Psychometric properties of the Drug Abuse Screening Test (DAST) in a clinical sample using a network analysis approach

Global drug use has increased substantially in recent years, accompanied by a rise in substance use disorders and related mental and behavioral conditions. Although the Drug Abuse Screening Test (DAST-10) has shown adequate psychometric properties in various countries, no validated version is currently available for the Peruvian clinical context, highlighting the need for robust psychometric tools to assess substance abuse. Therefore, this study aimed to evaluate the psychometric properties of the DAST-10 within the Peruvian context through a network psychometric approach. The sample included 212 service users at addiction rehabilitation centers (159 men and 53 women; mean age = 29.1 years, SD = 9.5), selected by non-probabilistic convenience sampling. Network analysis was used to evaluate the internal structure and reliability of the instrument. The Exploratory Graph Analysis showed that 5 items (1, 2, 6, 8, 9) presented a one-dimensional measure. Reliability, assessed by structural consistency, identified that one dimension was obtained 100% of the time, and items remained stable by consistently replicating within the empirical dimension. This study provides the first psychometric evidence of the DAST-10 in Peru using a network analysis approach, yielding a refined five-item version (DAST-5) with a stable one-dimensional structure, strong reliability, and alignment with DSM-5-TR criteria, highlighting its clinical relevance.