Objectives: ADDM record review methods are well operationalized and executed but have yet been validated by in-person developmental evaluations. This study is designed to measure sensitivity, specificity, and positive predictive value (PPV) of the ADDM system. This abstract will focus on PPV.
Methods: The Metropolitan Atlanta Developmental Disabilities Surveillance Program (MADDSP) is the gold standard for which ADDM methods are based. For this study, surveillance was conducted on a representative subset of the MADDSP population. A research reliable clinician (blinded to surveillance case status) collected data on the Social Communication Questionnaire (SCQ), Autism Diagnostic Observation Schedule (ADOS), and Autism Diagnostic Interview-Revised (ADI-R) for 177, 9 year olds. Clinical case status was based on results from the ADOS and ADI-R. Surveillance case status was based on an independent review of multiple evaluation records from both educational and clinical sources. Trained clinicians (blinded to clinical case status) with expertise in ASD systematically coded the surveillance records based on the DSM-IV-TR criteria for pervasive developmental disorders to determine surveillance case status.
Results: The study yielded an 85% PPV. The 28 clinically confirmed surveillance cases were male (79%), had recorded IQ < 70 (43%), received educational services or ASD (61%), scored above the SCQ cutoff (93%), and had concordant results on the ADOS and ADI-R (79%).
Conclusions: Clinical evaluations confirmed 85% of surveillance cases were correctly classified given ASD positive ADOS/ADI-R scores. The extent that the confirmed ADDM case profile is representative of characteristics of all identified clinical cases will be examined.