ORIGINAL RESEARCH
Anabolic Steroid Use in Weightlifters and Bodybuilders
An Internet Survey of Drug Utilization
Paul J. Perry, PhD, Brian C. Lund, PharmD, MS, Michael J. Deninger, PhD,
Eric C. Kutscher, PharmD, and Justin Schneider, PharmD
Purpose: Dietary supplements and ergogenic agents, including
anabolic steroids, are common components of present-day body-
builder and weightlifter training regimens. Prior reports of anabolic
steroid use suggest polypharmacy and high doses of injectable agents.
Hypothesis: To provide an updated description of anabolic steroid
regimens employed by weightlifters and bodybuilders and to deter-
mine the extent to which anabolic steroid–associated behaviors are
consistent with substance dependence.
Study Design: Web-based survey.
Methods: Links to the Web-based survey instrument were estab-
lished from leading bodybuilding and fitness web pages. The ques-
tionnaire included demographic information, anabolic drug use
history, adverse effects, information sources, and steroid use behavior
consistent with criteria for a substance dependence disorder.
Results: A total of 207 subjects provided a detailed anabolic steroid
drug history. Steroid regimens included a mean of 3.1 agents, in-
volved cycles ranging from 5 to 10 weeks, and often included doses
5 to 29 times greater than physiologic replacement doses. Behavior
consistent with a substance dependence disorder was endorsed by
33% of respondents.
Conclusions: These findings suggest that anabolic steroid use
among weightlifters and bodybuilders continues, generally involving
multiple steroids and additional dietary supplementary agents. The
adverse effects, polypharmacy, large dosages, and risk of substance
abuse are all major health care concerns that require further study.
Clinical Relevance: The survey findings provide sports medicine
practitioners a reasonable estimate of the expected drug history among
bodybuilders and weightlifters for the use of performance-enhancing
agents.
Key Words: anabolic steroids, internet survey, drug utilization,
adverse effects, substance dependence, disorder
(Clin J Sport Med 2005;15:326–330)
Received for publication March 2005; accepted July 2005.
From the Department of Psychiatry, College of Medicine, and the College of
Pharmacy, University of Iowa, Iowa City, IA.
Reprints: Paul J. Perry, PhD, S-415 Pharmacy Bldg., University of Iowa, Iowa
City, IA 52242-1112 (e-mail: paul-perry@uiowa.edu).
Copyright Ó 2005 by Lippincott Williams & Wilkins
326
T he hepatic, cardiovascular, endocrine, and psychiatric
problems related to their use are well documented.1,2 De-
spite these potential adverse effects, prevalence studies,3 sub-
stance dependence surveys,4 and anabolic steroid usage surveys5
suggest that the associated risks are probably minimized. A
small local area survey of 20 anabolic steroid users in the late
1980s suggested a polypharmacy usage pattern.5 Oral steroids,
primarily because of hepatoxicity risk, were ingested at the
usual recommended doses, whereas injectable steroids were in-
gested in doses many times higher than the recommended phy-
siologic replacement dose.
The first goal of the current study was to provide an up-
dated description of anabolic steroid regimens employed by
weightlifters and bodybuilders, including the types of agents
used, how agents are used together, for what duration they are
used, and in what doses. The intent of this study was to gen-
erate a drug history that described steroid regimens and
associated characteristics among steroid users. It should be
emphasized that this study was not designed to generate pre-
valence rates of steroid use. The second goal was to determine
the extent to which anabolic steroid–associated behaviors are
consistent with substance dependence. To gather data from a
broader population than our previous study,5 we employed a
Web-based survey instrument.
MATERIALS AND METHODS
Participants
In the hope of achieving a large and geographically di-
verse survey of anabolic steroid use in the United States,
participants were surveyed using a Web-based application.
Potential participants were able to find the address for the
survey page through links or postings on various bodybuilding
and fitness web pages. The authors identified potential posting
sites through commonly used Internet search engines. Web
pages that had message boards regarding bodybuilding or
athletic supplementation were contacted by the authors to
request their participation with the survey. Numerous sites
declined due to lack of interest in participation or were
dropped from consideration because they required payment to
create a link to our survey page. Participating sites created
a direct link to our survey page, either as a banner ad on the top
of their web page or as a link in their message boards. The
survey page was available for a 3-month period from April
through June 2002.
Clin J Sport Med  Volume 15, Number 5, September 2005
Clin J Sport Med  Volume 15, Number 5, September 2005
Survey Instrument
The questionnaire consisted of 4 pages and was esti-
mated to take about 30 minutes to complete. The questionnaire
was composed of several primary areas that included demo-
graphic information, anabolic drug use history, adverse effects,
information sources, and steroid use behavior consistent with
DSM-IV-TR criteria for Substance Dependence Disorder,6 and
use of nonsteroid performance-enhancing agents. Geograph-
ical regions of residence were categorized based on the group-
ings used by the US Census Bureau. The survey instrument
can be obtained by contacting the corresponding author.
As privacy of submitted information was essential, every
known and reasonable measure was taken to ensure the security
of the data collected by the server. No individually identifiable
information was collected as part of the questionnaire. The
server running the Web site by normal course of operation col-
lected the IP address from the computer submitting information
to the server in a server log. An IP address could in relatively
rare circumstances be used to identify a particular computer
(though not who was using the computer). However, the IP ad-
dress was not collected in the dataset with the survey responses
and therefore could not be linked back to individual responses.
Furthermore, a certified secure connection made it nearly im-
possible for outside parties to intercept and decrypt data as it
was being submitted. The issue of IP address and other security
measures was presented to potential survey participants in
the introductory part of the Web-based survey, prior to the
collection of any questionnaire information. The University of
Iowa Institutional Review Board approved this research.
Statistical Methods
Between-group comparisons of continuous data were
performed using a 2-sample t test, and the Pearson x2 test was
used for categorical comparisons. All statistical tests were
2-sided, with a significance level of a = 0.05, and were per-
formed using Statview version 5.0 (SAS Institute Inc., Cary, NC).
RESULTS
Respondents
Of 260 overall survey responses submitted, 207 included
detailed steroid regimen information and are the subject of
this report. The self-reported steroid users were a mean of
27.2 years of age (n = 207; SD 6 7.2; range = 16–51), 208.2
pounds in weight (n = 207; SD = 31.2; range = 106–315), and
70.9 inches in height (n = 204; SD = 3.1; range = 60–80). They
lifted weights for a mean of 8.4 years (n = 198; SD = 6.0;
range = 1–21) for 7.4 hours per week (n = 180; SD = 3.4;
range = 1–21). Of the users, all specified at least 1 athletic
specialty, of whom 19 were competitive body builders, 16
competitive weight lifters, 174 noncompetitive body builders
or weight lifters, and 37 other specialties. Of respondents, 37
included themselves in multiple categories. The category of
other specialties included football (n = 15), hockey (n = 5),
wrestling (n = 4), rugby (n = 3), basketball (n = 2), boxing
(n = 2), beach volleyball, fighting, softball, military, all sports,
kickboxing, firefighter, and track (n = 1 each).
Of the 207 surveys, Figure 1 describes the 199 steroid
survey respondents who designated their geographical area of
q 2005 Lippincott Williams & Wilkins
Anabolic Steroid Use
residence. Of the people who responded to the survey and were
users of steroids, the primary concentrations were in the South
Atlantic (36,17%), the Middle Atlantic (35, 17%), and the Pa-
cific (33, 16%) regions. The fewest number of respondents
came from the New England (12, 6%), East South Central (11,
5%), and Mountain (9, 4%) regions. The distributions for the
other regions among respondents were West South Central, 27
(13%), East North Central, 21 (12%) and West North Central,
15 (7%). This distribution demonstrates that the survey en-
gaged nationwide participation and that the findings can be
generalized across the United States. However, our method-
ology does not allow estimates of steroid prevalence or com-
parisons of prevalence rates across geographic areas.
Figure 2 characterizes the anabolic steroid sources of
information used by the users. Of the 207 respondents, 185
completely answered the question, ‘‘How frequently do you
get information about anabolic steroids from the following
sources?’’ with the categorical options being ‘‘a) Internet, b)
Books and magazines, c) other steroid users, and d) health
care providers.’’ The categorical responses were ‘‘often,’’
‘‘seldom,’’ and ‘‘never.’’ Unsurprisingly, since this was a Web-
based survey, the Internet was the most commonly reported
source of anabolic steroid information reported in this sample.
Obtaining information from other anabolic steroid users and
from books and magazines were also rather common. In con-
trast, information from health care providers was uncommon.
Usage Patterns of
Performance-Enhancing Agents
Use of anabolic steroids and other ergogenic agents
in the training regimens of weightlifters is a prime example of
frequent polypharmacy among the bodybuilding population.
FIGURE 1. Geographic distribution of anabolic steroid survey
respondents ascertained by 199 of 207 anabolic steroid using
bodybuilders for nine areas of the United States: East North
Central (ENC = IN, IL, MI, OH, WI), East South Central (ESC =
AL, KY, MS, TN), Middle Atlantic (MA = NJ, NY, PA), Mountain
(MTN = AZ, CO, ID, MT, NM, NV, UT, WY), New England (NE =
CT, MA, ME, NH, RI, VT), Pacific, (PAC = AK, CA, HI, OR, WA),
South Atlantic (SA = DC, DE, FL, GA, MD, NC, SC, VA, WV),
West North Central (WNC = IA, KS, MN, MO, ND, NE, SD), and
West South Central (WSC = AR, LA, OK, TX).
327
Perry et al Clin J Sport Med  Volume 15, Number 5, September 2005

injectable steroids was 9.1 weeks. Thus, the duration of the

injectable steroid cycles was 53% longer than the oral steroid
cycles. Mean doses of oral and injectable steroid products are
also presented in Table 2.
The nonsteroidal ergogenic agents included clomiphene
citrate (Clomid; 30/207), anastrozole (Arimidex; 12/207),
clenbuterol (12/207), liothyronine or T3 (Cytomel; 11/207),
human growth hormone (Genotropin, Humatrope, Nutropin,
and so forth; 11/207), insulin (11/207), tamoxifen (Nolvadex;
9/207), human chorionic gonadotropin (Novarel, Pregnyl,
Profasi; 6/207), metformin (Glucophage; 2/207), and finas-
teride (Proscar, Propecia; 2/207). Tamoxifen and clomiphene
citrate are selective estrogen receptor modulators developed
as antiestrogenic agents for the treatment of breast cancer,
but they are often used by bodybuilders to minimize possi-
ble side effects of excess estrogen such as gynecomastia in
addition to stimulating testosterone production. Clomiphene
FIGURE 2. Sources of information regarding anabolic steroids also binds to estrogen receptors to produce an antiestrogenic
described by 185 of 207 bodybuilders. effect. Anastrozole is a nonsteroidal aromatase inhibitor of
the conversion of androstenedione to estrone and testosterone to
Of the 207 respondents who provided the detailed steroid drug estradiol. It is often used as an alternative to tamoxifen to
histories, their current or most recent anabolic steroid cycle suppress estradiol. Clenbuterol is a b-adrenergic agonist used in
included a mean of 3.1 agents (SD = 1.5; range = 1–9). Nan- asthma that has also been shown to increase muscle strength.7
drolone decanoate and Sustanon, a combination product of 4 Of the 207 respondents, 126 described their use of die-
testosterone esters, were the most commonly reportedly used tary supplements. Of these, over 2/3 of the subjects used cre-
agents (Table 1). Almost half of the 207 steroid regimen re- atine, whereas glutamine and protein were reported in about
spondents in the survey used either Sustanon, nandrolone 1/3 of supplement regimens (Table 3). Ephedrine-containing
decanoate, or both, with about 1/4 using testosterone propi- products (Xenadrine, ECA stacks, Ripped Fuel, and so forth)
onate, stanozolol (Winstrol), and boldenone (Equipoise). Non- were also common in that about 1/4 of all participants used
steroidal ergogenic agents were less common, at a mean of 0.5 these products as part of their daily regimens.
agents (SD = 0.9; range = 0–5) among steroid users. Addition-
Adverse Physical Effects
ally, the 128 steroid users who gave detailed dietary supple-
ment histories on average were taking 2.7 (SD = 1.6) supple- Anabolic steroids can cause numerous adverse physical
ments in their training regimens. Thus, the typical respondent effects, with the most commonly reported events acne, testic-
reported use of 6 to 7 agents to sustain their training program. ular atrophy, loss of hair, and gynecomastia.1,8 Adverse phy-
As presented in Table 2, steroid cycles ranged from 5 sical effects reported by our survey respondents were consistent
to 10 weeks, depending on the particular steroid being used. with prior findings. Among the 207 subjects, 1/3 reported acne
The weighted mean cycle length for the 3 oral steroids was (35%; n = 72). Other adverse effects mentioned included water
5.9 weeks, whereas the weighted mean cycle length for the retention (10%; n = 20), testicular atrophy (8%; n = 17),
gynecomastia (7%; n = 14), oily skin (5%; n = 11), hair loss
(4%; n = 9), increased blood pressure (4%; n = 8), cramps (3%;
n = 6), increased appetite (2%; n = 4), hair growth (2%; n = 5),
TABLE 1. Steroid Agents Being Utilized in Training Program joint soreness/pain (2%; n = 4), and abnormal lipid panels
of 207 Bodybuilders (2%; n = 4). Some adverse effects reported by 3 or fewer re-
Steroid/Ergogenic Users (%) Dosage Form spondents were soreness from injection, headache, bloating,
Nandrolone decanoate circulation problems, increased perspiration, erectile dysfunc-
(Deca-Durabolin) 97 (46.9) Injectable tion, and increased heart rate, night sweats, and muscle
Testosterone propionate, soreness/tension.
phenylpropionate, isocaproate,
and decanoate (Sustanon) 83 (40.1) Injectable
Psychologic Effects
Boldenone (Equipoise) 50 (24.2) Injectable (veterinary)
Anabolic steroid use can affect mood and psychologic
Stanozolol (Winstrol) 46 (22.2) Oral
well being.9 Mood alterations, which can potentially be dan-
Trenbolone (Fina) 44 (21.3) Injectable (veterinary)
gerous, were common in our respondents. Of the 207 subjects,
Testosterone (not specified) 38 (18.4) Injectable
the majority (54%; 111) answered yes. Marked agitation or
Testosterone cypionate 28 (13.5) Injectable
irritability was reported in a minority of patients (11%; 23), as
Testosterone enanthate 25 (12.1) Injectable
Testosterone propionate 22 (10.6) Injectable
was the case with aggression (11%; 22). Other common mood
changes reported were characterized as increased mood (12%;
Methandrostenolone (Dianabol) 21 (10.1) Oral
24), increased libido (11%; 22), mood swings (6%; 13), de-
Oxandrolone (Anavar) 9 (4.3) Oral
creased mood (6%; 13), insomnia (6%; 12), decreased libido

328 q 2005 Lippincott Williams & Wilkins

Clin J Sport Med  Volume 15, Number 5, September 2005 Anabolic Steroid Use

TABLE 2. Anabolic Steroid Dosage and Usage Profile in 207 Bodybuilders

Usage/Cycle, Usage/Cycle,
Dosage Form Drug n wk Mean (SD) Dose (mg) Mean (SD)
Oral Methandrostenolone 21 5.1 (1.8) 39.8 (12.3)/d
Oxandrolone 9 8.7 (3.8) 26.9 (12.0)/d
Stanozolol 51 5.8 (2.0) 53.4 (23.7)/d
Injectable Nandrolone decanoate 97 9.8 (2.0) 423.3 (195.4)/wk
Testosterone cypionate 28 9.1 (2.4) 583.0 (301.1)/wk
Testosterone mix 83 9.6 (5.5) 516.0 (323.9)/wk
(Sustanon)
Testosterone enanthate 25 10.1 (3.3) 564.0 (211.9)/wk
Testosterone propionate 22 7.8 (4.4) 155.8 (54.6) QOD
Veterinary injectable
agents Boldenone 50 9.2 (7.0) 507.3 (300.8)/wk
Trenbolone 44 6.8 (2.1) 117.1 (48.7) QOD

(4%; 8), and euphoria (4%; 8). The increased mood and libido
commonly were noted while on steroids, whereas the de-
creased mood and libido were commonly described as steroid
withdrawal symptoms.
Substance Dependence
Among the 207 steroid use history respondents, 206
completed the section of the questionnaire that investigated
addictive patterns of use. The frequency of endorsement for
each criterion is provided in Table 4. Most commonly reported
were time related to acquisition (37.9%), escalated and pro-
longed use (28.2%), tolerance (26.7%), and withdrawal prob-
lems (26.7%). Three or more symptoms, consistent with a
DSM-IV-TR6 diagnosis of substance dependence, were re-
ported by 33.0% of the respondents.
Several steroid use characteristics were contrasted be-
tween respondents classified as substance dependent versus
nondependent, including years of steroid use, number of
stacking cycles, cycle length, number of ergogenic agents used,
years lifting, and hours per week lifting. The only significant
difference was mean stacking cycle length, which was 10.2
(SD = 2.4) weeks and 9.4 (SD = 2.5) weeks among substance-
dependent and nondependent steroid users, respectively (t =
2.111; df = 192; P = 0.0361).
DISCUSSION
In 1990, Perry et al5 examined the anabolic steroid use
histories of 20 competitive and noncompetitive weight lifters
in eastern Iowa. These 20 steroid users consistently practiced
polypharmacy. The subjects’ steroid use cycles ranged
TABLE 3. Dietary Supplement Use in Anabolic Steroid Users
Described in Detail by 126 of 207 Weightlifters
Dietary Supplement n (%)
Creatine 86 (70)
Protein supplements 43 (34)
Glutamine 42 (33)
Vitamins 18 (14)
Androstenedione 8 (6.3)
in duration from 7 to 14 weeks. The weightlifters commonly
used 2 or 3 oral agents and 2 long-acting injectable products.
Doses of oral preparations were consistent with pharmaceu-
tical manufacturer prescribing information recommendations,
whereas the doses of the long-acting injectable generally
ranging from 300 to 800 mg/wk5 were many times greater than
the physiologic dose of 28 to 56 mg/wk of testosterone.2
The goal of the current study was to provide an update
on usage patterns of individuals using anabolic steroids for
bodybuilding. Survey respondents used, on average, 3 anabolic
steroid agents in addition to 3 other ergogenic supplemental
agents. The weightlifters also tended to use a combination of
oral and long-acting injectable products in their steroid regi-
mens. The use of multiple agents shows current anabolic
steroid users continue to practice polypharmacy; however,
steroid cycles seem to be slightly shorter than previously
shown in the study by Perry et al.5 Similar to the previous
survey,5 current oral steroids such as oxandrolone and
stanozolol are being dosed at the package insert–recommended
TABLE 4. Self-Reported Symptoms of Substance
Dependence Described in 206 of 207 Anabolic
Steroid–Using Bodybuilders
DSM-IV-R Criteria for Substance Dependence n %
Spending great deal of time in activities
related to obtaining steroids 78 37.9
Taking larger amounts of steroids over longer
times than originally intended 58 28.2
Need increased amounts of steroids to achieve
the desired effect 55 26.7
Physical or emotional problems when steroid was stopped 55 26.7
Decreased effect with use of the same dose of steroid 42 20.4
Continued use of steroids despite experiencing
physical, emotional, or social problems 31 15.0
Restarting steroid to relieve problem that
occurred when you stopped 26 12.6
Quitting or reducing social, occupational, or
recreational activities because of steroid use 21 10.2
Having the desire to or making unsuccessful
efforts to decrease amounts taken 9 4.4
Meets criteria for substance dependence ($3 symptoms) 68 33.0

q 2005 Lippincott Williams & Wilkins 329

Perry et al
doses. However, the injectable steroids continue to be routinely
administered at doses many times greater than the replace-
ment dose of 28 to 56 mg/wk of testosterone.2 Unfortunately,
this study was not designed to estimate prevalence rates of
anabolic steroid use, or whether rates have increased over time.
However, this study demonstrates that, among anabolic steroid
using weightlifters, the doses and duration of regimens have
changed little over the past 10 to 15 years.
Brower et al4 described self-reported symptoms of ste-
roid dependence in 49 steroid users, in accordance to DSM-III-R
criteria.10 In the current study, 206 of the 207 respondents
completed the substance dependence section of the question-
naire. Overall, the current study observed fewer participants
meeting criteria for substance dependence (68/206; 33.0%),
defined as the endorsement of 3 or more criteria, compared
with the study by Brower et al4 (28/49; 57%; x2 = 8.820;
P = 0.003). The differences observed between the studies may
be a function of the slightly different definition of the 9 depen-
dence items in the DSM-IV-TR6 versus the DSM-III-R.10 It
also may indicate a change in societal steroid dependence, or
the differences may be the result of different sample sizes and
study population diversity.
There are several potential strengths and limitations with
using a Web-based survey that need to be considered. The
greatest strength of this study was its nationwide recruitment,
in contrast to our earlier report of 20 users in eastern Iowa or
other studies based in specific locations. Survey responses
were obtained from every region of the United States, sug-
gesting that our results reflected steroid regimen use nation-
wide, rather than being limited to a single isolated region. One
limitation, however, is that requiring response via the Web
necessitates access to a computer, which may have imposed a
bias toward a higher socioeconomic subpopulation of steroid
users. A further consideration is the anonymity of responses
allowed by Web-based data collection, which could produce
both positive and negative effects. Unfortunately, it is un-
known whether responders are more likely to provide distorted
answers regarding steroid use during an interview or during an
anonymous self-report on the Web. There did not appear to be
significantly distorted information regarding steroid dosing,
duration, and so forth, that suggested artificial responses. Fur-
thermore, there were no duplicate entries that suggested mul-
tiple responses by 1 individual. A better understanding of the
balance between these specific strengths and weaknesses will
develop as researchers expand the use of Web-based technology.
This study is also subject to the usual limitations and po-
tential biases involved with survey designs. Individuals who
choose to respond to a survey are often different from those
who do not. For example, ultrahigh-dose users, individuals
with a livelihood dependent on athletic performance, users
with or at risk for significant legal problems, or paranoid users
may be less likely to respond to the survey. As a result, the
findings of this study may not be generalizable to all subgroups
of steroid users. One typical method for assessing the potential
impact of this limitation is by reporting the participation rate.
As participation declines from 100%, the degree of general-
izability declines. Because there was no individual-level invi-
tation to participate in the study, there was no way to determine
330
Clin J Sport Med  Volume 15, Number 5, September 2005
the denominator for a participation rate. In other words, there
was no way to determine how many individuals viewed the
survey but chose not to respond. Of the 319 individuals who
submitted any response to the survey, only 207 (65%) offered
complete steroid regimens including drug, dose, and dura-
tion. Furthermore, only 129 (40%) steroid-using respondents
listed their nonsteroid supplement regimen in detail, and
only 163 (51%) completed the substance dependence section.
An additional consideration is that participation was purely
voluntary and without financial or any other nonaltruistic
compensation.
CONCLUSIONS
Weightlifters and bodybuilders have always kept to a
strict exercise regimen, incorporating performance enhancing
supplements and proper diet, to maintain the level of perfor-
mance that is desired. This survey updates previous work and
suggests that anabolic steroid use among weightlifters con-
tinues. Steroid users often took multiple agents at high doses
for extended periods of time, which indicates a disregard for
the significance of steroid adverse effects. The respondents
often described their steroid-related adverse effects in graphic
detail and used agents to counteract these effects. The adverse
effects findings, coupled with substance dependence findings,
suggest that this cohort of steroid users do not find credible the
public health warnings associated with the misuse of anabolic
steroids, because the bodybuilders and weightlifters are using
these drugs outside the readily available recommended guide-
lines suggested in manufacturers prescribing information. Fur-
ther study is warranted to evaluate the healthcare impact of
anabolic steroid polypharmacy and the potential for substance
dependence.
REFERENCES
1. Blue JG, Lombardo JA. Steroids and steroid-like compounds. Clin Sports
Med. 1999;18:667–698.
2. Kretser DM. Morphology and physiology of the testis. In: Becker KL, ed.
Principles and Practice of Endocrinology and Metabolism. 2nd ed.
Philadelphia, PA: JP Lippincott Co; 1995:1032–1041.
3. Johnston LD, O’Malley PM, Bachman JG. Monitoring the Future
National Survey Results on Adolescent Drug Use: Overview of Key
Findings, 2002 (NIH Publication No. 03-5374). Bethesda, MD: National
Institute on Drug Abuse; 2003:61.
4. Brower KJ, Blow FC, Young JP, et al. Symptoms and correlates of
anabolic-androgenic steroid dependence. Br J Addict. 1991;86:759–768.
5. Perry PJ, Anderson KH, Yates WR. Illicit anabolic steroid use in
weightlifters: a case series analysis. Am J Sports Med. 1990;18:422–428.
6. American Psychiatric Association. Diagnostic and Statistical Manual of
Mental Disorders. 4th ed, Text Revision. Washington, DC: American
Psychiatric Association; 2000.
7. Matlin CA, Delday MI, Watson JS, et al. Clenbuterol, a beta-adrenoceptor
agonist, increase relative muscle strength in orthopedic patients. Clin Sci.
1994;84:651–654.
8. Kutscher EC, Lund BC, Perry PJ. Anabolic steroids: a review for the
clinician. Sports Med. 2002;32:285–296.
9. Pope HG, Katz DL. Psychiatric and medical effects of anabolic-androgenic
steroid use: a controlled study of 160 weightlifters. Arch Gen Psychiatry.
1994;51:375–382.
10. American Psychiatric Association. Diagnostic and Statistical Manual of
Mental Disorders. 3rd ed, Revised. Washington, DC: American Psychi-
atric Association; 1987.
q 2005 Lippincott Williams & Wilkins