Serum testosterone levels decline gradually and progressively with aging in men. Many manifestations associated with aging in men, including muscle atrophy and weakness, osteoporosis, reduced sexual functioning, and increased fat mass, are similar to changes associated with testosterone deficiency in young men. These similarities suggest that testosterone supplementation may prevent or reverse the effects of aging. A MEDLINE search was performed to identify studies of testosterone supplementation therapy in older men. A structured, qualitative review was performed of placebo-controlled trials that included men aged 60 and older and evaluated one or more physical, cognitive, affective, functional, or quality-of-life outcomes. Studies focusing on patients with severe systemic diseases and hormone deficiencies related to specific diseases were excluded. In healthy older men with low-normal to mildly decreased testosterone levels, testosterone supplementation increased lean body mass and decreased fat mass. Upper and lower body strength, functional performance, sexual functioning, and mood were improved or unchanged with testosterone replacement. Variable effects on cognitive function were reported, with improvements in some cognitive domains (., spatial, working, and verbal memory). Testosterone supplementation improved exercise-induced coronary ischemia in men with coronary heart disease, whereas angina pectoris was improved or unchanged. In a few studies, men with low testosterone levels were more likely to experience improvements in lumbar bone mineral density, self-perceived functional status, libido, erectile function, and exercise-induced coronary ischemia with testosterone replacement than men with less marked testosterone deficiency. No major unfavorable effects on lipids were reported, but hematocrit and prostate specific antigen levels often increased. Based on these results, testosterone supplementation cannot be recommended at this time for older men with normal or low-normal testosterone levels and no clinical manifestations of hypogonadism. However, testosterone replacement may be warranted in older men with markedly decreased testosterone levels, regardless of symptoms, and in men with mildly decreased testosterone levels and symptoms or signs suggesting hypogonadism. The long-term safety and efficacy of testosterone supplementation remain uncertain. Establishment of evidence-based indications will depend on further demonstrations of favorable clinical outcomes and symptomatic, functional, and quality-of-life benefits in carefully performed, long-term, randomized, placebo-controlled clinical trials.