Psychometric and diagnostic properties of the Drug Abuse Screening Test (DAST): Comparing the DAST-20 vs. the DAST-10

Abstract

Abstract Background. The use of reliable and valid self-report questionnaires to identify drug use disorders (DUD) is a strategy that has shown usefulness for screening. One of the instruments more used for detection is the Drug Abuse Screening Test (DAST). The psychometric properties in the 20- and 10-item versions have been evaluated in other countries but in Mexico the psychometric and diagnostic properties of both versions are yet to be evaluated. Objective. The purpose of this study was to evaluate the psychometric and diagnostic properties of DAST-20 and -10. Method. The sample included 565 participants receiving care in addiction residential centers. The DAST-20 was used as a measure to screen for DUD, and the Mini International Neuropsychiatric Interview 5.0 was used as “gold standard” for the DUD diagnosis. Cronbach’s α and CFA were estimated in order to evaluate the psychometric properties. The Receiver Operator Characteristic (ROC) analysis was used to examine the diagnostic properties of each version. Results. Both versions obtained a Cronbach’s α ≥ .80, an optimal goodness of fit for the one factor model and Areas Under the Curve ≥ .90 (95% CI 87-93) for both versions. Discussion and conclusion. DAST-20 and -10 versions are reliable and valid tools for DUD assessment and screening.