Factor structure of the BDI-II in Parkinson's disease.

Abstract

There is substantial heterogeneity in depressive symptomology for individuals with Parkinson's disease (PD). It is unknown whether the Beck Depression Inventory-Second Edition (BDI-II) is capable of identifying such phenotypic variations of depression. We investigated the factor structure of the BDI-II and its associations with demographic characteristics and other nonmotor symptoms in PD. We reviewed the cases of 236 patients with a confirmed PD diagnosis. Evaluations included the BDI-II, Montreal Cognitive Assessment (MoCA), Apathy Scale (AS), and Geriatric Anxiety Inventory (GAI). We used exploratory structural equation modeling (ESEM) with target rotations as this method integrates aspects of exploratory and confirmatory factor analysis. We conducted hierarchical regressions to assess for associations between the BDI-II factors and gender, age, education, disease duration, cognition, anxiety, and apathy. ESEM supported the retention of a Somatic factor and an Affective factor that accounted for 53% of the model variance. Model goodness-of-fit measures were within normal limits. Higher AS scores were positively associated with the Somatic and Affective factors. Higher GAI scores were positively associated only with the Affective factor. There were no other significant relationships with factor scores. This study supports the retention of a two-factor model of the BDI-II in PD. These unique clusters of depressive symptoms in PD can be used to guide clinical decisions about the need for further psychiatric evaluation and the appropriateness of different therapeutic interventions. (PsycInfo Database Record (c) 2021 APA, all rights reserved).