Screening for Depression in Elderly Hemodialysis Patients

<i>Background:</i> Depression is common and associated with increased morbidity and mortality in elderly (≧65 years) hemodialysis patients. Beck’s Depression Inventory (BDI) and the Geriatric Depression Scale (GDS) have been used in different cohorts to screen for depression. <i>Objectives:</i> We aimed to evaluate the 15-item GDS (GDS-15) as such a tool in elderly hemodialysis patients and compare it with BDI, a previously validated tool in younger hemodialysis patients. <i>Design:</i> Cross-sectional study. <i>Setting:</i> Four out-patient hemodialysis units; 1 based in a university hospital and 3 based in the community. <i>Participants:</i> Hemodialysis patients aged 65 years and older. <i>Intervention:</i> Both tools were administered to all participants, and a geriatric psychiatrist blinded to the results evaluated them for depression by the gold standard psychiatric interview. <i>Measurements:</i> The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for both tools were assessed against the psychiatric interview (n = 62). <i>Results:</i> Patients who were depressed according to the psychiatric interview had significantly higher GDS-15 and BDI scores compared to those not depressed (p < 0.01 both). ROC curves showed high predictive accuracy of the GDS-15 and BDI (area under the curve: 0.808 and 0.729) versus the psychiatric interview. The GDS-15 cutoff with the best diagnostic accuracy was 5 with a sensitivity of 63%, specificity of 82%, PPV of 60% and NPV of 83%. The BDI cutoff with the best diagnostic accuracy was 10 with a sensitivity of 68%, specificity of 77%, PPV of 57% and NPV of 85%. <i>Conclusion:</i> These results provide evidence that the GDS-15 shows validity in comparison to a gold standard and can be used to screen for depression in the elderly hemodialysis population.