Psychometric analysis of the prefrontal symptom inventory abbreviated: evidence of its validity and reliability in the general Venezuelan population
*Correspondencia: Dr. Vicente Cancino. Universidad de La Frontera. Calle Montevideo, 0830. Temuco, Chile. E-mail: vicente.cancino@ufrontera.cl
Introduction: Few tools exist to objectively measure dysfunctions of prefrontal origin self-reported by the general population. The Prefrontal Symptom Inventory (PSI) is a test with excellent psychometric properties that allows such assessment and so far, no robust analysis of its abbreviated version in Spanish for Latin America has been performed.
Aims: To analyze the psychometric properties of the abbreviated PSI in terms of reliability and validity in the general population in the Venezuelan context.
Subjects and methods: 300 subjects from the general population participated. The factor structure of the abbreviated ISP was determined through confirmatory factor analysis (CFA); construct validity was assessed by contrasting groups with no risk of MCI and the convergence of scores with the domains that make up the Montreal Cognitive Assessment (MoCA). Likewise, internal consistency was estimated through McDonald's ? and Cronbach's a.
Results: Five factorial models were contrasted and a version of the PSI composed of 18 items was obtained, which presented excellent indicators of goodness of fit (?2 (132) = 200.057, p < 0.001, CFI=0.955, TLI=0.948, SRMR=0.042, RMSEA=0.041) and internal consistency (? = 0.90; a = 0.89). Likewise, statistically significant differences between groups and inverse correlations were evidenced with the sections evaluated in the MoCA except for abstraction.
Conclusion: The PSI-18 is a valid and reliable measure to be used in the studied population. Consistently, previous studies show its versatility to be used in research and health contexts.
Objetivo Analizar las propiedades psicométricas del ISP abreviado, en términos de fiabilidad y validez en la población general dentro del contexto venezolano.
Sujetos y métodos Participaron 300 sujetos de población general. La estructura factorial del ISP abreviado se determinó a través del análisis factorial confirmatorio; la validez de constructo se evaluó a partir del contraste de grupos con/sin riesgo de deterioro cognitivo leve y la convergencia de puntuaciones con los dominios que componen la prueba cognitiva de Montreal (MoCA). Asimismo, se estimó la consistencia interna través de la omega de McDonald y el alfa de Cronbach.
Resultados Se contrastaron cinco modelos factoriales y se obtuvo una versión del ISP compuesta por 18 ítems, que presentó excelentes indicadores de bondad de ajuste: ?2 (132) = 200,057, p < 0,001, índice de ajuste comparativo = 0,955, índice de Tucker Lewis = 0,948, raíz cuadrática estandarizada de las medias residuales = 0,042, raíz cuadrática de la media del error de aproximación = 0,041; y consistencia interna (omega = 0,9; alfa = 0,89). Asimismo, se evidenciaron diferencias estadísticamente significativas entre grupos y correlaciones inversas con los apartados evaluados en la MoCA a excepción de la abstracción.
Conclusión El ISP-18 es una medida válida y confiable para ser utilizada en la población estudiada. Consistentemente, estudios previos dan cuenta de su versatilidad para que se use en investigación y en contextos de salud.