determine the effectiveness of a package of deimplementation strategies for reducing the use of inappropriate feeding practices in ECE. The hypothesis is that, compared to usual practice, the deimplementation strategies will be effective at reducing the total number of observed inappropriate feeding practices per meal, our primary deimplementation outcome (N= 88 sites, 528 educators). This project will also identify effects on secondary effectiveness outcome measures including child body mass index, willingness to try new foods, fear of new foods (i.e., neophobia), and dietary intake (N=1320 children).

Aim 2: Identify mechanisms of deimplementation using a mixed methods approach. The hypothesis is that factors specified by the Implementation Trust Building Theory of Change22 will mediate the effect of strategies on deimplementation outcomes. Using an explanatory, sequential design, we will collect qualitative data with purposively selected educators (N = 32) to identify emergent mediators and moderators.

Inclusion Criteria:

• Sites within a 100-mile radius of staff offices serving 12+ children ages 3 to 5 years
• Agreeing to participate in data collection
• Sites will serve meals.
• All classrooms at a site will receive the same deimplementation strategies and participate in data collection. We will select one classroom at random per site to participate in the collection of child outcomes (N= 264 classrooms, 15 children per classroom= 1320 children total).

Exclusion Criteria:

• We will exclude sites that have participated in intervention studies with our team in the last 3 years.