Context: Mobility limitation is associated with increased morbidity and mortality. The relationship between circulating testosterone and mobility limitation and physical performance is incompletely understood.
Objective: Our objective was to examine cross-sectional and prospective relations between baseline sex hormones and mobility limitations and physical performance in community-dwelling older men.
Design, setting, and participants: We conducted cross-sectional and longitudinal analyses of 1445 men (mean age 61.0 +/- 9.5 yr) who attended Framingham Offspring Study examinations 7 and 8 (mean 6.6 yr apart). Total testosterone (TT) was measured by liquid chromatography tandem mass spectrometry at examination 7. Cross-sectional and longitudinal analyses of mobility limitation and physical performance were performed with continuous (per SD) and dichotomized [low TT and free testosterone (FT) and high SHBG vs. normal] hormone levels.
Main outcome measures: Self-reported mobility limitation, subjective health, usual walking speed, and grip strength were assessed at examinations 7 and 8. Short physical performance battery was performed at examination 7.
Results: Higher continuous FT was positively associated with short physical performance battery score (beta = 0.13; P = 0.008), usual walking speed (beta = 0.02; P = 0.048), and lower risk of poor subjective health [odds ratio (OR) = 0.72; P = 0.01]. In prospective analysis, 1 SD increase in baseline FT was associated with lower risk of developing mobility limitation (OR = 0.78; 95% confidence interval = 0.62-0.97) and progression of mobility limitation (OR = 0.75; 95% confidence interval = 0.60-0.93). Men with low baseline FT had 57% higher odds of reporting incident mobility limitation (P = 0.03) and 68% higher odds of worsening of mobility limitation (P = 0.007).
Conclusions: Lower levels of baseline FT are associated with a greater risk of incident or worsening mobility limitation in community-dwelling older men. Whether this risk can be reduced with testosterone therapy needs to be determined by randomized trials.