Background: The aim of the present study was to determine the validity of the Brazilian version of the Geriatric Depression Scale (GDS) with 30 (GDS-30), 15 (GDS-15), 10 (GDS-10), 4 (GDS-4) and 1 (GDS-1) items and to calculate the optimum cutoff points for identifying depression among elderly primary care subjects.
Methods: A cross-sectional study was carried out involving 220 elderly patients recruited from four primary care clinics in northeastern Brazil. The following measurements were obtained: sociodemographic variables, Katz scale of independence in activities of daily living, and the GDS with 30, 15, 10, 4 and 1 item(s). A psychiatrist blinded to the results of the GDS applied the mood module of the Structured Clinical Interview for the DSM-IV for the diagnosis of major depressive episodes as the "gold standard."
Results: The use of the cut-off point of 10/11 for the GDS-30 produced sensitivity and specificity rates of 92.0% (95% CI: 70-98) and 79% (95% CI: 73-85), respectively. The positive predictive value (PPV) and the negative predictive value (NPV) were 49% and 98%, respectively. The optimum cut-off point for the GDS-15 was 4/5, at which sensitivity was 87% (95% CI: 71-95) and specificity was 82% (95% CI: 76-91), PPV was 51% and NPV was 97%. At the cut-off point of 3/4 the sensitivity, specificity, PPV and NPV for the GDS-10 were 76% (95% CI: 60-89), 81% (95% CI: 75-87), 46% (95% CI: 33-59%), and 94% (95% CI 89-97%), respectively. The optimum cut-off point for the GDS-4 was 0/1, at which sensitivity was 84% (95% CI: 68-93%); specificity was 75% (95% CI; 68-91%); PPV was 41% and NPV was 96%. For the GDS-1, sensitivity was 47%, specificity was 96%; PPV was 69% and NPV was 90%.
Conclusions: The GDS-30, GDS-15, GDS-10 and GDS-4 proved to be good screening instruments for depression in primary care clinics in Brazil, whereas the GDS-1 failed to perform adequately.