85 - Teaching Primary Care Clinicians to Diagnose Young Children at High Risk for Autism on a National Scale: Feasibility and Acceptability of a Multi-Site Training Model

Publication Number: 1077.85

Holly Harris, Baylor College of Medicine, Houston, TX, United States; Nikki Gambhir, Baylor College of Medicine, HOuston, TX, United States; Gabriela V. Miller, Boston Children's Hospital, Brookline, MA, United States; Julie O. Bickel, Boston Children's Hospital, Ashland, MA, United States; Sarah Canale, Boston University School of Medicine, Winchester, MA, United States; Bridget Poznanski, Boston University School of Medicine, Boston, MA, United States; Marilyn Augustyn, Boston University School of Medicine, SOmerville, MA, United States; Katherine Pawlowski, Boston Children's Hospital, Boston, MA, United States; Carol Weitzman, Boston Children's Hospital, Boston, MA, United States

Background: Training primary care clinicians (PCCs) to diagnose young children at high risk for autism within the medical home can improve early access to intervention. Several groups have developed educational programs for this purpose, yet the feasibility and acceptability of scaling such training across multiple geographic sites are not well established.

Objective: To determine the feasibility and acceptability of implementing an educational program to train PCCs in the diagnosis of autism in high-risk children aged 18 months to 3 years across three academic medical centers.

Design/Methods: PCCs were recruited from three academic sites to participate in an 8-session, 12-hour hybrid training (in-person and virtual) delivered over 8 months. The curriculum included didactic instruction, case review, and supervised practice in completing a Brief Autism Structured Observation in Children (BASIC) and scoring a Childhood Autism Rating Scale-2 (CARS-2). PCCs completed baseline and post-training surveys capturing demographics, practice characteristics, attendance, and satisfaction. Attendance was tracked for each session, and PCCs rated their satisfaction after each session and at the completion of training. Descriptive statistics summarized PCC characteristics, attendance, and satisfaction, and ANOVA examined site differences.

Results: Fifty-six PCCs participated (61% in practice >10 years); 87.5% had never completed an autism diagnostic evaluation before training. PCCs represented urban and suburban practices with varied payer mix. Significant differences across sites were observed in PCC ethnicity and practice size. (Table 1). Overall, 82.4% attended ≥7 sessions, with attendance varying by site (94%, 65%, and 90%, p< 0.05). Fifty-one pediatricians completed the post-training survey (N = 51), with 84% rating the training as excellent or very good, with no significant site differences in satisfaction (Table 2). Most (78%) found the duration of training and number of sessions appropriate, and the highest-rated contributor to satisfaction was instructional quality and instructor expertise (84%).

Conclusion(s): A 12-hour hybrid training program to prepare PCCs to diagnose young children at risk for autism was feasible, well-received, and scalable across diverse academic settings. Overall high satisfaction and moderate to high attendance suggest that a concise, structured program can effectively expand the PCC workforce equipped to conduct autism diagnostic evaluations in early childhood.

Table 1: Primary Care Clinician Characteristics
Sites with significant differences via Tukey's test
*Site 1 - Site 3, Difference between means: 1.56, 95% CI (0.11, 3.01)
**Site 1 - Site 3, Difference between means: 1.00, 95% CI (0.40, 1.60)
**Site 2 - Site 3, Difference between means: 0.83, 95% CI (0.35, 1.41)

Table 2: Primary Care Clinician Satisfaction
*Contributors were check all that apply