Declines in Hb levels over time are associated with poor functional outcomes. The risks and benefits of interventions to increase Hb among older men warrant further investigation to differentiate whether this is an active contributor to age-related debility or a passive biomarker of it.

Dynamic temporal changes in serum T, cFT, DHT, and E2 (but not E1, LH, FSH, and SHBG) in older men are associated with all-cause and cause-specific mortality in distinct patterns.

Our findings show that decline in androgen status is associated with cognitive decline in older men, but the causality of this association requires further elucidation.

The objectives of this study were to examine relationships between baseline levels of reproductive hormones in older men and (1) change in bone mineral density (BMD) over 5 years and (2) incident fractures over an average of 6 years' follow-up. A total of 1705 men aged 70 years and older from the Concord Health and Ageing in Men Project (CHAMP) study were assessed at baseline (2005)(2006)(2007), 2 years follow-up (2007)(2008)(2009), and 5 years follow-up (2010-2013). At baseline, testosterone (T), dihydrotestosterone (DHT), estradiol (E2), and estrone (E1) were measured by liquid chromatography-tandem mass spectrometry (LC-MS/MS), and sex hormone-binding globulin (SHBG), luteinizing hormone (LH), and follicle-stimulating hormone (FSH) by immunoassay. Hip BMD was measured by dual X-ray absorptiometry (DXA) at all three time-points. Fracture data were collected at 4-monthly phone calls and verified radiographically. Statistical modeling was by general estimating equations and Cox model regression. Univariate analyses revealed inverse associations for serum SHBG, FSH, and LH and positive association for E1 but not DHT or E2 with BMD loss at the hip across the three time points. Serum levels of SHBG (b ¼ -0.071), FSH (b ¼ -0.085), LH (b ¼ -0.070), and E1 (b ¼ 0.107) remained significantly associated with BMD loss in multivariate-adjusted models; however, we were unable to identify any thresholds for accelerated BMD loss according to reproductive steroids. Incident fractures (all, n ¼ 171; hip, n ¼ 44; and nonvertebral, n ¼ 139) were all significantly associated with serum SHBG, FSH, and LH levels in univariate models but none remained significantly associated in multivariate-adjusted model. Serum T, DHT, E2, and E1 levels were not associated with incident fractures in univariate or multivariate-adjusted analyses. In older men, lower serum SHBG, FSH, and LH and higher E1 levels protected against loss of BMD without increasing fracture rate. This means these reproductive variables may be considered as novel biomarkers of bone health during male aging.