Our study examines the Diagnostic and Statistical Manual-Fifth Edition (DSM-5) and Diagnostic and Statistical Manual-Fourth Edition, Text Revision (DSM-IV-TR) when applied concurrently against the best estimate clinical diagnoses for 110 children (5.1-19.6 years old) referred for diagnostic assessments of Autism Spectrum Disorder (ASD) in a Singaporean outpatient speciality clinic. DSM-IV-TR performed slightly better, yielding sensitivity of 0.946 and specificity of 0.889, compared to DSM-5 (sensitivity = 0.837; specificity = 0.833). When considering the ASD sub-categories, sensitivity ranged from 0.667 to 0.933, and specificity ranged from 0.900 to 0.975. More participants with a PDD-NOS best estimate clinical diagnosis (40%) were misclassified on DSM-5. Merits and weaknesses to both classification systems, and implications for access to services and policy changes are discussed.
Keywords: Autism Spectrum Disorders; DSM-5; DSM-IV-TR; Diagnosis.