Leveraging translational insight: Empowering preschool in-service teachers to identify children with autism spectrum disorder (ASD) in classrooms

Abstract

<p>Preschool teachers, with specialized training and opportunities to regularly observe children's peer interactions, are well-positioned to recognize symptoms of autism spectrum disorder (ASD). However, they often hesitate to communicate their concerns to parents. To aid in this process, the Classroom Observation Scale for Autism in Preschool (COSA-P), formerly known as the Classroom Observation Scale (COS), was developed as a screening tool for identifying ASD in preschool children (Au et al., 2021; Shum et al., 2022). The COSA-P consists of 13 items that enable teachers and non-clinically trained observers to observe and assess ASD symptoms in young children. Shum et al. (2022) found that non-clinically trained research assistants exhibited greater accuracy in using the COSA-P for ASD screening compared to Chinese-speaking preschool teachers.</p><p>This study aimed to enhance preschool teachers' accuracy in using the COSA-P for early ASD identification. It involved 1114 preschool children aged 2-6 and 161 teachers from 14 regular preschools. Teachers in the intervention schools received training with high metacognitive engagement (more group discussions and case illustrations), while teachers in the control schools received a didactic form of training. Teachers in the intervention schools reported significantly higher levels of perceived mastery of ASD knowledge (<em>F </em>= 20.66, <em>p </em>< .001,<em> η² </em>= .03), belief in the importance of early identification (<em>F </em>= 11.56, <em>p </em>< .001, <em>η² </em>= .02), self-efficacy in pedagogical accommodation (<em>F </em>= 11.10, <em>p </em>< .05, <em>η² </em>= .01), and motivation in using COSA-P (<em>F </em>= 11.12,<em> p </em>< .05, <em>η² </em>= .01).</p><p>Subsequently, 102 screen-positive children and 93 randomly selected typically developing peers were assessed using the Autism Diagnostic Observation Schedule (ADOS-2) two months later. According to the receiver operating characteristic curve analysis, the intervention group had a significantly higher Area Under Curve (<em>AUC </em>= .86) than the control group (<em>AUC </em>= .71), indicating a stronger predictive accuracy of COSA-P for the intervention group. This research underscores the transformative potential of in-service teacher training, by improving beliefs, self-efficacy, and accuracy in using screening tools for early ASD identification. It also highlights the positive impact of effective in-service training, empowering educators and improving outcomes for children with ASD in educational settings.</p>