The use of signal detection theory to assess DSM-III-R criteria for autistic disorder.

Sixty cases from the DSM-III-R field trials for autistic disorder (AD) were studied to derive an optimally sensitive and specific combination of scoring criteria based on the 16 DSM-III-R criteria for AD. Signal detection theory, a statistical method new to psychiatric research, was applied. Two of the sixteen criteria tested with this methodology yielded a sensitivity of 0.82, a specificity of 0.89, and a total predictive value of 0.91 relative to clinical diagnosis. Excluded criteria were redundant; unrelated to diagnosis; or highly correlated with degree of mental retardation, mental age, and/or chronological age.