26 June 2007
Aims: We studied the relationship between the Beck Depression Inventory (BDI) and the 15-item Geriatric Depression Scale (GDS-15) in young–old hemodialysis and hospitalized patients in order to evaluate the possible usefulness of GDS-15 in hemodialysis patients. Methods: Thirty-one hospitalized and 31 young–old hemodialysis patients aged 65–74 (young–old) were enrolled in the study. Comprehensive geriatric assessment (Mini Mental State Examination (MMSE), BDI, GDS-15, Cumulative Illness Rating Scale (CIRS) and Activities of Daily Living (ADL)) was made for all patients. The internal consistency between BDI and GDS-15 was evaluated with Cronbach’s α coefficient. Sensitivity, specificity and receiver operating characteristic (ROC) curves for GDS-15 were determined using BDI as the standard. Results: In the hospitalized group, the prevalence of depressive symptoms, as evaluated by BDI (≧14) and GDS-15 (≧6), were 29 and 32%, respectively. In the hemodialysis group, the prevalence of depressive symptoms, as evaluated by BDI and GDS-15, were 61 and 58%, respectively. A significantly positive correlation between the BDI and GDS-15 was found in hospitalized (r = 0.808; p < 0.001), hemodialysis (r = 0.692; p < 0.001) and both patient groups together (r = 0.777; p < 0.001). The area under the ROC curve was 0.99 in the hospitalized and 0.95 in the hemodialysis groups. The ROC curves indicate a best effectiveness cutoff point (balancing sensitivity and specificity) of ≧6 for GDS-15 compared to BDI. Conclusions: The GDS-15 could be a useful instrument for evaluating depressive symptoms in young–old hemodialysis patients.