Objective: To compare very short scales for screening for depression with longer, widely used scales.
Methods: Eighty-seven patients over the age of 60 who were admitted to rehabilitation wards or were attending a day rehabilitation facility at a British teaching hospital were screened for depression using the 1-item mental health inventory, and the 4-item, 15-item and 30-item geriatric depression scales. The sensitivity, specificity, and areas under receiver operating characteristic curves were compared, with the diagnostic criteria for research of ICD-10 providing the criterion diagnosis of depressive episode.
Results: All the scales had comparable sensitivity (82.4-100%), specificity (60.0-71.4%), and positive predictive values (33.3-42.9%). Comparison of receiver operating characteristic curves for each scale showed no statistically significant difference between them (range 0.80-0.88).
Conclusions: The very short scales performed just as well as the widely used longer screening scales in this population. They are worthy of further examination in elderly populations at risk of depression, and may be particularly suitable for older adults due to their brevity and ease of use.