Objectives: To determine the prevalence of depression in a cohort of elderly men as assessed using a 30-item Geriatric Depression Scale (GDS) score and to describe the association between this score and sex steroids, androgen receptor (AR) polymorphism, and general health status.
Design: Observational study on the relationship between sex steroid status and health-related parameters.
Setting: Community-based.
Participants: Ambulatory men (n=236 in 1997, n=192 in 2000) aged 70 and older at inclusion in 1996, interviewed in 1997 and 2000.
Measurements: Serum levels of testosterone, estradiol, sex hormone binding globulin (SHBG), dehydroepiandrosterone-sulfate (DHEAS), cortisol, and the AR gene cytosine, adenine, guanine (CAG)-repeat length polymorphism were determined. Free testosterone and free estradiol were calculated. Questionnaires included GDS, 36-item Short Form, and Rapid Disability Rating Scale-2.
Results: Median age was 75.3 years (interquartile range=73.5-78.5). A GDS score of 11 or greater was found in 30 (12.7%) men. Age and GDS score were significantly interrelated (P<.01), as were all health-assessment scores. GDS scores were not related to (free) testosterone or AR polymorphism in 1997 or 2000. In 1997 only (n=236), higher GDS scores were related to higher estradiol, free estradiol, and DHEAS levels.
Conclusion: The data did not support a role for testosterone in depression in elderly community-based men as assessed using the GDS.