Agreement Between the Geriatric Depression Scale and the General Health Questionnaire in a Population-Based Elderly Cohort

Objective: To investigate the agreement between the screening scales General Health Questionnaire (GHQ) and Geriatric Depression Scale (GDS) in a cohort of elderly residents in Bambui, Southeast Brazil. Methodology: The study population included all individuals ≥ 60 years who comprised the baseline cohort in 1997 and who answered the GHQ-12 and GDS-30 items in 2001, the fourth year of follow-up. The prevalences of depressive symptoms were determined for different cut-points and correction methods. The agreement between the scales was measured using the Kappa coefficient. Results: 1,291 individuals took part in the 2001 follow-up, and 1,172 (90.8%) answered both scales: 38% were male, 62% female, 67% under 75 years of age and 22% with 75 years and above. Median scores were 2.7 for the GHQ-12 and 11.7 for the GDS-30. The prevalences of depressive symptoms based on the GHQ-12 were 40% and 33%, with cut-points 2/3 and 3/4, respectively. Based on the GDS-30 these rates were 50% and 38%, with 10/11 and 13/14 cut-points, respectively. Prevalences were significantly higher among women, those aged > 75 years and those with low educational level. Agreement between the scales ranged from 0.50 to 0.60, with no variation according to demographic factors. Conclusion: To our knowledge, this is the first study comparing the GHQ-12 and GDS-30. We found very high prevalences of depressive symptoms among this elderly cohort, especially with the GDS-30 scale. The agreement between the scales was only reasonable. These results indicate the importance and urgency of validating these scales and investigating the factors that contributed to these results.