PRISM-5 Update: Adaptation and validation of the Spanish Version Interview for Dual Diagnosis Assessment Based on DSM-5 Criteria.

Introduction: Diagnosing individuals with co-occurring substance use and other mental disorders (SUDs and MDs) is challenging due to overlapping symptoms and masking effects. Semi-structured interviews, like the Psychiatric Research Interview for Substance and Mental Disorders (PRISM), are essential for diagnostic accuracy. This study aims to update and validate the Spanish PRISM-IV to DSM-5 criteria, providing evidence of reliability based on its ability to assess Dual Disorders (DDs).

Material and methods: The PRISM-5 was translated and culturally adapted to Spanish to ensure equivalence with the original English version. The interview was computerized using BLAISE© software and pilot-tested. A cross-sectional study with patients recruited from specialized treatment centers in Spain and Argentina compared PRISM-5 diagnoses to those established through the Longitudinal, Expert, All Data (LEAD) method, providing evidence of diagnostic agreement through Cohen's kappa coefficient.

Results: 197 patients were recruited (69% male, mean age = 35.6). PRISM-5 showed substantial evidence of diagnostic agreement for most SUDs (k = 0.62-0.77), including alcohol, cannabis, cocaine, heroin, sedative, and amphetamine use disorders, with excellent agreement for past heroin use disorder (k = 0.83). For MDs, substantial agreement was found for major depression, ADHD and psychotic, panic, and personality disorders (k = 0.63-0.73), while moderate agreement was observed for substance-induced and persistent depression (k = 0.59-0.60).

Conclusions: PRISM-5 provided strong evidence of diagnostic reliability in its Spanish version, with substantial agreement across most diagnoses. Compared to the PRISM-IV, it showed similar evidence of reliability, with some notable improvements in areas such as alcohol and cannabis use disorders. These findings underscore the instrument's robustness and value for diagnosing DDs.