Full text of DOI:10.1093/gerona/57.5.M321

Journal of Gerontology: MEDICAL SCIENCES
Copyright 2002 by The Gerontological Society of America
2002, Vol. 57A, No. 5, M321–M325
Effects of Transdermal Testosterone on Cognitive
Function and Health Perception in Older Men
With Low Bioavailable Testosterone Levels
Anne M. Kenny,1 Sandra Bellantonio,1 Cynthia A. Gruman,2 Ruben D. Acosta,1
and Karen M. Prestwood1
1Center on Aging, University of Connecticut Health Center, Farmington.
2Braceland Center for Mental Health and Aging, Institute of Living/Hartford Hospital, Connecticut.
Background. Many men older than 50 years have bioavailable testosterone levels below the reference range for young
adult men. The impact of the decreased androgen levels on cognition and health perception has received little attention.
Methods. Sixty-seven men (mean age 76 ꢀ 4 years, range 65–87) with bioavailable testosterone levels below 128
ng/dl (lower limit for adult normal range) were randomized to receive transdermal testosterone (2–2.5 mg patches/d) or
placebo patches for 1 year. All men received 500 mg supplemental calcium and 400 IU vitamin D. Outcome measures
included sex hormones [testosterone, bioavailable testosterone, sex hormone binding globulin (SHBG), estradiol, and
estrone], cognitive tests (Digit Symbol, Digit Span, Trailmaking A and B), health perception (Medical Outcome Survey
Short-form 36 or SF-36), lower extremity muscle strength and power, and calcium intake.
Results. Twenty-three men (34%) withdrew from the study; 44 men completed the trial. Bioavailable testosterone
levels increased from 93 ꢀ 34 (SD) to 162 ꢀ 100 ng/dl (p ꢁ .002) at 12 months in the testosterone group (n ꢂ 24) while
no change occurred in the control group (n ꢂ 20). While there was no change in estradiol levels in either group, estrone
levels increased in the testosterone group (28 ꢀ 7 to 32 ꢀ 9 pg/dl, p ꢂ .017). Scores on the Digit Symbol test improved
in both the testosterone and placebo groups. Scores on Trailmaking B improved in men treated with testosterone (p ꢁ
.005), although the changes were not statistically different from the changes seen in the placebo group. Twelve-month
scores on Trailmaking B for the entire group were correlated with 12-month testosterone levels (p ꢂ .016). Scores for
health perception measured by SF-36 did not change significantly, though scores of mental and general health declined
in both groups during the 12-month intervention. Twelve-month bioavailable testosterone scores were directly corre-
lated with scores for physical role (p ꢂ .022), vitality (p ꢂ .036), and the physical composite score (p ꢂ .010).
Conclusions. Transdermal testosterone treatment in men with low bioavailable testosterone levels does not impair
and may improve cognitive function. Treatment did not improve health perception but this may have been due to the
side effects of skin irritation suggested by similar reactions in both the testosterone and placebo groups.
ANY men older than the age of 50 have bioavail-
able testosterone levels below the normal young
Health perceptions related to testosterone replacement
have not been extensively investigated in older men. Young
hypogonadal men reported decreased irritability, tiredness,
anxiety, and sadness and increased energy level, cheerful-
ness, and sense of well being on testosterone replacement
(17,18). Similarly, untreated young men with hypogo-
nadism reported higher scores for anger, fatigue, and confu-
sion than eugonadal control subjects (19). However, no as-
sociation has been found between health perception and
baseline testosterone levels in other studies (18,20), includ-
ing studies of older men (21). In studies of men receiving
testosterone replacement, no changes in health perception
have been detected in men with testosterone levels 1 SD be-
low the young adult mean (22) or in men with normal tes-
tosterone levels (23).
This study is part of a larger study to assess the effects of
testosterone on bone and muscle (24) and in addition, we
sought to determine the effects of transdermal testosterone
supplementation on cognition and health perception in men
older than 65 years with low levels of bioavailable testoster-
one. We hypothesized that transdermal testosterone replace-
M
adult range. There is a progressive decrease in bioavail-
able testosterone levels with increasing age and further
declines associated with medical comorbidity (1–3). The
significance of the gradual decline in testosterone levels
on cognition and health perception has received little at-
tention. In animal studies, testosterone replacement re-
verses memory loss and decreases amyloid precursor pro-
tein production (4,5). Young hypogonadal men with low
testosterone levels have impaired cognitive function,
compared with eugonadal men, that improved following
testosterone replacement (6). Similarly, in studies of
younger men with severe testosterone deficiency, testos-
terone levels and spatial cognition are inversely associ-
ated (7–11). While few studies exist in older men, higher
testosterone levels were predictive of performance in sev-
eral domains of cognition (12,13), and testosterone sup-
plementation improved performance on spatial cognition
and verbal and working memory compared to placebo in
three small studies (14–16).
M321

M322
KENNY ET AL.
ment would improve cognitive function and health percep-
tion in older men compared with placebo.
strength, and skin effect (rash and itching) scores. All analy-
ses were done using SPSS version 10.0 (SPSS, Inc., Chi-
cago, IL).
METHODS
RESULTS
Sixty-seven men were randomized to treatment. Twenty-
three men (33%) discontinued due to rash (n ꢂ 4), elevation
in prostate specific antigen level (n ꢂ 5), intercurrent illness
(n ꢂ 9), and personal reasons (n ꢂ 5) with no significant
differences between the groups. Forty-four men completed
the study: 24 men in the testosterone group and 20 men in
the placebo group. The baseline characteristics of the study
sample are presented in Table 1; no significant differences
existed in baseline information between the treatment and
control groups.
Following treatment, bioavailable testosterone increased
in the testosterone group from 93 ꢀ 34 ng/dl to 162 ꢀ 100
ng/dl (p ꢂ .002) and was unchanged in the placebo group
(101 ꢀ 26 ng/dl, p ꢂ .86). Estrone levels increased from
28 ꢀ 7 pg/dl to 32 ꢀ 9 pg/dl (p ꢂ .017) in the testosterone
treatment group but did not change in the placebo group
(p ꢂ .12). There were no significant changes in SHBG or
estradiol in either group.
Results of cognitive testing are summarized in Table 2.
Digit Symbol test scores improved in both groups. Trail-
making B scores improved in the testosterone group com-
pared to baseline (p ꢁ .01). No changes were found in
scores on Trailmaking B in the placebo group, and no sig-
nificant difference could be detected between scores from
the testosterone and placebo groups (p ꢂ .19). After con-
trolling for baseline scores, the change in Trailmaking B
score in the entire group was significantly associated with
12-month testosterone levels by linear regression analysis
(R2 ꢂ .18, p ꢂ .016).
Study Population and Design
Men were recruited to take part in a study to determine
the effects of testosterone on bone and muscle (24). As part
of that study, we also measured cognition and health per-
ception. Information on recruitment and eligibility are de-
scribed elsewhere (24). A total of 67 men were enrolled in
the study.
Men were randomized to receive 12 months of treatment
with either transdermal testosterone supplementation (5 mg/d)
using the nonscrotal testosterone patch or a placebo patch;
all men received 500 mg of calcium and 400 IU of cholecal-
ciferol. We measured total and bioavailable testosterone,
sex hormone binding globulin (SHBG), estrone, and estra-
diol, and we administered the Medical Outcome Survey
Short-form 36 (SF-36) and standardized tests of cognitive
function (Digit Symbol, Digit Span, Trailmaking A, Trail-
making B) at baseline and 12 months. We assessed skin ef-
fects of the transdermal patch and symptoms of itching us-
ing a 5-point scale (0 ꢂ no rash; 4 ꢂ erythema, induration,
and bullae) and a 4-point scale (1 ꢂ none; 4 ꢂ persistent),
respectively (25).
Biochemical Assays
Total and bioavailable testosterone and SHBG measure-
ments were performed at Endocrine Sciences Inc., Calaba-
sas Hills, California. Testosterone levels were measured by
radioimmunoassay, SHBG by competitive binding assay,
and bioavailable testosterone by competitive binding of the
non-SHBG-bound portion of testosterone following ammo-
nium sulfate precipitation of the SHBG-bound steroid as de-
scribed by Nankin (26). Estradiol and estrone were per-
formed using radioimmunoassay (Diagnostic Systems Lab,
Inc., Webster, TX).
Table 1. Baseline Characteristics of 44 Men Selected for Low
Testosterone Levels Completing 1 Year of Testosterone or
Placebo Supplementation
Statistical Analysis
Testosterone
Placebo
Baseline and clinical characteristics were reported using
means and standard deviations stratified by treatment group.
One-way analysis of variance (ANOVA) or chi-square anal-
ysis was used to test the difference in baseline characteris-
tics between the treatment groups. Wilcoxon signed ranks
test was used to assess change within groups over time, and
ANOVA was used to compare the percent change in health
perception or cognitive scores between groups. Multiple lin-
ear regression analysis was performed using 12-month
scores of the domains of the SF-36, controlling for baseline
scores, as the dependent variable; these domains included
physical function, role-physical, bodily pain, general health,
vitality, social function, role-emotional, and mental health.
The changes in cognitive test scores from baseline to 12
months were also entered as dependent variables, including
change in Digit Scan, Digit Symbol, Trailmaking A, and
Trailmaking B scores. Independent variables in the regres-
sion models were those variables found to correlate with the
dependent variables; variables entered into the correlation
matrix included sex hormone levels, physical activity,
Variable
n ꢂ 24
n ꢂ 20
Age (years)
Testosterone (nmol/l)
BioT (nmol/l)
Estradiol (pmol/l)
Estrone (pmol/l)
Calcium intake (mg/d)
Strength (Newtons)
Power (Watts)
Marital status (n married )
Education (% with college degree or higher)
History of depression
History of heart disease
History of hypertension
Antihypertensive treatment
Cholesterol-lowering agent
Antidepressant therapy
76 ꢀ 4
13.5 ꢃ 6.1
3.23 ꢀ 1.28
70 ꢀ 26
28 ꢃ 8
805 ꢀ 322
735 ꢀ 223
368 ꢃ 114
21 (87%)
12 (50%)
0
75 ꢀ 5
13.5 ꢃ 3.6
3.47 ꢀ 0.80
58 ꢀ 21
27 ꢃ 7
876 ꢀ 333
755 ꢀ 220
380 ꢃ 140
17 (85%)
15 (75%)
1
12
12
8
3
15
15
4
4
0
1
Notes: Values are mean and standard deviation. Comparison made either by
analysis of variance for continuous variables or chi-square analysis for dichoto-
mous variables. BioT ꢂ bioavailable testosterone.
p ꢁ .05.

TESTOSTERONE EFFECTS ON COGNITION AND HEALTH PERCEPTION
M323
Table 2. Comparison of Cognitive Test Scores Prior to and
Following 12 Months of Testosterone or Placebo Treatment
the Trailmaking B test. The inability to detect differences in
Trailmaking B scores between the testosterone and placebo
groups may have been due to the small sample size. How-
ever, an association between the 12-month testosterone lev-
els and the changes in scores of the Trailmaking B test sug-
gests that our results are not spurious. Others who used
Trailmaking B in their evaluations have not found differ-
ences or changes associated with testosterone levels or sup-
plementation, but the study populations were not selected
for low testosterone levels and are not directly comparable
with our study (12,14). Trailmaking B is a test that requires
both right and left cerebral hemisphere and overall general
brain functions for integration (27,28). Testosterone defi-
ciency may impair performance on this test, although this
theory has not been directly tested. Testosterone may also
exhibit its effect indirectly via estrogen after aromatization.
Estrone increased following testosterone supplementation
in our study while Janowsky and colleagues report a de-
crease in estrogen levels following testosterone replacement
(14). An inverse relationship between estrogen levels and
spatial cognition has been found in several studies of
women (29).
There is evidence of cognitive improvement associated
with higher testosterone levels in the literature. Barrett-Con-
nor and Goodman-Gruen, in a population-based study of
547 older men, found that older men with higher bioavail-
able testosterone levels had better scores on tests of verbal
memory and mental control (12). Morley and colleagues
similarly found bioavailable testosterone levels correlate
with cognitive abilty including tests of visual and verbal
learning, memory, and naming ability (13). Janowsky and
colleagues found that testosterone supplementation im-
proved spatial cognition, but the improvement was more
closely associated with the decrease in estradiol levels pos-
sibly due to central inhibition of gonadotropins; no differ-
ences were found between groups in verbal memory,
speeded cognitive flexibility, or fine motor dexterity (14).
In another study by Janowsky and Chavez (15), working
memory improved after 1 month of testosterone supplemen-
tation to scores of young men. Cherrier and colleagues also
reported significant improvement in spatial and verbal
memory in testing during intramuscular testosterone re-
placement (16). In contrast, Sih and colleagues (30) found
no change in memory in a small group of older hypogonadal
men receiving testosterone. Our work does not answer the
question of whether or by what mechanism testosterone af-
fects neurocognitive function but suggests, along with the
work of Janowsky and colleagues and Cherrier and col-
leagues, that further study of the role of testosterone on cog-
nition in older men is warranted.
Testosterone
Placebo
12 mo
11.8 ꢀ 1.8 12.4 ꢀ 1.9
Test
Baseline
12 mo
Baseline
Digit Span
11.4 ꢀ 2.6 11.5 ꢀ 2.5
Digit Symbol
Trailmaking A (sec)
Trailmaking B (sec)
42 ꢀ 8
42 ꢀ 14
104 ꢀ 39
46 ꢀ 9*
38 ꢀ 8
87 ꢀ 29*
43 ꢀ 8
39 ꢀ 16
95 ꢀ 30
47 ꢀ 7**
38 ꢀ 17
90 ꢀ 38
Notes: Values are mean and SD. No differences between groups by analysis
of variance.
*p ꢄ .01 compared to baseline; **p ꢄ .05 compared to baseline.
SF-36 scores are shown in Table 3. Overall, there were
few changes in health perception. Significant declines in the
general and mental health domains are seen in both groups,
and the changes were similar. Twelve-month bioavailable
testosterone levels predicted 12-month scores for role-phys-
ical (R2 ꢂ .39, p ꢂ .022), vitality (R2 ꢂ .17, p ꢂ .036), and
physical component composite (R2 ꢂ .17, p ꢂ .010) by lin-
ear regression analysis, after controlling for baseline scores.
In similar analysis, the 12-month score for bodily pain, after
correcting for baseline score, was negatively associated
with itching score (R2 ꢂ .08, p ꢂ .05).
There were no significant differences in rash irritation be-
tween the testosterone groups (1.9 ꢀ 0.3 vs 0.8 ꢀ 0.2; p ꢂ
.09), but 77% of the men in the testosterone group com-
pared with 40% of men in the placebo group reported a
score of 2 or greater, signifying a rash with induration. Itch-
ing was more frequent in the men in the testosterone group
compared with the placebo group (3.2 ꢀ 0.2 vs 2.6 ꢀ 0.2,
p ꢂ .001).
DISCUSSION
Treatment with testosterone did not result in overall
changes in cognition compared with placebo, though signif-
icant improvements from baseline scores were detected on
Table 3. Comparison of Scores on Individual Domains of Health
Perception Measured by Medical Outcome Survey Short-Form 36
Prior to and Following 12 Months of Testosterone or
Placebo Treatment
Testosterone
Placebo
12 mo
Variable
Baseline
12 mo
Baseline
Physical function
Role-physical
Bodily pain
General health
Vitality
Social function
Role-emotional
Mental health
Physical composite score 47 ꢀ 10
Mental composite score 58 ꢀ 5
79 ꢀ 25
73 ꢀ 42
76 ꢀ 26
79 ꢀ 15
69 ꢀ 18
96 ꢀ 10
93 ꢀ 22
89 ꢀ 9
78 ꢀ 23
71 ꢀ 39
72 ꢀ 31
71 ꢀ 24*
68 ꢀ 21
91 ꢀ 19
85 ꢀ 30
85 ꢀ 12** 88 ꢀ 9
47 ꢀ 10
56 ꢀ 8
81 ꢀ 21
64 ꢀ 41
74 ꢀ 23
72 ꢀ 18
67 ꢀ 17
96 ꢀ 11
96 ꢀ 11
77 ꢀ 16
59 ꢀ 37
74 ꢀ 20
67 ꢀ 17**
61 ꢀ 15
85 ꢀ 21
82 ꢀ 31
82 ꢀ 14*
44 ꢀ 10
54 ꢀ 10*
The men in this study reported baseline health perception
by SF-36 scores comparable to those of age- and gender-
matched subjects (31). These results are similar to other re-
ports of self-perceived health in older men with both normal
and low testosterone levels (21) and older men with low
bioavailable testosterone levels (22,23).
We expected testosterone supplementation to improve
health perception due to reports of improved mood and en-
ergy following testosterone supplementation in young, hy-
pogonadal men (17,18). However, two aspects of health
45 ꢀ 11
58 ꢀ 6
Notes: Values are mean and SD. No differences between groups by analysis
of variance.
*p ꢄ .05 compared to baseline; **p ꢄ .01 compared to baseline.

M324
KENNY ET AL.
perception, general health and mental health, declined over
12 months in both the placebo and treatment groups. Others,
similarly, have found no improvement in health perception
following testosterone or dihydrotestosterone replacement
in older men (22,23,32). The decline in health perception
scores may be due to the decline in health perception re-
ported with increasing age (31), though this is unlikely
given the relatively short duration of our study. More likely,
testosterone replacement may be associated with improved
health perception only in states of severe deficits as seen in
studies of younger, hypogonadal men. Further studies are
required to understand whether testosterone replacement is
associated with better health perception in this population.
Another explanation for the lack of effect of testosterone
therapy on self-perceived health may be due to the delivery
system. Both the testosterone and placebo patches were as-
sociated with skin irritation (40% in placebo group and 77%
in testosterone group) and itching. Itching predicted self-
perceived bodily pain at the 12-month survey, suggesting
that the rash and itching induced by the placebo and testos-
terone patches may be responsible for the overall lack of
improvement in health perception and for the deterioration
in self-perceived general and mental health in our study
sample. There has been another report of decline in quality-
of-life scores in a study involving allergic symptom induc-
tion (33). Other studies of transdermal testosterone or dihy-
drotestosterone delivery have also failed to demonstrate
change in health perception (22,32), but one study of 13
older, hypogonadal men receiving intramuscular testoster-
one reported a general increase in sense of well-being (34)
while another study of eugonadal men did not (23). Again,
further studies of testosterone supplementation using vari-
ous delivery systems are needed to determine whether there
are associated changes in health perception and whether
they are dependent on the mode of delivery.
Bioavailable testosterone levels at 12 months predicted
some aspects of health perception, including self-perceived
role limitations due to physical problems and self-perceived
vitality, as well as the physical component composite score.
This suggests that testosterone levels in older men may pos-
itively influence health perception associated with per-
ceived physical function. Snyder and colleagues also re-
ported improved self-perceived physical functioning in men
receiving testosterone compared with placebo (22). Testos-
terone has been implicated in the process of sarcopenia and
in the associated loss of muscle mass and strength (35), al-
though we did not find that strength predicted self-
perceived physical function scores. Further studies to
understand the mechanism by which testosterone predicts
physical function, other than by strength, are required.
The study has several limitations. The sample size is
small with a large number of men discontinuing prior to
study completion. This may have biased the results to only
the most robust individuals answering questions. While
baseline measures of changes in cognitive function were
only compared in the men who completed the study, we
may have missed an effect in less cognitively robust men.
Similar effects may have been missed in relation to health
perception.
available testosterone levels does not impair and may im-
prove cognitive function. Treatment did not improve health
perception, but this may have been due to the side effects of
skin irritation suggested by similar reactions in both the tes-
tosterone and placebo groups. Further studies of testoster-
one supplementation will be required to fully address the
role of testosterone in cognition and health perception in
older men with testosterone decline.
Acknowledgments
This work has been supported by the Patrick and Catherine Donaghue
Research Foundation, General Clinical Research Center (MO1-RR06192),
Claude Pepper OAIC (5P60-AG13631), and Dr. Kenny has been supported
with fellowships from the Brookdale Foundation and the Paul Beeson Fac-
ulty Scholar Program. In addition, we thank Pamela Fall and Christine
Abreu for assistance in running the biochemical assays and Julie Fenster for
assistance with data management. Testosterone and placebo patches were
provided by SmithKline Beecham at no charge to the study.
Address correspondence to Anne Kenny, MD, Center on Aging, MC-
5215, University of Connecticut Health Center, Farmington, CT 06030-
5215. Email: kenny@nso1.uchc.edu
References
1. Korenman SG, Morley JE, Mooradian AD, et al. Secondary hypogo-
nadism in older men: its relation to impotence. J Clin Endocrinol
Metab. 1990;71:963–969.
2. Khosla S, Melton LJ, Atkinson EJ, O’Fallon WM. Relationship of se-
rum sex steroid levels to longitudinal changes in bone density in young
versus elderly men. J Clin Endocrinol Metab. 2001;86:3555–3561.
3. Morley JE, Kaiser FE, Perry HM, et al. Longitudinal changes in tes-
tosterone, luteinizing hormone, and follicle-stimulating hormone in
healthy older men. Metabolism. 1997;46:410–413.
4. Flood JF, Morley PM, Morley JE. Age-related changes in learning,
memory, and lipofuscin as a function of the percentage of SAMP8
genes. Physiol Behav. 1995;58(4):819–822.
5. Gouras GK, Xu H, Gross RS, et al. Testosterone reduces neuronal se-
cretion of Alzheimer’s beta-amyloid peptides. Proc Natl Acad Sci
USA. 2000;97(3):1202–1205.
6. Alexander GM, Swerdloff RS, Wang C, et al. Androgen-behavior cor-
relations in hypogonadal men and eugonadal men. II. Cognitive
abilities. Horm Behav. 1998;33(2):85–94.
7. Vogel W, Broverman DM, Klaiber EL, Abraham G, Cone FL. Effects
of testosterone infusions upon EEGs of normal male adults. Electroen-
cephalogr Clin Neurophysiol. 1971;31(4):400–403.
8. Kertzman C, Robinson DL, Sherins RJ, Schwankhaus JD, McClurkin
JW. Abnormalities in visual spatial attention in men with mirror
movements associated with isolated hypogonadotropic hypogonadism.
Neurology. 1990;40:1057–1063.
9. Cappa SF, Guariglia C, Papagno C, et al. Patterns of lateralization and
performance levels for verbal and spatial tasks in congenital androgen
deficiency. Behav Brain Res. 1988;31(2):177–183.
10. Christiansen K, Knussmann R. Sex hormones and cognitive function-
ing in men. Neuropsychobiology. 1987;18(1):27–36.
11. Gouchie C, Kimura D. The relationship between testosterone levels
and cognitive ability patterns. Psychoneuroendocrinology. 1991;
16(4):323–334.
12. Barrett-Connor E, Goodman-Gruen D. Cognitive function and endog-
enous sex hormones in older women. J Am Geriatr Soc. 1999;47(11):
1289–1293.
13. Morley JE, Kaiser F, Raum WJ, et al. Potentially predictive and ma-
nipulable blood serum correlates of aging in the healthy human male:
progressive decreases in bioavailable testosterone, dehydroepiandros-
terone sulfate, and the ratio of insulin-like growth factor 1 to growth
hormone. Proc Natl Acad Sci USA. 1997;94:7537–7542.
14. Janowsky J, Oviatt K, Orwoll E. Testosterone influences spatial cogni-
tion in older men. Behav Neurosci. 1994;108(2):325–332.
15. Janowsky J, Chavez B. Sex steroids modify working memory. J Cog-
nitive Neuroscience. 2000;12(3):407–414.
Transdermal testosterone treatment in men with low bio-
16. Cherrier MM, Asthana S, Plymate S, et al. Testosterone supplementa-

TESTOSTERONE EFFECTS ON COGNITION AND HEALTH PERCEPTION
M325
tion improves spatial and verbal memory in healthy older men.
Neurology. 2001;57(1):80–88.
17. O’Carroll R, Bancroft J. Testosterone therapy for low sexual interest
and erectile dysfunction in men: a controlled study. Br J Psychiatry.
1984;145:146–151.
26. Nankin HR, Calkins JH. Decreased bioavailable testosterone in aging
normal and impotent men. J Clin Endocrinol Metab. 1986;63:1418–
1420.
27. Reitan RM. The relation of the Trail Making Test to organic brain
damage. J Consulting Psychol. 1955;19:393–394.
18. Wang C, Alexander G, Berman N, et al. Testosterone replacement
therapy improves mood in hypogonadal men—a clinical research cen-
ter study. J Clin Endocrinol Metab. 1996;81:3578–3583.
19. Burris A, Banks S, Carter S, Davidson J, Sherins R. A long-term, pro-
spective study of the physiologic and behavioral effects of hormone
replacement in untreated hypogondal men. J Androl. 1992;13:297–
304.
20. Anderson RA, Bancroft J, Wu FCW. The effects of exogenous testos-
terone on sexuality and mood of normal men. J Clin Endocrinol
Metab. 1992;75:1503–1507.
21. Dunbar N, Gruman C, Reisine S, Kenny A. Comparison of two health
status measures and their associations with testosterone levels in older
men. Aging Male. 2001;3:1–7.
22. Snyder PJ, Peachey H, Hannoush P, et al. Effect of testosterone treat-
ment on bone mineral density in men over 65 years of age. J Clin Med
Endocrinol. 1999;84:1966–1972.
23. Reddy P, White CM, Dunn AB, Moyna NM, Thompson PD. The ef-
fect of testosterone on health-related quality of life in elderly males—
a pilot study. J Clin Pharm Ther. 2000;25:421–426.
24. Kenny AM, Prestwood KM, Gruman CA, Marcello KM, Raisz LG.
Effects of transdermal testosterone on bone and muscle in older men
with low bioavailable testosterone levels. J Gerontol Med Sci. 2001;
56A:M266–M272.
28. Reitan RM. The validity of Trail Making Test as an indicator of or-
ganic brain damage. Percept Mot Skills. 1958;8:271–276.
29. Hampson E. Spatial cognition in humans: possible modulation by an-
drogens and estrogens. J Psychiatry Neurosci. 1995;20(5):397–404.
30. Sih R, Morley JE, Kaiser FE, Perry HM III, Patrick P, Ross C. Tes-
tosterone replacement in older hypogonadal men: a 12-month ran-
domized controlled trial. J Clin Endocrinol Metab. 1997;82(6):1661–
1667.
31. Ware JE. SF-36 Health Survey Manual and Interpretation Guide. Bos-
ton, MA: The Health Institute, New England Medical Center; 1993.
32. Ly LP, Jimenez M, Zhuang TN, Celermajer DS, Conway AJ, Handels-
man DJ. A double-blind, placebo-controlled, randomized clinical trial
of transdermal dihydrotestosterone gel on muscular strength, mobility,
and quality of life in older men with partial androgen deficiency. J
Clin Endocrinol Metab. 2001;86(9):4078–4088.
33. Ellis A, Day J, Lundie M. Impact on quality of life during an allergen
challenge research trial. Ann Allergy Asthma Immunol. 1999;83:33–
39.
34. Tenover JS. Effects of testosterone supplementation in the aging male.
J Clin Endocrinol Metab. 1992;75(4):1092–1098.
35. Morley JE. Editorial—Andropause: is it time for the geriatrician to
treat it? J Gerontol Med Sci. 2001;55A:M489–M491.
25. Meikle AW, Arver S, Dobs AS, Sanders SW, Rajaram L, Mazer NA.
Pharmacokinetics and metabolism of a permeation-enhanced testoster-
one transdermal system in hypogonadal men—influence of application
site—a clinical research center study. J Clin Endocrinol Metab. 1996;
81(5):1832–1840.
Received November 29, 2001
Accepted December 10, 2001