Cognitive Biases Questionnaire for Psychosis (CBQp): Spanish Validation and Relationship With Cognitive Insight in Psychotic Patients

Abstract

Introduction: Cognitive biases are key factors in the development and persistence of delusions in psychosis. The Cognitive Biases Questionnaire for Psychosis (CBQp) is a new self-reported questionnaire of 30 relevant situations to evaluate five types of cognitive biases in psychosis. In the context of the validation of the Spanish version of the CBQp, our objectives were to (1) analyze the factorial structure of the questionnaire with a confirmatory factor analysis (CFA), (2) relate cognitive biases with a widely used scale in the field of delusion cognitive therapies for assessing metacognition, specifically, Beck's Cognitive Insight Scale (BCIS) (1), and, finally, (3) associate cognitive biases with delusional experiences, evaluated with the Peters Delusions Inventory (PDI) (2).Materials and Methods: An authorized Spanish version of the CBQp, by a translation and back-translation procedure, was obtained. A sample of 171 patients with different diagnoses of psychoses was included. A CFA was used to test three different construct models. Associations between CBQp biases, the BCIS, and the PDI were made by correlation and mean differences. Comparisons of the CBQp scores between a control group and patients with psychosis were analyzed.Results: The CFA showed comparative fit index (CFI) values of 0.94 and 0.95 for the models with one, two, and five factors, with root mean square error of approximation values of 0.031 and 0.029. The CBQp reliability was 0.87. Associations between cognitive biases, self-certainty, and cognitive insight subscales of the BCIS were found. Similarly, associations between total punctuation, conviction, distress, and concern subscales of the PDI were also found. When compared with the group of healthy subjects, patients with psychoses scored significantly higher in several cognitive biases.Conclusion: Given the correlation between biases, a one-factor model might be more appropriate to explain the scale's underlying construct. Biases were associated with a greater frequency of delusions, distress, conviction, and concern as well as worse cognitive insight in patients with psychosis.