25135
A Smartphone Application to Measure Response to Name in Everyday Environments

Friday, May 12, 2017: 10:00 AM-1:40 PM
Golden Gate Ballroom (Marriott Marquis Hotel)
R. P. Thomas1, L. A. Wang1, J. Miller2, J. W. Pennington2, S. Hassan2, A. Grasmeder2, J. Swanick2, N. Minyanou1 and J. Miller3, (1)Center for Autism Research, The Children's Hospital of Philadelphia, Philadelphia, PA, (2)Department of Biomedical and Health Informatics, The Children's Hospital of Philadelphia, Philadelphia, PA, (3)The Center for Autism Research, The Children's Hospital of Philadelphia, Philadelphia, PA
Background:  Reduced response to name is a hallmark feature of Autism Spectrum Disorder (ASD) but our current methods of identifying it are not very sensitive. Measurement of response to name is often qualitative, relying on the child’s performance during an in-person evaluation (ADOS) or a yes/no response on parent questionnaires (M-CHAT). Measured as a quantitative variable, or as a rateof responding rather than a dichotomous summary judgment, response to name may be both a more sensitive indicator of ASD and a potential moderator of response to treatment.

Objectives:  The objective of the study was to develop a smartphone application to record children’s response to name in everyday environments. We examined fidelity, accuracy of parent ratings, and user experience.

Methods:  Parents of children between the ages of 18-48 months participated in the study. Children could be typically developing, or have ASD or another developmental concern. Parents downloaded the app onto their smartphone and were asked to complete 3 sessions. The app included a built-in tutorial; otherwise no instructions were provided. The tutorial displayed illustrations and text guiding parents to stand 3-5 feet behind the child, call the child’s name once, and then press Yes or No to rate whether the child responded. Within each session, the app prompted up to 3 name call trials in case the child did not respond right away. After 3 sessions, we provided feedback to the parent about fidelity. Then parents were asked to record up to 30 additional sessions over the course of one week. Sessions were automatically uploaded to our database, with both a video file and time-stamped text file of parent ratings. After completing all sessions, parents completed online questionnaires about user experience and satisfaction.

Results:  So far, 4 families have begun participation, and 17 more have expressed interest (study goal is 30 families). We currently have 16 pre-feedback trials and 135 post-feedback trials. Two raters viewed each video recording and text file to establish parent fidelity (the parent’s ability to administer the trial correctly) and accuracy (of child response ratings). Inter-rater agreement was 100% for fidelity and 98% for accuracy. Across participants’ first three sessions, parent fidelity averaged 78% (range 50%-100%). Two families needed feedback to say the child’s name only once, and not to shout the child’s name. Across post-feedback trials, parent fidelity averaged 89% (range=77%-97%). Parent accuracy for rating their child’s response was 97% (range=95%-100%). Parent survey responses indicated parents “strongly agreed” the app was easy to use, that it measured a meaningful behavior, and that they would recommend the app to a friend.

Conclusions:  The app appears easy for parents to use correctly, though instructions do need to highlight calling the child’s name only once and not shouting the child’s name. Ongoing data collection will allow us to also develop formulas for calculating a child’s rate of responding, which can then be examined across children with ASD, other delays, and typical development.