Beck Depression Inventory (BDI-IA) adapted for HIV: Psychometric properties, sensitivity & specificity in depressive episodes, adjustment disorder & without symptomatology

Abstract

Introduction. The Beck Depression Inventory (BDI-IA) is the most widely used instrument for assessing depression symptoms. Although it has been validated in the Mexican population, it has not been tested in people living with HIV (PLWH), who tend to have symptoms difficult to distinguish from those associated with viral infection. Objective. We obtained the psychometric properties, sensitivity, specificity, and cut-off points to distinguish between a depressive episode, adjustment disorder and no symptoms. Method. Prospective study with 2,022 PLWH (88% men), who completed the BDI-IA between 2016 and 2017. Subjects had a mean age of 31.9 ± 9.3 years, with 12.0 ± 5.6 years of schooling, and 4.5 ± 4.3 years since diagnosis. The differentiation of items, internal consistency, factor analysis, and calculation of sensitivity and specificity were tested. Results. A Cronbach’s alpha coefficient of .91 was obtained. Through factorial analysis with orthogonal rotation (average intercorrelations r = .40, KMO .929), we obtained three factors: general factor of depression, somatic, and cognition, which explained 39.7%, 6.01%, and 5.49% of the variance, respectively. Only the items in the first factor (the short version with 12 items) were tested. With a cut-off point of 11, it had 85.5% sensitivity and 76% specificity [(AUC) = .865, 95% CI [.83, .90], p ≤ .001], and distinguished major depressive disorder from cases without mental symptoms. Discussion and conclusion. We show that the short version of the BDI-IA is reliable, valid, sensitive, and specific for evaluating depression symptoms comorbid with HIV infection.