The aims of this study were to examine the psychometric properties of The Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) in psychiatric inpatients, due to the scarcity of screening instruments validated in this population. Patients from Hospital Clínic's psychiatric ward (n = 202) completed: ASSIST, Addiction Severity Index (ASI), MINI-International Neuropsychiatric Interview (MINI), Alcohol Use Disorders Identification Test (AUDIT), Fagerström Test for Nicotine Dependence (FTND), Severity of Dependence Scale (SDS), and Drug Abuse Screening Test (DAST). Reliability and validity evidences based on internal structure (Exploratory and Confirmatory Factor Analyses) and on the relation to other variables were obtained. Excellent internal consistency was found for Total Substance Involvement (TSI) (α = .92 and ω = .93) and for Specific Substance Involvement (SSI) scores (α = .88 - .96 and ω = .89 - .95). Analysis of internal structure for tobacco, alcohol and cannabis subscales resulted in unidimensional models with adequate goodness-of-fit indices. ASSIST scores were significantly correlated with those of ASI (r = .795 to r = .953), AUDIT (r = .864), FTND (r = .808), DAST (r = .831), SDS (r = .519) and with "number of diagnoses of abuse/dependence" in MINI-Plus (TSI: r = .857 to r = .862; SSI: r = .646 to r = .834). Receiver operating characteristic analysis (ROC) and Mann-Whitney's U test found good discriminative validity evidences. ASSIST scores showed good reliability and there were validity evidences that support its use for identifying risk levels of tobacco, alcohol and other substance use in psychiatric patients.
Los objetivos fueron examinar las propiedades psicométricas de la prueba de detección de consumo de alcohol, tabaco y sustancias (ASSIST) en pacientes con trastorno psiquiátrico. Un total de 202 pacientes ingresados en psiquiatría del Hospital Clínic completaron: ASSIST, Índice de gravedad de la adicción (ASI), MINI-Entrevista Neuropsiquiátrica Internacional (MINI), cuestionario de identificación de los trastornos debidos al consumo de alcohol (AUDIT), Test de Fagerström (FTND), Escala de gravedad de la dependencia (SDS) y Prueba de detección de abuso de drogas (DAST). Se obtuvieron la fiabilidad y evidencia de validez de la estructura interna (análisis factorial exploratorio/confirmatorio) y de la relación con otras variables. Se encontró excelente consistencia interna en puntuaciones de riesgo total (TSI) (α = ,92 y ω = ,93) y de cada sustancia (SSI) (α = ,88 - ,96 y ω = ,89 - ,95). La estructura interna de tabaco, alcohol y cannabis resultó en modelos unidimensionales con índices de bondad de ajuste adecuados. Las puntuaciones del ASSIST correlacionaron significativamente con: ASI (r = ,795 a r = ,953), AUDIT (r = ,864), FTND (r = ,808), DAST (r = ,831), SDS (r = ,519) y «número de diagnósticos de abuso/dependencia» en MINI-Plus (TSI: r = ,857 - ,862; SSI: r = ,646 - ,834). El análisis de curva ROC y U de Mann-Whitney mostraron evidencias de validez discriminativa. Las puntuaciones del ASSIST tienen buena fiabilidad y existen evidencias de validez para su uso en la detección del nivel de riesgo de consumo de tabaco, alcohol y sustancias en pacientes con trastorno psiquiátrico.