U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Committee on Assessing the Need for Clinical Trials of Testosterone Replacement Therapy; Liverman CT, Blazer DG, editors. Testosterone and Aging: Clinical Research Directions. Washington (DC): National Academies Press (US); 2004.

Cover of Testosterone and Aging

Testosterone and Aging: Clinical Research Directions.

Show details


Testosterone is often equated in the popular culture with the macho male physique and virility. Viewed by some as an anti-aging tonic, the growth in testosterone's reputation and increased use by men of all ages in the United States has outpaced the scientific evidence about its potential benefits and risks. Scientific questions of safety and effectiveness are best answered by randomized clinical trials, the gold standard in clinical research. The Women's Health Initiative (WHI) and other large-scale clinical trials, for example, have provided new insights into the benefits and risks of postmenopausal hormone therapy in women that are quite different from what had been assumed during decades of widespread use of estrogen-progestin therapy. Now, as large-scale clinical trials of testosterone therapy are being considered by the National Institutes of Health (NIH) and other research organizations, it is important to carefully assess the rationale for such studies so that the research can be designed to best answer questions regarding benefits and risks in a timely and cost-effective manner.


In 2002, the National Institute on Aging (NIA) and the National Cancer Institute (NCI) asked the Institute of Medicine (IOM) to conduct a 12-month study to review and assess the current state of knowledge related to the potential beneficial and adverse health effects of testosterone therapy in older men, and to make recommendations regarding clinical trials of testosterone therapy, including the parameters that should be considered in study design and conduct. More specifically, the committee was asked to review and consider:

  • epidemiologic data on normal levels of testosterone during the lifespan and the associations with morbidity and mortality;
  • the risks and benefits of testosterone therapy;
  • the potential public health impact of testosterone therapy in the United States; and
  • the ethical issues related to the conduct of clinical trials of testosterone therapy.

The committee members included experts from many fields including bioethics, endocrinology, internal medicine, urology, oncology, epidemiology, biostatistics, clinical trials research, geriatrics, and behavioral science. The committee held four meetings over the course of the 12-month study and convened a public scientific workshop in Phoenix, Arizona, on March 31, 2003.


Increases in life expectancy are resulting in an aging global and U.S. population. In 1900, persons age 65 years and older accounted for only 4 percent of the U.S. population. By 2000, that proportion had risen to 12.4 percent, or 35 million people, and it is projected to rise to 19.6 percent, or 71 million people, by 2030 (CDC, 2003). It has been noted that of all the people who have ever lived to the age of 65 years, more than half are now alive (Resnick, 2001).

The oldest age group—those over age 85—are the fastest growing segment of the older population. It is estimated that the number of persons age 80 and older will increase from 9.3 million in 2000 to 19.5 million in 2030 (CDC, 2003). The ratio of older men to women will narrow slightly over the next few decades in the United States. Men represented 41 percent of those over age 65 in 2000; by 2030 that percentage is projected to increase to 44 percent (CDC, 2003).

Life expectancy continues to rise as well. Male life expectancy at birth in the United States reached a record 74.4 years in 2001 (Arias and Smith, 2003). The growing number of older individuals increases demands on public health and medical and social services. Chronic diseases disproportionately affect older people, who are also more prone to frailty and disabilities (CDC, 2003). In addition, many older people have sensory, mobility, and cognitive impairments that affect their quality of life and may predispose them to falls, injuries, and fractures. In the United States, approximately 80 percent of all persons over age 65 have at least one chronic condition, and 50 percent have at least two (CDC, 2003).

Changes in the levels of many hormones occur naturally with aging, and these changes have long been associated with a variety of chronic conditions. In women, estrogen and progesterone levels drop sharply after ovulation ends. For some time it has been observed that such declines are associated with increased bone loss leading to osteoporosis, and possibly with greater risk for cardiovascular disease and stroke. Thus, hormone replacement therapy (estrogen or estrogen in combination with progestin) was widely prescribed as a preventive agent. Recent information, particularly the analysis of the results of the estrogen plus progestin component of the WHI randomized trial, has provided insights into the risks and benefits of hormone treatment. Although women taking orally administered estrogen plus progestin in this study experienced fewer hip and other fractures and were less likely to develop colorectal cancer, they were more likely to develop heart disease events, stroke, blood clots, and breast cancer (Rossouw et al., 2002). More recently it has been reported that women taking hormones are at greater risk for developing dementia (Shumaker et al., 2003).

Although the focus of this report is on testosterone, it is important to remember that testosterone is but one of many hormones that change with aging in men. The terms adrenopause, somatopause, and andropause have been used to indicate the gradual decline in the adrenal compounds (dehydroepiandrosterone [DHEA] and its sulfate [DHEAS]), the somatotropic hormone (growth hormone [GH]) secreted by the pituitary, and androgens (particularly testosterone). There is some controversy regarding the use of these terms, as the declines are gradual, and there is a great deal of variability between individuals in the extent and nature of declining levels (Gould and Petty, 2000). Some assert that the term true andropause can only be used to describe situations in which testosterone levels drop precipitously, for example following ablative treatment for advanced prostate cancer (Morales et al., 2000).

In many cases, physiological changes seen with aging (such as decreased muscle strength and increased percent body fat) are also seen in individuals with specific hormone deficiencies or excesses (Table 1-1). Thus, these correlations suggest the potential for hormonal effects on aging. However, there are many unknowns regarding how hormone changes may interrelate with or contribute to the overall decline in physiologic function with aging.

TABLE 1-1. Similarities of Changes in Body Composition, Muscle Strength, Aerobic Capacity, and Metabolic Variables with Aging and in Hormone Deficiency/Excess States.


Similarities of Changes in Body Composition, Muscle Strength, Aerobic Capacity, and Metabolic Variables with Aging and in Hormone Deficiency/Excess States.

Beyond the endocrine system, numerous other factors contribute to the physiologic changes associated with aging. To illustrate this point, a recent review (Matsumoto, 2002) listed some of the multiple factors that may contribute to decreased bone mass (potentially increasing the risk for fracture) including low estradiol (E2) concentrations, vitamin D deficiency, low growth hormone and IGF-1(insulin-like growth factor 1) levels, low testosterone levels, poor nutrition, use of certain medications, smoking, excessive alcohol intake, inactivity and lack of exercise, inadequate calcium intake, certain illnesses, and genetic predisposition. The multifactorial etiology of reduced bone mass puts into perspective the complexities involved in diagnosing, treating, and preventing age-related adverse clinical outcomes.


The synthesis of testosterone in men occurs primarily in the Leydig cells of the testes, with a small percentage produced in the adrenal cortex. The testes are the male reproductive glands that also produce sperm. Testosterone, the primary androgenic hormone is synthesized through a series of five enzymatic reactions that convert cholesterol to testosterone (Figure 1-1). Testosterone biosynthesis is up-regulated by luteinizing hormone (LH), a gonadotrophic hormone secreted from the pituitary gland. The hypothalamus secretes gonadotropin-releasing hormone (GnRH), which stimulates the pituitary to secrete LH and follicle-stimulating hormone (FSH). In men, LH stimulates Leydig cells to produce testosterone and FSH acts on Sertoli cells, stimulating spermatogenesis (Figure 1-2). Approximately 5 to 6 mg of testosterone is secreted into plasma daily in men (Griffin and Wilson, 1998). In men, LH and testosterone are secreted in a pulsatile manner every 60 to 90 minutes in a diurnal rhythm, with peak levels occurring in the morning. This circadian pattern appears to be less pronounced in older men (Bremner et al., 1983; Tenover et al., 1988).

FIGURE 1-1. Pathways of testosterone synthesis in human testis.


Pathways of testosterone synthesis in human testis. SOURCE: Reprinted from Griffin JE, Wilson JD. 1998. Disorders of the testes and the male reproductive tract. In: Wilson JD, Foster DW, Kronenberg HM, Larsen PR, eds. Williams Textbook of Endocrinology. (more...)

FIGURE 1-2. Regulation of testosterone and sperm production by LH and FSH (C = cholesterol, T = testosterone).


Regulation of testosterone and sperm production by LH and FSH (C = cholesterol, T = testosterone). SOURCE: Griffin JE, Wilson JD. 2001. Disorders of the testes. In: Braunwald E, Fauci AS, Kasper DL, Hauser SL, Longo DL, Jameson JL, eds. Harrison's Principles (more...)

Testosterone can act directly on target cells, or it can be converted into its primary metabolites, dihydrotestosterone (DHT) and estradiol. Both testosterone and DHT bind to the androgen receptor, but DHT has a higher affinity for the receptor and is therefore a more potent androgen (Bagatell and Bremner, 1996; Bruchovsky and Wilson, 1999). The 5α-reductase enzymes, which convert testosterone to DHT, are most abundant in prostate, skin, and reproductive tissues. The aromatase enzyme complex, which converts testosterone to estradiol, an estrogen, is most abundant in adipose tissue, liver, and certain central nervous system nuclei (Mooradian et al., 1987; Simpson and Davis, 2001). Thus, there are numerous endpoints that may be affected by testosterone and its metabolites.

Approximately 98 percent of testosterone circulates in the blood bound to protein, of which approximately 60 percent is bound weakly to albumin and other proteins and 40 percent is bound with higher binding affinity to sex hormone binding globulin (SHBG) (Figure 1-3) (Dunn et al., 1981; Bhasin et al., 1998). The remaining 2 percent is free or unbound. The fraction available to the tissues (also termed bioavailable testosterone) is believed to be the free plus the albumin-bound testosterone, consisting of approximately half of the total plasma testosterone (Griffin and Wilson, 2001). Testosterone bound to albumin is biologically available due to rapid dissociation.

FIGURE 1-3. Testosterone partitions in the serum.


Testosterone partitions in the serum.

Testosterone and its metabolites play a crucial role in the health and development of the male. At approximately seven weeks of gestation, the male embryo begins production of testosterone, and levels of the hormone are maintained at a high level through most of gestation (Griffin and Wilson, 2001). During fetal development, testosterone and DHT are needed for normal differentiation of male internal and external genitalia. Late in gestation the levels drop, and, at birth, serum testosterone levels are only slightly higher in males than in females. After birth, plasma testosterone levels in male infants rise and are elevated for approximately the first three months, after which the testosterone levels decrease and remain only slightly higher in boys than in girls until the beginning of puberty (Griffin and Wilson, 2001).

During puberty, testosterone is required for the development of male secondary sexual characteristics, stimulation of sexual behavior and function, and initiation of sperm production. Levels of plasma testosterone increase in males reach normal male adult levels of 10 to 35 nmol/L (approximately 300 to 1,000 ng/dL) by about age 17 (Griffin and Wilson, 2001; Merck, 2003). Levels of bioavailable testosterone remain level until men are in their 30s to 40s then the levels begin to decline about 1.2 percent per year (Griffin and Wilson, 2001; Harman et al., 2001). In adult males, testosterone is involved in maintaining muscle mass and strength, fat distribution, bone mass, red blood cell production, male hair pattern, libido and potency, and spermatogenesis (Bagatell and Bremner, 1996).


Several laboratory assays and methods of calculation are used to assess the three testosterone measures: total testosterone (protein bound plus free), free testosterone (not bound to proteins), and bioavailable testosterone (free plus albumin bound). The methods used to conduct the measurements vary in their accuracy, standardization, the extent of validation, and the reproducibility of results. Additionally, there are issues regarding the timing and number of samples needed to provide accurate data that can be compared across studies. Further complicating this issue are the fluctuations of an individual's testosterone levels during the day and the wide range of normal testosterone levels between individuals.

Total testosterone (serum testosterone) is generally measured by radioimmunoassay, which is a validated, standardized, and reproducible assay. However, because the level of the high-affinity binding protein SHBG increases with age (and therefore a greater percentage of the total testosterone is bound to SHBG and is not available to the tissues), this measure may not be as useful in studies of aging populations as are measures of bioavailable testosterone.

Bioavailable testosterone (free plus albumin bound) is measured or calculated in several ways. SHBG in serum can be precipitated with ammonium sulfate and the bioavailable testosterone is then measured in the supernate (SHBG is precipitated by a lower concentration of ammonium sulfate than albumin) (Rosner, 1972). Alternatively, bioavailable testosterone can be calculated using measures of total testosterone and immunoassayed SHBG concentrations.

Measures of free testosterone are more controversial. Laboratory measurements of free testosterone have generally been conducted by equilibrium dialysis. This method is standardized and validated, but is only available through reference laboratories (Matsumoto, 2002) and is costly. Direct nondialysis measures of free testosterone using analog immunoassays are widely used in local laboratories; however, the results appear to be less accurate (Winters et al., 1998; Rosner, 2001) with either high or low SHBG levels. Free testosterone can also be calculated using measurements of total testosterone, albumin, and SHBG concentrations (Vermeulen et al., 1999).

As noted above, the timing of the sampling may influence comparisons among individual testosterone levels, due to diurnal variations, particularly in younger men.


Since the time of the ancient Egyptians and Romans, the products of the testis have been thought to act as aphrodisiacs and as a fountain of youth to boost physical strength and reverse the effects of aging (Hoberman and Yesalis, 1995). The modern field of endocrinology emerged at the turn of the 20th century as researchers working on “internal secretions” (termed hormones in 1905 by the British scientist Ernest Henry Starling) explored how those compounds act as physiological regulators. One of the early experiments was reported in 1889 by French physiologist Charles Edouard Brown-Séquard who attributed increases in his physical strength and intellectual energy to self-injections of an extract from the testicles of dogs and guinea pigs (Medvei, 1982). The continued use of crude (possibly inactive) gonadal preparations continued into the 1930s, to be gradually replaced with periodic injections of testosterone. In 1939 Leopold Ruzicka and Adolf Butenandt shared the Nobel Prize for Chemistry for their work on isolating and synthesizing testosterone and other reproductive hormones (Malmström and Andersson, 2003).

A number of testosterone compounds have been approved by the Food and Drug Administration (FDA) as treatments for specific conditions, particularly hypogonadism. Testosterone products must be prescribed and are designated as Schedule III controlled substances due to abuse potential. Because testosterone is weakly soluble in water, limiting absorption, and is rapidly metabolized by the liver, bioavailability via the oral route is limited. Therefore, a variety of non-oral delivery methods have been developed (e.g., gel, patch, injection). The goal in developing testosterone formulations has been to produce a product that will deliver physiological levels for prolonged periods of time; is safe, effective, easy to use, and inexpensive; and has few local side effects (e.g., skin irritability) (Handelsman, 1996).

The oral forms of alkylated androgen compounds available in the United States are generally not recommended for use as testosterone therapy because they may produce deleterious effects, including hepatotoxicity (hemorrhagic liver cysts, cholestasis, and hepatocellular adenoma) and unfavorable alterations in the lipid profile (Wang, 1996; AACE, 2002; Swerdloff and Wang, 2002). Orally administered testosterone is almost completely inactivated by its first pass through the liver, and this rapid metabolism makes it difficult to sustain constant levels of circulating hormone. In Europe, testosterone undecanoate is available and is considered a more acceptable oral alternative, as it is absorbed from the gastrointestinal tract into the lymphatic system due to its lipophilic side chain and, thus, partially escapes hepatic inactivation (Matsumoto, 2002). However, the absorption is rather variable, and the dose required is best determined on the basis of plasma levels and clinical effects.

There are several testosterone formulations that can be delivered by intramuscular injection. Testosterone enanthate and testosterone cypionate are testosterone esters that are available in oil suspension preparations. Esterification increases the lipid solubility of the compound and extends its action (Winters, 1999). This form of administration may yield transient supraphysiological levels the first two to three days after injection and then decline toward the end of the dosing interval to subphysiologic levels (Sokol et al., 1982; AACE, 2002). The high levels may result in acne and polycythemia; at low levels men may experience fluctuations in sexual function, energy, and mood. The usual dose for adults is 150 to 200 mg administered every two to three weeks (Winters, 1999). Lower doses given at more frequent intervals (50 to 100 mg every 7 to 10 days) produce more sustained levels, but may be more inconvenient, particularly if the injections are not self-administered and require visits to the physician's office (AACE, 2002).

Transdermal delivery of testosterone allows for absorption directly into systemic circulation at a controlled rate, thus alleviating the fluctuations in levels (Winters, 1999). Transdermal scrotal or permeation-enhanced nonscrotal patches deliver 4 to 6 mg of testosterone per day. The scrotal patch was developed to take advantage of the high permeability of scrotal skin (at least five times more permeable to testosterone than other skin sites). However, the high concentration of the enzyme 5α-reductase present in the scrotal skin may result in higher than normal levels of DHT, a testosterone metabolite of concern regarding prostatic hyperplasia and cancer. Accordingly, nonscrotal patches have been developed that can be applied to sites on the back, abdomen, upper arms, and thighs (Findlay et al., 1989). Because enhancers are needed to increase absorption, local skin irritation has been the most common adverse effect reported with the patch delivery method. Second generation torso patches have reduced skin side effects. Recently, several gel formulations have been approved by the FDA. Gel formulations are applied daily to nongenital skin, generally the shoulders and upper arms. The gel dries quickly, but there is potential for transfer of the gel from person to person through direct skin contact. The transdermal delivery systems have the advantage of immediate cessation of drug delivery when the product is removed or not reapplied. A transbuccal (gum surface) delivery system recently received FDA approval. This method uses a tablet that adheres to the gum surface; testosterone is absorbed through the buccal mucosa into the bloodstream.

Other forms of testosterone supplements have been used, are in use in other countries, or are in development, including testosterone pellets implanted subcutaneously every four to six months, variations of orally administered preparations, transdermal gels, and long-acting injectables.

A new class of compounds may provide an alternative to testosterone. Selective androgen receptor modulators (SARMs) are a class of compounds that have been reported to have androgenic effects similar to testosterone on muscle mass, sexual function, and bone density in animal models, while apparently causing little or no harm to the prostate (Orwoll, 2001). SARMs exhibit moderate-to-high androgen receptor binding affinity similar to testosterone while maintaining selective tissue effects (Yin et al., 2003). SARMs do not affect or act as a substrate for 5α-reductase, so they do not metabolize to DHT (Negro-Vilar, 1999). Although they appear to be promising, these compounds are still in the developmental stages.

Treating Hypogonadism and Other Medical Conditions

Testosterone products have been approved by the FDA for the treatment of primary and secondary hypogonadism in males. Some products are also approved for use in delayed puberty in males or metastatic breast cancer in females.

The benefits of testosterone therapy for markedly hypogonadal males have been well established. Hypogonadism is defined as “inadequate gonadal function, as manifested by deficiencies in gametogenesis and/or the secretion of gonadal hormones” (Stedman's Medical Dictionary, 2000). Male hypogonadism is categorized as primary or secondary (also termed central) based on the location of the disorder. In primary hypogonadism, the testes do not function properly for reasons including surgery, radiation, genetic and developmental disorders, infection, or liver and kidney disease. The most common genetic disorder resulting in primary hypogonadism in men is Klinefelter's syndrome, in which there is an extra sex chromosome, XXY. Primary hypogonadism is characterized by low levels of testosterone with elevated levels of the gonadotropins, FSH and LH.

Secondary (or hypogonadotropic) hypogonadism is the result of disorders in the pituitary gland or hypothalamus. Causes of secondary hypogonadism include pituitary tumors, surgery, radiation, infections, inflammation, trauma, bleeding, genetic problems, nutritional deficiency, and iron excess (hemochromatosis) (Medline Plus, 2002). In secondary hypogonadism testosterone levels are low, while the levels of FSH and LH remain in the low to low-normal range.

The clinical manifestations of androgen deficiency depend on the age at onset and the severity and duration of the deficiency. In the first trimester of fetal development, androgen deficits in the male can result in inadequate differentiation of external genitalia. During puberty, male teens with hypogonadism may have poor muscle development and sparse body hair, and there may be continued long bone growth due to delayed fusion of the epiphyses. In adult males, hypogonadism can result in decreased libido, decreased strength, sparse body hair, and—depending on the degree and length of the deficiency—osteopenia and gynecomastia (Merck, 2003).

Hypogonadism is diagnosed easily when the usual signs and symptoms of androgen deficiency are present or when the patient has a history of a predisposing condition (e.g., mumps orchitis, orchiectomy, radiation to the pelvis or head). The diagnosis of hypogonadism in adult males involves a comprehensive history and physical examination in addition to laboratory tests for levels of testosterone and gonadotropins, and possible further testing to determine the cause.

Testosterone levels alone are not considered sufficient evidence to define hypogonadism. A recent review of guidelines for the evaluation and treatment of male hypogonadism by a task force of the American Association of Clinical Endocrinologists (AACE) stated that men with total testosterone levels less than 200 ng/dL and with symptoms of hypogonadism may be candidates for testosterone therapy, but the report did not issue specific recommendations (AACE, 2002). The report of the Endocrine Society's Second Annual Andropause Consensus Meeting (Endocrine Society, 2002) delineated three categories for consideration in screening and diagnosing hypogonadism in men over 50 years of age: 1) total testosterone less than or equal to 200 ng/dL: “diagnosis of androgen deficiency is confirmed. Rule out serious hypothalamic or pituitary disease in men with hypogonadotropic hypogonadism” prior to initiating testosterone therapy; 2) total testosterone levels greater than 200 but less than 400 ng/dL: recommended additional measures of testosterone and further evaluation before considering testosterone therapy; and 3) total testosterone levels greater than 400 ng/dL: considered not to have testosterone deficiency. Many studies have used the 300 to 350 ng/dL range of total testosterone as a cutoff for identifying hypogonadal patients, although there is not a clearly defined standard, and other factors such as SHBG, LH, and FSH levels and the clinical presentation and physical findings are key in making a diagnosis of hypogonadism. Some studies define andropausal or androgen-deficient levels of testosterone in older men as those 2 standard deviations or more below normal laboratory values for young men (approximately 320 ng/dL total testosterone, 7 ng/dL free testosterone, 90 to 230 ng/dL bioavailable testosterone) (Heaton, 2003).

The prevalence of androgen deficiency is not known with certainty, and hypogonadism is probably underdiagnosed (Winters, 1999). It has been estimated that 4 to 5 million Americans have hypogonadism,1 of which 5 percent receive testosterone therapy (FDA, 2001).

In addition to its use for treating hypogonadism, testosterone has been used in men and women to treat the wasting syndrome of advanced AIDS, the pronounced muscle wasting associated with glucocorticoid therapy, and debilitating illnesses such as emphysema and cirrhosis. Testosterone is also being evaluated as a male hormonal contraceptive because it suppresses the production of pituitary gonadotropins and therefore spermatogenesis (Amory and Bremner, 2000).

Use of Testosterone Therapy in Aging Men

Unlike estrogen declines in women, which are precipitous with the end of ovulation, the testosterone decline in men is gradual (Griffin and Wilson, 2001; Harman et al., 2001). This decrease in testosterone levels results from a decline in the testicular production of testosterone as well as from reduced hypothalamic secretion of gonadotropin-releasing hormone (and consequently reduced LH secretion by the pituitary) (Matsumoto, 2003). The number of Leydig cells in the testes may decline with aging. However, some of the declines in testosterone production are mitigated by decreases in the metabolic clearance rate of circulating testosterone with aging (Matsumoto, 2002). Further, testosterone levels may be decreased due to increased body mass index, alcohol use, presence of chronic disease (e.g., diabetes, endocrine disorders), or use of some medications (e.g., glucocorticoids) (Kaufman and Vermeulen, 1997).

While testosterone production declines with age, many older men have levels well within the normal range for younger men. Normal aging is associated with some of the same symptoms as hypogonadism (e.g., decreases in muscle strength and fat-free body mass), but not all older men meet the clinical definitions of hypogonadism. As noted above, the diagnosis of testosterone deficiency in older men involves attention to a range of symptoms and an extensive history and physical examination, in addition to tests of hormone levels. Further complicating the diagnosis is the fact that low testosterone levels in some cases may be a marker, rather than a cause, of ill health. Thus, there are a number of issues confounding a clear interpretation of the meaning of diminishing testosterone levels in older men and their relationship to aging and health. These issues include the vagueness of the definition of hypogonadism/androgen deficiency in older men, the overlap with normal aging symptoms and health status, the wide range of normal levels in a given population, and the uncertainty as to which measure of testosterone should be used to diagnose hypogonadism in older men.

However, the association of lower testosterone levels with lower muscle mass and other age-related conditions suggests that testosterone therapy might be beneficial in some older men. As discussed further in Chapter 2, there are varying degrees of evidence of potential benefits of testosterone treatment in older men, including positive changes in body composition; improved strength; positive effects on fatigue, mood, and sexual function; and increased bone mineral density. Further, there are potential adverse effects, including obstructive sleep apnea, urinary obstruction, gynecomastia, polycythemia, benign prostatic hyperplasia, and prostate cancer.

For men who have extreme testosterone deficiencies, testosterone therapy may offer substantial benefit. Although some older men who have tried testosterone therapy report feeling “more energetic” or “younger,” testosterone supplementation remains a scientifically unproven method for preventing or relieving any physical or psychological change that men may experience as they get older. The levels at which testosterone therapy might be indicated are unclear, that is, it is uncertain whether men who are at the lower end of the normal range of testosterone production would benefit most from treatment. Experts are also concerned about potential long-term harmful effects (e.g., prostate cancer) that testosterone might have on the aging body. Until more scientifically rigorous studies are conducted, the questions regarding the nature and extent of benefits of testosterone therapy and whether benefits outweigh the potential negative effects will remain unanswered. It is uncertain whether all men being prescribed testosterone have been diagnosed clinically and biochemically as hypogonadal, although off-label use of approved drugs is allowed, based on the physician's decision. Skepticism about treatment in the absence of disease has been heightened by the experience of women prescribed postmenopausal hormone therapy.


In recent years there has been growing concern about an increase in the use of testosterone by middle-aged and older men who have borderline testosterone levels—or even normal testosterone levels—in the absence of adequate scientific information about its risks and benefits. More than 1.75 million prescriptions for testosterone products were written in 2002,2 an estimated increase of 30 percent over the approximately 1.35 million prescriptions in 2001, and an increase of 170 percent from the 648,000 prescriptions in 1999 (Rose, 2003) (Figure 1-4). This trend has been seen since at least the early 1990s; a cumulative 500 percent increase in prescription sales of testosterone was reported from 1993 to 2000 (Bhasin and Buckwalter, 2001). Testosterone product sales in the United States, stable at about $18 million until 1988, were projected to reach $400 million before the end of 2002 (Bhasin et al., 2003).

FIGURE 1-4. Testosterone therapy prescription trend.


Testosterone therapy prescription trend. SOURCE: Rose, 2003. Reprinted with permission of IMS Consulting, a division of IMS Health, from a report to Solvay Pharmaceuticals.

Growth in the use of testosterone can also be seen in the data on the number of people purchasing testosterone products. According to data collected by IMS Consulting, there were more than 800,000 testosterone-treated patients (men and women) in 2002, an increase of 29 percent over the number for 2001 (Figure 1-5) (Rose, 2003). Approximately 58 percent of testosterone therapy retail patients in 2002 were between 46 and 65 years of age, with 28 percent in the 18- to 45-year age range, 13 percent over 65 years, and about 1 percent under 18 years old (Figure 1-6) (Rose, 2003). An analysis by Dendrite International found similar data, with 70 percent of sales to 40- to 69-year-olds and 12 percent to people age 70 and older (Rose, 2003). An analysis using U.S. census data estimated that use of testosterone products increased from 4.7 per 1,000 males over 65 years of age in 2001 to 5.6 per 1,000 in 2002, a 19 percent increase (Rose, 2003). This compares with a 23 percent increase in use by 46- to 65-year-old men from 9.3 per 1000 in 2001 to 11.4 per 1,000 in 2002. Given the size and projected growth of the aging male population, it is important to know the effects of testosterone therapy before more men are treated at considerable cost and uncertain benefit or safety.

FIGURE 1-5. Growth in the number of testosterone treated patients.


Growth in the number of testosterone treated patients. SOURCE: Rose, 2003. Reprinted with permission of IMS Consulting, a division of IMS Health, from a report to Solvay Pharmaceuticals.

FIGURE 1-6. Total retail patient count by age for 2002.


Total retail patient count by age for 2002. SOURCE: Rose, 2003. Reprinted with permission of IMS Consulting, a division of IMS Health, from a report to Solvay Pharmaceuticals.

There are few data available on testosterone use or on prescribing practices. It would be helpful in planning future research efforts to quantify the increased use of testosterone in clinical practice. Information on the outcomes that the physician and patient consider the “desired action” would also be of interest. In particular, it would be useful to establish whether the primary reasons for taking (or prescribing) testosterone (or other androgens) include concerns about muscle mass or strength, vitality, and sexual function as is often heard anecdotally, as opposed to cardiovascular health or osteroporosis, for which alternative therapies are available.

It is also important to note the increased abuse of anabolic-androgenic steroids in the United States, primarily among healthy athletes but also by adolescents (NIDA, 2003). Further, there are also a variety of androgenic compounds available over the counter, which may confound some research efforts. Certain steroidal supplements can be converted to testosterone or its metabolites, and abuse may lead to numerous health problems. Use of androgen supplements is an often unrecognized cause of male infertility (Schover and Thomas, 2000).


As mentioned previously, the federal government and the private sector have sponsored and conducted long-term, large-scale trials of the relative risks and benefits of postmenopausal hormone therapy in women. In 1991, the National Heart, Lung, and Blood Institute and other units of the NIH launched the WHI, one of the largest studies of its kind ever undertaken in the United States. It includes a factorial clinical trial, an observational study, and a community prevention study, which together involve more than 161,000 healthy postmenopausal women. The observational study is examining predictors and biological markers for disease and is being conducted at more than 40 centers across the United States, while the community prevention study, which has ended, sought to find ways to encourage women to adopt healthful behaviors. WHI's factorial clinical trial, conducted at the same U.S. centers, is designed to test the effects of postmenopausal hormone therapy, diet modification, and calcium and vitamin D supplements on cardiovascular disease, osteoporotic fractures, and breast and colorectal cancer risk. In May 2002 the component of the trial examining the combination of estrogen plus progestin was halted because of evidence of increased risk for breast cancer, coronary events, stroke, and venous thromboembolism (Rossouw et al., 2002). The ongoing estrogen component of the trial will provide further clarity about the risks and benefits of unopposed estrogen therapy in older women.

Existing data on testosterone treatment in older men are derived from many small studies. The results of these studies are difficult to summarize because of differences in design and methodology. The studies are generally of short duration, are conducted in a variety of populations, and often do not include adequate controls. A large-scale clinical trial of testosterone treatment has been proposed. The ESTEEM (Efficacy and Safety of Testosterone in Elderly Men) trial proposes to follow 6,000 men age 65 years and older with low serum testosterone levels for seven years. The trial would examine multiple outcomes, including the effects of testosterone therapy on physical function, bone mineral density, fractures, and clinical prostate cancer. As a result of discussions regarding large-scale trials of testosterone therapy, the National Institute of Aging and the National Cancer Institute asked the IOM to assess the current state of knowledge regarding the potential risks and benefits of testosterone therapy and provide recommendations on directions for further clinical research on testosterone and its effects on human health.


This report examines the state of current scientific knowledge regarding testosterone therapy in older men and assesses the types of clinical research needed to determine the benefits and risks of testosterone therapy in the aging male population. Chapter 2 provides an overview of the research that has been conducted on changes in endogenous testosterone levels with aging and on the associations of testosterone therapy with a range of health outcomes including bone mineral density, body composition, physical function, sexual function, cardiovascular outcomes, prostate outcomes, cognitive function, mood, depression, and quality of life. Chapter 3 addresses issues in clinical trials of testosterone therapy and provides the committee's recommendations regarding future research directions. The committee's concluding remarks are contained in Chapter 4.


  • AACE (American Association of Clinical Endocrinologists). 2002. American Association of Clinical Endocrinologists medical guidelines for clinical practice for the evaluation and treatment of hypogonadism in adult male patients–2002 update. Endocrine Practice 8(6):439–456. [PubMed: 15260010]
  • Amory JK, Bremner WJ. 2000. Newer agents for hormonal contraception in the male. Trends in Endocrinology Metabolism 11(2):61–66. [PubMed: 10675892]
  • Arias E, Smith BL. 2003. Deaths: preliminary data for 2001 . National Vital Statistics Report 51(5). [PubMed: 12670054]
  • Bagatell CJ, Bremner WJ. 1996. Androgens in men: uses and abuses. New England Journal of Medicine 334(11):707–714. [PubMed: 8594431]
  • Bhasin S, Bagatell CJ, Bremner WJ, Plymate SR, Tenover JL, Korenman SG, Nieschlag E. 1998. Issues in testosterone replacement in older men. Journal of Clinical Endocrinology and Metabolism 83(10):3435–3448. [PubMed: 9768643]
  • Bhasin S, Buckwalter JG. 2001. Testosterone supplementation in older men: a rational idea whose time has not yet come. Journal of Andrology 22(5):718–731. [PubMed: 11545281]
  • Bhasin S, Singh AB, Mac RP, Carter B, Lee MI, Cunningham GR. 2003. Managing the risks of prostate disease during testosterone replacement therapy in older men: recommendations for a standardized monitoring plan. Journal of Andrology 24(3):299–311. [PubMed: 12721204]
  • Bremner WJ, Vitiello MV, Prinz PN. 1983. Loss of circadian rhythmicity in blood testosterone levels with aging in normal men. Journal of Clinical Endocrinology and Metabolism 56(6):1278–1281. [PubMed: 6841562]
  • Bruchovsky N, Wilson JD. 1999. Discovery of the role of dihydrotestosterone in androgen action. Steroids 64(11):753–759. [PubMed: 10577831]
  • CDC (Centers for Disease Control and Prevention). 2003. Public Health and Aging: Trends in Aging: United States and Worldwide. [Online]. Available: http://www​.cdc.gov/mmwr​/preview/mmwrhtml/mm5206a2.htm [accessed February 2003].
  • Dunn JF, Nisula BC, Rodbard D. 1981. Transport of steroid hormones: binding of 21 endogenous steroids to both testosterone-binding globulin and corticosteroid-binding globulin in human plasma. Journal of Clinical Endocrinology and Metabolism 53(1):58–68. [PubMed: 7195404]
  • Endocrine Society 2002. Summary from the Second Annual Andropause Consensus Committee.
  • FDA (Food and Drug Administration). 2001. Skin Patch Replaces Testosterone. [Online]. Available: http://www​.fda.gov/fdac/departs/196_upd​.html [accessed November 2002].
  • Findlay JC, Place V, Snyder PJ. 1989. Treatment of primary hypogonadism in men by the transdermal administration of testosterone. Journal of Clinical Endocrinology and Metabolism 68(2):369–373. [PubMed: 2493029]
  • Gould DC, Petty R. 2000. The male menopause: does it exist? Western Journal of Medicine 173(2):76–78. [PMC free article: PMC1070997] [PubMed: 10924412]
  • Griffin JE, Wilson JD. 1998. Disorders of the testes and the male reproductive tract. In: Wilson JD, editor; , Foster DW, editor; , Kronenberg HM, editor; , Larsen PR, editor. , eds. Williams Textbook of Endocrinology . 9th ed. Philadelphia, PA: W.B. Saunders Co. Pp.819–875.
  • Griffin JE, Wilson JD. 2001. Disorders of the testes. In: Braunwald E, editor; , Fauci AS, editor; , Kasper DL, editor; , Hauser SL, editor; , Longo DL, editor; , Jameson JL, editor. , eds. Harrison's Principles of Internal Medicine . 15th ed. New York: McGraw Hill. Pp.2143–2154.
  • Handelsman DJ. 1996. Androgen delivery systems: testosterone pellet implants. In: Bhasin S, editor; , Gabelnick HL, editor; , Spieler JM, editor; , Swerdloff JM, editor; , Wang C, editor; , Kelly C, editor. , eds. Pharmacology, Biology, and Clinical Applications of Androgens: Current Status and Future Prospects . New York: Wiley-Liss. Pp.459–469.
  • Harman SM, Metter EJ, Tobin JD, Pearson J, Blackman MR. 2001. Longitudinal effects of aging on serum total and free testosterone levels in healthy men. Baltimore Longitudinal Study of Aging. Journal of Clinical Endocrinology and Metabolism 86(2):724–731. [PubMed: 11158037]
  • Heaton JP. 2003. Hormone treatments and preventive strategies in the aging male: whom and when to treat? Reviews in Urology 5(S1):S16–S21. [PMC free article: PMC1502318] [PubMed: 16985938]
  • Hoberman JM, Yesalis CE. 1995. The history of synthetic testosterone. Scientific American 272(2):76–81. [PubMed: 7817189]
  • Kaufman JM, Vermeulen A. 1997. Declining gonadal function in elderly men. Bailliere's Clinical Endocrinology and Metabolism 11(2):289–309. [PubMed: 9403124]
  • Malmström BG, Andersson B. 2003. The Nobel Prize in Chemistry: The Development of Modern Chemistry. [Online]. Available: http://www​.nobel.se/chem-istry​/articles/malmstrom/index.html [accessed May 2003].
  • Matsumoto AM. 2002. Andropause: clinical implications of the decline in serum testosterone levels with aging in men. Journals of Gerontology. Series A, Biological Sciences & Medical Sciences 57(2):M76–M99. [PubMed: 11818427]
  • Matsumoto AM. 2003. Fundamental aspects of hypogonadism on the aging male Reviews in Urology 5(S1):S3–S10. [PMC free article: PMC1502324] [PubMed: 16985941]
  • Medline Plus. 2002. Medical Encyclopedia: Hypogonadism. [Online]. Available: http://www​.nlm.nih.gov​/medlineplus/print​/ency/article/001195.htm [accessed May 2003].
  • Medvei VC. 1982. A History of Endocrinology . Lancaster, England: MTP Press.
  • Merck. 2003. Male hypogonadism. In: Beers MH, editor; , Berkow R, editor. , eds. The Merck Manual of Diagnosis and Therapy . 17th ed. Whitehouse Station, NJ: Merck & Co., Inc.
  • Mooradian AD, Morley JE, Korenman SG. 1987. Biological actions of androgens. Endocrine Reviews 8(1):1–28. [PubMed: 3549275]
  • Morales A, Heaton JP, Carson CC 3rd. 2000. Andropause: a misnomer for a true clinical entity. Journal of Urology 163(3):705–712. [PubMed: 10687961]
  • Negro-Vilar A. 1999. Selective androgen receptor modulators (SARMs): a novel approach to androgen therapy for the new millennium. Journal of Clinical Endocrinology and Metabolism 84(10):3459–3462. [PubMed: 10522980]
  • NIDA (National Institute on Drug Abuse). 2003. Research Report: Anabolic Steroid Abuse. [Online]. Available: http://www.drugabuse.gov/Research Reports/Steroids/AnabolicSteroids.html [accessed May 2003].
  • Orwoll ES. 2001. Equal time for the older male: pathophysiology, evaluation, and management of male osteoporosis. Annals of Long-Term Care 9(8).
  • Resnick NM. 2001. Geriatric medicine. In: Braunwald E, editor; , Fauci AS, editor; , Kasper DL, editor; , Hauser SL, editor; , Longo DL, editor; , Jameson JL, editor. , eds. Harrison's Principles of Internal Medicine . 15th ed. New York: McGraw Hill. Pp.36–46.
  • Rose K. 2003. Extent and Nature of Testosterone Use . Presentation at the March 31, 2003 Workshop of the IOM Committee on Assessing the Need for Clinical Trials of Testosterone Replacement Therapy, Phoenix, AZ.
  • Rosner W. 1972. A simplified method for the quantitative determination of testosterone-estradiol binding globulin activity in human plasma. Journal of Clinical Endocrinology and Metabolism 34(6):983–988. [PubMed: 4112000]
  • Rosner W. 2001. An extraordinarily inaccurate assay for free testosterone is still with us. Journal of Clinical Endocrinology and Metabolism 86(6):2903. [PubMed: 11397908]
  • Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC, Kotchen JM, Ockene J. 2002. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. Journal of the American Medical Association 288(3):321–333. [PubMed: 12117397]
  • Schover LR, Thomas, AJ. 2000. Overcoming Male Infertility . New York: John Wiley & Sons.
  • Shumaker SA, Legault C, Thal L, Wallace RB, Ockene JK, Hendrix SL, Jones BN 3rd, Assaf AR, Jackson RD, Kotchen JM, Wassertheil-Smoller S, Wactawski-Wende J. 2003. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. Journal of the American Medical Association 289(20):2651–2662. [PubMed: 12771112]
  • Simpson ER, Davis SR. 2001. Minireview: aromase and the regulation of estrogen biosynthesis—some new perspectives. Endocrinology 142(11):4589–4594. [PubMed: 11606422]
  • Sokol RZ, Palacios A, Campfield LA, Saul C, Swerdloff RS. 1982. Comparison of the kinetics of injectable testosterone in eugonadal and hypogonadal men. Fertility & Sterility 37(3):425–430. [PubMed: 6800853]
  • Stedman's Medical Dictionary. 2000. Philadelphia, PA: Lippincott Williams & Wilkins.
  • Swerdloff RS, Wang C. 2002. Androgens and the aging male. In: Lunenfeld B, editor; , Gooren L, editor. , eds. Textbook of Men's Health . Boca Raton, FL: Parthenon Publishing. Pp.148–157.
  • Tenover JS, Matsumoto AM, Clifton DK, Bremner WJ. 1988. Age-related alterations in the circadian rhythms of pulsatile luteinizing hormone and testosterone secretion in healthy men. Journal of Gerontology 43(6):M163–M169. [PubMed: 3183306]
  • Vermeulen A, Verdonck L, Kaufman JM. 1999. A critical evaluation of simple methods for the estimation of free testosterone in serum. Journal of Clinical Endocrinology and Metabolism 84(10):3666–3672. [PubMed: 10523012]
  • Wang C. 1996. Androgen delivery systems: Overview of existing methods and applications. In: Bhasin S, editor; , Gabelnick HL, editor; , Spieler JM, editor; , Swerdloff RS, editor; , Wang C, editor; , Kelly C, editor. , eds. Pharmacology, Biology, and Clinical Applications of Androgens: Current Status and Future Prospects . New York: Wiley-Liss. Pp.433–435.
  • Winters SJ. 1999. Current status of testosterone replacement therapy in men. Archives of Family Medicine 8(3):257–263. [PubMed: 10333822]
  • Winters SJ, Kelley DE, Goodpastor B. 1998. The analog free testosterone assay: are the results in men clinically useful? Clinical Chemistry 44(10):2178–2182. [PubMed: 9761253]
  • Yin D, Gao W, Kearbey JD, Xu H, Chung K, He Y, Marhefka CA, Veverka KA, Miller DD, Dalton JT. 2003. Pharmacodynamics of selective androgen receptor modulators. Journal of Pharmacology and Experimental Therapeutics 304(3):1334–1340. [PubMed: 12604714]



Criteria used to define hypogonadism are not provided.


Since testosterone products are Schedule III controlled drugs, prescription renewal may require a new prescription.

Copyright 2004 by the National Academy of Sciences. All rights reserved.
Bookshelf ID: NBK216164


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (4.9M)

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...