Using signal detection methodology to revise DSM-III-R: re-analysis of the DSM-III-R national field trials for autistic disorder

Abstract

Signal detection theory which takes into account the relative prevalence, sensitivity, and specificity of each of a set of criterion symptoms is used to determine an algorithm optimally predictive of a clinical diagnoses of autistic disorder using the data from the DSM-III-R national field trials. Findings support inclusion of one diagnostic criterion (marked lack of awareness of others) as mandatory, and four more (impaired imitation, abnormal social play, abnormal nonverbal communication, and abnormal speech) as alternate, associated criteria. Advantages of signal detection over other statistical methods for empirically deriving diagnostic standards are given, and implications for DSM-IV are discussed.