Reducing Challenging Behaviors in Children With Autism Through Digital Health

Investigating the Impact of an Emotional Stress Alarm App (KeepCalm) on Managing Stress and Preventing Challenging Behaviors in Children With Autism

School-based behavioral approaches to managing challenging behaviors in children with ASD are limited by three key factors: 1) children with ASD often have difficulties communicating their emotions; 2) it is challenging to implement evidence-based, personalized strategies for individual children, and; 3) it is difficult for teachers to track which strategies are successful for individual children. The investigators' personalized mobile-health emotion regulation application (m-health app) will pair heart rate tracking with digital tools to help reduce challenging behavior by supporting stress detection, reminding teachers of specific behavioral strategies and helping teachers to track progress.

The project will develop and pilot test a personalized medicine mobile health emotion regulation application (the m-health KeepCalm app) that incorporates physiological stress measurement to support evidence-based practices for reducing challenging behavior in children with autism spectrum disorder (ASD). As much as 80% of children with ASD exhibit challenging behaviors that can have a devastating impact on personal and family well-being, contribute to teacher burnout and require frequent hospitalization. Evidence-based practices for reducing these behaviors emphasize uncovering triggers, yet parents and teachers often report that challenging behaviors surface without warning. Challenging behaviors caused by emotion dysregulation can be the most difficult to predict, as children with ASD often have difficulty communicating their distress before it results in challenging behavior. Exciting recent advances in digital technology now allow measurement of momentary emotion dysregulation, using physiological indices. The investigators' pilot data from four separate samples demonstrate that increased heart rate predicts onset of challenging behavior in children with ASD. In order to tailor the m-health KeepCalm app to end users and avoid potential barriers to its adoption, in Aim 1, the investigators will evaluate the acceptability, feasibility, and appropriateness of app, and the needs of educational teams in managing stress in children with ASD and challenging behaviors, by conducting interviews with teachers of children with ASD, parents of children with ASD and school administrators, and conducting structured in-class observations with teachers. Through the activities of Aim 2, the researchers will improve the m-health KeepCalm app, building on their initial app prototype, in collaboration with their established research partner, the Translational Informatics Unit, Children's Hospital of Philadelphia, and their established community partner, the School District of Philadelphia. The investigators will do this through 1) exploratory work on the specificity of heart rate increase to challenging behavior, on app clinical decision support timing, and on the association of app false positives and negatives to movement or child factors; 2) monthly advisory board meetings with expert stakeholders for app development guidance, and; 3) rapid-cycle prototyping of the app with 10 educational teams (i.e. 1-2 children with ASD, and their teacher and classroom aide, if they have one). This will allow for iterative improvement based on each user's experience. Through Aim 3, the researchers will test the app for usability, acceptability, feasibility and appropriateness, as well as preliminary effectiveness with 30 educational teams in a randomized waitlist field trial over a 3-month period. Successful completion of these aims will result in a novel m-health app designed to help teachers support emotion regulation, and reduce or prevent challenging behavior in children with ASD, using evidence-based strategies. These activities will lay the foundation for an R01 to evaluate the effectiveness of the m-health KeepCalm app in a full- scale randomized field trial. This proposal aligns with the strategic plan of the Interagency Autism Coordinating Committee, to maximize the potential for technology-based interventions to improve the lives of people with ASD.

Participants will be grouped into teams (child with autism, their teacher, and 1:1/aide) and teams will be randomized to one of the two arms. Teams (n = 10) randomized to the intervention will be given the KeepCalm application to use for 3 months. Teams randomized to the waitlist control condition (n = 10) will be given access to the app following the study's completion.

The KeepCalm app is designed to help manage stress and prevent challenging behaviors in children with autism. The app uses physiological stress tracking, assessed using heart rate monitoring, to indicate when challenging behaviors are likely to occur. The app then recommends strategies to help prevent the onset of challenging behaviors. The app can also be used to track trends in triggers, behaviors, and strategies, and to communicate this information to parents and other members of a child's educational team.

Individuals assigned to the waitlist control condition will gain access to the KeepCalm app following the study's completion. During the trial, they will not receive any intervention, but will complete baseline and post-intervention measures during the trial.

The KeepCalm app is designed to help manage stress and prevent challenging behaviors in children with autism.

The System Usability Scale (SUS) measures perceptions of usability of a given system, in this case, the KeepCalm app. The scale consists of 10 items, each rated on a 5-point Likert Scale from 1 ("Strongly disagree") to 5 ("Strongly agree"). Responses are transformed to get a total range of possible values from 0-100, with higher scores indicating higher perceptions of usability. Typically, a score above 68 is regarded as a good score.

The Acceptability of Intervention Measure (AIM) is used to measure the acceptability of an intervention. The scale consists of 4 items rated on a 5-point Likert Scale from 1 ("Completely disagree") to 5 ("Completely agree"). The score consists of the sum of the items, ranging from 5 to 20, with higher scores indicating greater acceptability of the intervention.

The Feasibility of Intervention Measure (FIM) is used to measure the feasibility of an intervention. The scale consists of 4 items rated on a 5-point Likert Scale from 1 ("Completely disagree") to 5 ("Completely agree"). The score consists of the sum of the items, ranging from 5 to 20, with higher scores indicating greater feasibility of the intervention.

The Intervention Appropriateness Measure (IAM) is used to measure the appropriateness of an intervention. The scale consists of 4 items rated on a 5-point Likert Scale from 1 ("Completely disagree") to 5 ("Completely agree"). The score consists of the sum of the items, ranging from 5 to 20, with higher scores indicating greater appropriateness of the intervention.

This interview will assess the intervention's usability, appropriateness, feasibility, and acceptability in a qualitative interview format.

The School Situations Questionnaire (SSQ) assessing challenging behaviors within the context of school. The scale assesses both the presence and the severity of challenging behaviors by asking the teacher whether or not a given behavior occurs for that child and to rate the severity. The scale yields two summary scores. The first is the number of problems score which is a sum of the number of problems endorsed, ranging from 0-8 with higher scores indicating greater problems endorsed. The second is the mean severity, calculated as the mean severity for the items endorsed, ranging from 1 to 9 with higher numbers indicating more severe problems.

The Institute for Basic Research Modified Overt Aggression Scale (IBR-MOAS) measures the frequency and severity of aggressive behavior episodes. The scale is divided into four components: verbal aggression, aggression against property, autoaggression, and physical aggression. There are 4 items in each category, scored from 0-4 (with higher numbers indicating greater severity) and raters can select as many items as are relevant in the category. The items are then summed within each category to create a summed category score, and then each category is multiplied by its specified weight to give a total weighted aggression score, ranging from 0 to 100, with higher scores indicating greater problems with aggression.

The Emotion Dysregulation Inventory (EDI) measures emotion dysregulation by asking raters to identify how much of a problem certain emotions and behaviors are for the person being rated. The scale consists of 66 items that are rated for how much of a problem the behavior has been in the last 7 days ("not at all", "mild", "moderate", "severe", "very severe"). This ratings are converted into a numerical rating from 0 ("not at all") to 4 ("very severe"), and then summed to create a total score ranging from 0-264, which higher scores indicating greater emotion dysregulation.

Proportion of time offline mode is activated over the testing week based on accelerometry data threshold

Proportion of alarms/emotion regulation strategy app push notifications that result in timely strategy implementation, as measured during the observation session

Proportion of alarms that appear not to be associated with an oncoming challenging behavior (e.g. due to child movement)/challenging behaviors that occurred without an alarm, measured during the observation session

Inclusion Criteria: Preschool or elementary school child with autism along with a parent and teacher and/or classroom aide Child must have challenging behaviors including but not limited to aggression, escape behavior, loud noises, non-compliance, property destruction, rigid/inflexible behavior, self-injury, and transition difficulties Adult participants must have access to an iPhone in order to test the app or be willing to use a provided study iPhone Exclusion Criteria: Child does not exhibit challenging behaviors