The Geriatric Depression Scale, Short Form (GDS-15) is a widely-used depression rating scale for elderly adults. It might be useful for persons across the adult lifespan, but more research is needed to support its clinical utility with young and middle-aged adults.
We examined the classification accuracy of the GDS-15 in identifying depression cases and non-cases in adults aged 18-54 (n = 199) compared to those aged 55-80 (n = 112), using the standard cutoff score of 5. Criterion-related validity of the GDS-15 was examined based on its chance-corrected agreement with a clinical diagnostic interview.
Classification accuracy based on receiver operating characteristic (ROC) analysis was strong in younger (area under the curve; AUC = 0.92) and older adults (AUC = 0.94). Sensitivity and specificity of the GDS-15 for identifying depression were 72% and 97% for younger adults and 86% and 91% for older adults, respectively. Classification accuracy did not differ between age cohorts (z = 0.74, p = 0.46). Chance-corrected agreement (kappa) between the GDS-15 and the criterion was 71% for younger and 74% for older adults.
Analyses are based on a convenience sample aggregated from three community mental health studies. Minor procedural inconsistencies may be present. Group sizes were uneven and accentuated cell size differences in the confusion matrices.
The GDS-15 is brief depression rating scale that shows good diagnostic sensitivity and specificity for adults aged 18 and older.