SCH: Real-Time Engagement of Children for Individualizing Behavior Management with Wearables

About This Grant

Approximately 4.5 million children in the United States (US) display severe emotional and behavioral disturbance. While it is not unusual for preschoolers to have occasional temper tantrums, it is considered symptomatic when temper outbursts (characterized by sudden, violent expression of strong feeling, anger, and aggression) occur most days that are severe enough to impair their academic, social, and family functioning. Evidence-based therapies such as parent child interaction therapy (PCIT) reduce behavioral challenges in children by improving parent-child relationships through parenting practices taught over a multi-week period. Despite the widespread availability of behavioral interventions, there are significant challenges: (a) the effectiveness of interventions is contingent upon parents remembering parenting practices taught during weekly therapy sessions to engage with their children, (b) there is limited education for affected children and their parents to help preempt temper outbursts and regulate their emotion, and (c) families from rural areas and populations with limited specialized pediatric mental health providers are less likely to have access to and utilize evidence-based therapies when it is available. The overarching goal of this convergent research proposal is to investigate the development of generative methods with closed-loop feedback from parents to individualize real-time interventions for children when a temper outburst is predicted. The project will accomplish the goal through the following aims. Aim 1: This project will develop generative algorithms with closed-loop feedback using 5.4 million minutes of smartwatch data collected from 50 children (aged 3 – 7 years) and parent-provided timestamps (closed-loop feedback) of disruptive behavior (characterized by temper outbursts). Aim 2: The developed technology will then be evaluated in a cohort of 50 new children to assess if parenting practices combined with child-initiated mindfulness (i.e., new patient-education) upon a predicted temper outburst could improve behavioral outcomes. Aim 3: Elucidate the perspectives of stakeholders (e.g., parents, schoolteachers) on the use of continuous monitoring devices for adaptive generative intelligence algorithms.