Anabolic-androgenic steroids are critical to the growth and development of the adolescent male. This article reviews the endocrine interrelationships at adolescence, discusses the pharmacology and toxicology of anabolic-androgenic steroids, reviews some of the epidemiological studies of abuse of these drugs in adolescents, and provides a view of some important ethical considerations surrounding this subject.
Growth and adolescent development have been extensively reviewed by Tanner and colleagues.1 These investigators have described a staging system that defines the normal progression of adolescent sexual development for girls and boys as well as describing some of the variations (especially in onset and tempo) that are considered to be within the range of normal. This staging system grades several aspects of pubertal development (genital and pubic hair development) individually. These investigators have also reported variability in the rate of pubertal progression among children. Thus, children differ in their entrance into puberty, in the length of time spent in a particular genital, pubic hair, or breast stage, and in the transition to the next stage. In general, however, the greatest variability is in the age of commencement of adolescent changes, and once begun, an orderly sequence of events is generally entrained. For example, in normal boys, testicular enlargement may begin at any time between 9.5 and 13.5 years, and breast budding in girls may begin between 8 and 13 years.
These structural changes are but the outward manifestations of the pubertal neuroendocrine developmental process and are roughly coordinated with the rising levels of circulating sex steroid hormones. The height spurt in boys occurs later in puberty than it does in girls. Menarche is a relatively late development of the pubertal process, attained well after the maximal growth velocity occurs. Gonadal steroid hormones are undoubtedly involved in the generation of this growth spurt, probably mediated by the effects of these hormones on the amount and secretory pattern of growth hormone (GH).
As normally growing boys enter puberty, there is a significant interaction between the rising androgen levels and GH secretion. This interaction is a major factor in the initiation and progression of the pubertal growth spurt.2 Although the proximal cause of the augmented GH secretion has been questioned (androgens acting through the androgen receptor or after aromatization to estrogens and activity through the estrogen receptor), there is little doubt that gonadal steroid hormones augment GH secretion during puberty in both sexes.2,6
In addition, there is little doubt that the GH secretory mechanism plays a central role in growth during the prepubertal period, but especially during the adolescent growth spurt. Bierich reported GH secretion during 5.5 hours of nocturnal deep sleep in 87 prepubertal children with constitutional delay of growth.7 The integrated (mean) GH concentration was diminished to 53% compared with healthy, normally growing (control) children. The highest peak GH concentrations were also lower (17 µg/L versus 37 µg/L). This disorder in GH concentration did not disappear as children entered and progressed through puberty, leading Bierich to conclude that the condition was permanent.
Augmentation of the circulating GH pattern is predominantly a pulse amplitude mediated phenomenon relatively independent of pulse frequency. It occurs due to the action of gonadal steroid hormones on the responsivity of the GH-secreting cells of the anterior pituitary to endogenous GH-releasing hormone, the amount of GH-releasing hormone, or to decreased tonic inhibitory tone from GH-release inhibiting hormone. The neurophysiologic mechanisms include:
* augmented daily production rate,
* greater mass of GH secreted per burst and a higher maximal rate of GH secretion within each burst, and
* amplified magnitude of the night/day rhythm in mass (but not frequency) of GH secretory events.
These effects of androgens on the mode of growth hormone secretion are apparently specific because the number and duration of GH secretory bursts and the subject-specific GH half-life are unaltered by androgen treatment. Similar conclusions (increased GH production rate without change in metabolic clearance) were presented by Mauras and colleagues5 for prepubertal girls with Turner's syndrome administered very low doses of ethinyl estradiol (100 ng/kg/day). Taken together, the data strongly suggest that early pubertal concentrations of gonadal steroids may be physiologically relevant to the maintenance and later augmentation of growth hormone secretion. The altered secretory dynamics rise to a maximum at approximately peak height velocity (data are more convincing for boys than girls) and then return to preand early pubertal levels as bony epiphyseal closure occurs. The alterations in GH are reflected in increasing levels of insulin-like growth tactor-I (IGF-I, also called somatomedin C)1 and the GH-dependent major IGF-I binding protein, IGFBP-3.
Anabolic-androgenic steroids can initiate adolescent development in the prepubertal male or augment the effects of endogenous androgens to accelerate delayed puberty. These drugs are all derivatives of testosterone, the natural male sex steroid hormone responsible for the androgenic and anabolic effects noted during male adolescence and adulthood. Androgenic refers to the effects that relate to the growth of the male reproductive tract and to the development of male secondary sexual characteristics. In the pubertal male, there are increases in the length and diameter of the penis, development of the prostate and scrotum, and the appearance of the pubic, axillary, and facial hair. Anabolic effects are those changes that occur in the somatic or nonreproductive tract tissues. These include an acceleration of linear growth that occurs before bony closure, enlargement of the larynx and thickening of the vocal cords, the development of libido and sperm production, and finally an increase in muscle bulk and strength. During normal male pubertal development, testosterone is responsible for the accelerated growth, increased muscle bulk (strength), and the decrease in body fat. This androgen is also probably responsible for the increase in aggressive and sexual behavior, although its role in psychological and behavioral personality characteristics is controversial.8
These drugs are medically unquestioned as replacement therapy in men with hypogonadism of central or peripheral origin and in some young men with marked delay of pubertal development. The amounts used either orally or by injection are within several-fold of the normal male daily testosterone production rate. At these levels, normal pubertal progression and the maintenance oí adult male strength and sexual function are obtained and untoward side effects are minimal. Somewhat increased amounts are used to stimulate anabolism or to produce positive metabolic. effects in some patients with neoplastic diseases. They may also be prescribed for patients with certain hormone dependent tumors, such as breast cancer.
Who Takes These Agents?
A large number of adolescents have been queried about anabolic-androgenic steroid use with anonymous surveys. The results have been summarized by Yesalis8 and are presented in the Table.
In 1987, the first nationwide study of anabolic steroid use at the high school level was conducted by Buckley and colleagues.9 They found that 6.6% of male high school seniors reported having used these agents (Table). There was no difference in the level of reported anabolic steroid use between urban and rural areas, however there was a small, but significant, difference in the size of enrollment at a particular school - students at larger high schools had higher rates of reported anabolic steroid use. Almost 40% of the anabolic steroid users reported five or more cycles of use (a cycle is an episode of use of anabolic steroids, which usually lasts from 6 to 12 weeks or more; see below).
In addition, of the self-reported anabolic steroid users, 38% had initiated use before age 16, 44% used more than one steroid at a time ( "stacking"), and 38% used injectable anabolic steroids. More than one-third of the anabolic steroid users did not intend to participate in interscholastic sports. Only 15% of the public and private high schools that participated in this study had no reported anabolic steroid use. However, in a survey of 472 head football coaches in Michigan high schools, only 12% admitted that they suspected or knew of at least one player who used anabolic steroids before 1988.8
Six studies at the state level (Table) have confirmed the findings of Buckley and coworkers,9 and they generally show that 5% to 12% of high school males admit having used anabolic steroids. These six studies also examined the use of anabolic steroids among high school females and found that approximately 1% to 2% reported using them. By contrast, one study of 295 high school athletes, including males, reported only a 1 .4% prevalence of anabolicandrogenic steroid use; however, distributing the questionnaire during a pre-participation examination may have led to significant underreporting. Recently, two other national surveys of high school seniors showed that approximately 5% of males and 0.5% to 1% of females acknowledged prior use of anabolic steroids; use in the prior 30 days was 1 .4% to 1 .7% and 0.1% to 0.3%, respectively (Table).
What Agents Are Taken?
There are three general classes of anabolicandrogenic steroids: testosterone, alkylated testosterones (oral), and alternatively structured agents, including testosterone esters (injectable) and 19 nortestosterone derivatives. All are available by legal prescription, but are more commonly obtained from black-market sources from which the identity and purity of the "active" agent may be suspect.8 Some anabolic steroids are manufactured primarily for veterinary use. Because most adolescents are not tested for the use of anabolic-androgenic steroids (as would be the case for some participants in national and international athletic contests), there is no worry about the rate of clearance from the body. Thus, the agents taken are usually those most readily available - often from friends, other steroid users, or sellers of other illicit drugs.
The alkylated testosterones are cleared less rapidly than testosterone itself, but are significantly more hepatotoxic as manifested by cholestatic jaundice peliosis of the liver (blood lakes), or, rarely, hepatoma. The latter may only occur in those with a propensity to develop malignancies, eg, individuals with Fanconi's and other anemias.8 The testosterone esters are more lipid soluble because of esterification with a fatty acid at the 17-hydroxyl position. This permits slower absorption from an intramuscular depot when injected in oil. The enanthate (hepanoate) and cypionate (cyclopentylpropionate) esters are the most common since they permit every 2- or 3-week administration. The only medical indication for these legitimate prescription drugs in adolescent boys is delayed puberty. In girls, there is good clinical evidence for efficacy (growth) in patients with Turner's syndrome, while in infant boys there is some evidence that anabolic-androgenic steroid therapy is effective in patients with microphallus.10
How Are These Agents Taken?
Anabolic steroids have traditionally been taken in "cycles," which are episodes of use of 6 to 12 weeks or more duration. However, there are athletes, such as some power weight lifters, who use the drugs on a relatively continuous basis and increase their dose at certain times of the year to prepare for a competition. Often athletes will take more than one steroid at a time; this is referred to as stacking. The purported rationale for stacking is that the user will activate more receptor sites than if only one steroid is used or that a synergistic effect can be achieved with certain combinations of steroids. To avoid developing tolerance to a particular anabolic steroid (plateauing), some users stagger the timing of their drugs and will take the steroids in an overlapping pattern or stop taking one drug and start taking another. Steroid users will often pyramid their dosing patterns such that they incrementally move from low daily doses at the beginning of the cycle to higher doses and then taper their doses downward toward the end of the cycle.
In addition, the athlete may use a number of other drugs concurrently with anabolic steroids to further enhance physical capacities or to counteract the common side effects of steroids. These drugs include stimulants, diuretics, antiestrogens, human chorionic gonadotropin (hCG), human growth hormone (hGH), and antiacne medications, as well as antiinflammatories. This polypharmacy is termed an array.8
What Are the Expected Physical Side Effects?
When administered in the usual therapeutic dose range, acne and gynecomastia can be expected, just as they appear in a large percentage of normally developing teenage boys. As the doses escalate, one may expect the more serious side effects of priapism (sustained penile erection) and edema secondary to sodium and chloride retention and even those noted above (eg, liver dysfunction), but these usually occur only after prolonged sustained high dose use - which may not be the usual pattern for the adolescent who wishes to get "big" and "strong* for sports or to "look good." In fact, there are relatively few significant side effects from the low-dose, intermittent-use patterns typical of adolescent athletes and nonathletes.
It is apparent that much of the liver toxicity, especially a consistent and substantial reduction in the ratio of high- to low-density lipoprotein cholesterol, is attributable to the 17-alkyl-substituted androgens (oral).8 Higher doses of any variety of anabolicandrogenic steroids would lead to premature epiphyseal closure and diminished adult height in growing children - not unlike the results noted in children with congenital virilizing adrenal hyperplasia in the precortisone period.11
There are profound and mainly irreversible effects for adolescent females. Clitorimegaly, hirsutism, and amenorrhea are common, although the menstrual dysfunction may be partially reversible following cessation of use. Effects on the larynx lead to a permanently deepened (masculine) voice.
Self-Reported Anabolic Steroid Use Among High School Students*
Are There Psychological and Behavioral Effects of Anabolic-Androgenic Steroids?
Perhaps the more serious side effects of anabolicandrogenic steroids occur in the behavioral sphere. The adolescent age group is at the transition to adulthood in terms of psychological make-up; therefore, during this period, the adolescent may be particularly vulnerable to the psychological effects of heightened aggression.
The psychologic and behaviorial aspects of the male personality have been noted for centuries, but studied only in an uncontrolled manner until the present century. In the last century, the field of endocrinology was highlighted by Brown-Sequard who attempted (he was his own experimental subject) to show that aqueous extracts of dog and guinea pig testes contained substances that in addition to their androgenic effects, positively affected vitality, energy, and youthfulness. Effects on mood and mental disorders were explored in the mid- to late 1900s and have recently been reviewed by Bahrke and coworkers.12 Several investigators have suggested that some of the purported ergogenic benefits of anabolic-androgenic steroids may derive from their psychological effects, ie, they may permit arousal, increased self-confidence and pain threshold, and facilitate expression of the "all-out" physical effort of training and competition. The majority of psychological/behavioral studies have been performed with small numbers of selected athletes, and the results are difficult to generalize. Some data indicate different mood changes following dynamic resistance exercise alone. These alterations in mood can be enhanced by athletes taking anabolicandrogenic steroids. Reduction in fatigue following exercise in athletes receiving anabolic-androgenic steroids has been cited in several investigations (mainly self-report information). However, as Bahrke and coworkers summarize: "Despite the suggestions and selfreports [italics added] scientific data supporting the notion that psychological changes (enhanced arousal, confidence, aggression, motivation) play a primary [italics added] role in mediating any ergogenic effects of anabolic-androgenic steroids is lacking."12
Are Affective and Psychotic Symptoms Associated With the Use of Anabolic-Androgenic Steroids?
For many years, anecdotes and even more formal studies have been reported that indicate that some violent men may have higher-than-average circulating levels of testosterone and that exogenous anabolicandrogenic steroids may increase natural aggressive tendencies. Several investigators have reported on the relationship between hormone levels (gonadotropins, gonadal steroid hormones, and adrenal androgens) and the emotional dispositions and aggressive behaviors of adolescents. The results indicate that high hormone levels were related to potentially adverse psychological consequences for boys and girls.12 None of these studies could be considered definitive based on design, numbers of subjects, inventories used, or variance in results.
While studies in lower animals appear more adequately controlled and do help explain naturally sexually dimorphic behavior, those in the human athlete designed to evaluate the effects of anabolicandrogenic steroids on affective and psychotic symptoms have not confirmed specific behavioral disturbances. In the main, these few studies lack statistical power and do not account for the "premorbid" condition of the athlete. Most do not account for the delivery of the pharmacologic agent, in that type, dose, "stacking" regimen, and concurrent use of other hormones or drugs of abuse were not carefully controlled. No study has shown that symptoms and concurrent use are causally related.
Thus, there is a great gulf between undoubted sexually dimorphic and aggressive behavior due to available androgenic steroids in lower animals and clinical studies in athletes. The absence of evidence is not the same as evidence of absence of effect, and more sophisticated longitudinal studies (probably placebo- and training regimen-controlled) are required to dissect the role of anabolic-androgenic steroids in behavioral disturbances.
Can Anabolie-Androgenic Steroids Produce a Psychological Dependence Syndrome?
Preliminary data from several investigators8 indicate psychological dependence on anabolicandrogenic steroids as demonstrated by: preoccupation with drug use, difficulty stopping despite psychological side effects, and drug craving. Purported withdrawal effects include mood swings, violent behavior, heightened aggression, and depression possibly severe enough to lead to thoughts of suicide. Such reports have led Kashkin and Kleber13 to propose an anabolicandrogenic steroid addiction hypothesis that the psychoactive effects, withdrawal symptoms, and underlying biological mechanisms of steroid hormones appear similar to the mechanisms and complications accompanying alcohol or opioid abuse. If this hypothesis is confirmed, then specific treatment modalities, eg, alpha-2 adrenergic agonist therapy as employed for abusers of alcohol or opioids, may be indicated. However, the possibility of an opioid-like withdrawal from anabolic steroids is supported by only one case study that has not been confirmed by others.8
How Should Pediatricians Approach the Issue of Anabolic-Androgenic Steroid Use by Their Adolescent Patients?
Our standard answer to such questions (as, for example, immunization and poisoning prevention) is to use anticipatory guidance and education, whether person-to-person or by any one of a number of audiovisual or literature aids. Pediatricians themselves must become knowledgeable about the drugs, their sources, the likely side effects, and the motivations of their patients for taking them. Especially important is to look for subtle side effects because most adolescents will not volunteer information about their (illegal) drug-taking behavior. Very rapid weight gain and increased muscle bulk should alert the pediatrician to the possibility of anabolic-androgenic steroid use. Increased acne and behavioral changes (especially heightened aggression and severe mood swings) ought to make the physician seriously consider steroid use. Would that it were so simple! One must remember that these signs and symptoms are but some of the more significant manifestations of the natural physical and behavioral pubertal changes that occur secondary to the rising levels of endogenous testosterone in the adolescent male. It is indeed difficult to factor out the abuse from the physiologic changes.
In adolescent girls the deepened voice and clitoral changes are major clues, but few adolescent girls have genital exams with each visit to the pediatrician. Although a relatively small percentage of adolescents participate in individual or team sports, the preparticipation examination is an ideal time to elicit both symptoms and signs of anabolic-androgenic steroid abuse. This is a focused time in which the physician should educate the patients concerning the dependency potential of anabolic-androgenic drug and address the issue that taking these drugs is a form of cheating in athletic competition.
Why not propose a general education program through the school system and athletic coaches as well as the primary care physicians? The difficult issues of appropriate curriculum and process must be anticipated; however, there is disquieting evidence for a counterproductive aspect of this type of educational program such as one directed at a group of high school football players in Oregon. Their propensity to use anabolic-androgenic steroids actually increased following certain educational programs at the high school level. On the other hand, a more recent study found that an educational program that emphasized alternatives to anabolic-androgenic steroid use, such as nutrition principles and strength training techniques, was more effective in improving attitudes toward potential anabolic-androgenic steroid use than either an educational program in which no alternatives were discussed or no intervention program proposed. Another study found an educational program that emphasized only the negative consequences of anabolic-androgenic steroid use (ie, scare tactics) was ineffective.8
The American Academy of Pediatrics (AAP) has expressed great concern over the use of anabolicandrogenic steroids in adolescent athletes. The AAP's position paper condemns the use of these agents because
... of their known toxic side effects, because of our belief that taking ergogenic drugs is another form of cheating and because competitors who enhance their athletic performance with anabolic steroids put the competitors in the difficult position of either not taking them and conceding a perceived advantage to the abusing competitor or taking them as well and accepting the risks of untoward side effects. Young athletes should not be placed in the situation of having to make such a choice.14
With the great emphasis on sports in our society, especially in the early pubertal years, one might expect nearly all adolescents, both male and female, to state that they wish to perform better in their chosen sport, perhaps as an avenue to scholarships or to professional teams. This phenomenon, in and of itself, is not problematic. However, when coupled with the "win at any cost" philosophy, which at times appears dominant in our society, the use of anabolic-androgenic steroids is an understandable, albeit regrettable, behavior. Our devotion to physical appearance, viewed as low body fat and muscularity, and our children's role models who embody these characteristics, provide additional incentive to use these drugs.
SUMMARY AND CONCLUSION
This article has reviewed some of the hormonal and behavioral maturation that occurs during adolescence, which are characterized by remarkable physical changes and behavioral vulnerability. Risk taking of many varieties is common and drugs (including anabolic-androgenic steroids) form a part of the prevailing culture in many places. These steroids probably are not severe health hazards when taken intermittently and in low to moderate doses. The 1 7-alkylated derivatives are clearly the more likely to cause hepatotoxicity. Thus, the scare tactics formerly used (severe constitutional side effects) are doomed to failure. The tenuous link between these drugs and objective behavioral and addictive effects must be strengthened before health strategies based on this issue can be validated. Clearly, the lack of scientific information has impeded, if not precluded, the formulation of an effective health education strategy.
The most potent deterrent to the use of steroid drugs by athletes must be the moral issue o( fair play and maintaining a "level playing field." We strongly support directed research in these areas and hope that the credibility of the scientific community can be regained after its faulted "stop steroid use" campaigns based on the lack of steroid efficacy in bringing about desired results or on their dire consequences have been replaced with credible evidence to refute their use on these and other grounds.
1. Tanner JM. Growth at Adolescence. 2nd ed. Oxford, England: Blackwell; 1962.
2. Martha PM Jr1 Rogol AD, Veldhuis JD, Kerrigan JR, Goodman DW, Blizzard RM. Alterations in the pulsatile properties of circulating growth hormone concentrations during puberty in boys. J Clin Endocrinol Metab. 1989;69:563-570.
3. Ulloa-Aguirre A, Christie CM, Garcia-Rubi E, et al. Testosterone and oxandrolone a non-aromatizable androgen specifically amplify the mass and rate of growth hormone (OH) release secreted per burst without altering GH secretory burst duration or frequency or the GH half-life, j Clin Endocrmoi Metab. 1990;71:846-854.
4. Mauras N, Blizzard RM, Link K, Johnson ML, Rogol AD, Veldhuis JD. Augmentation of growth hormone secretion during puberty: evidence for a pulse amplitudemodulated phenomenon. J Con Endocrinol Metab. 1987;64:596-601,
5. Mauras N, Rogol ?G? Veldhuis J. Increased hGH production rate after low dose estrogen therapy in pre-pubertal girls with Turner's syndrome. PediarrRes. 1990; 2 8:626· 630.
6. Rose SR, Municchi G, Barnes KL, et al. Spontaneous growth hormone secretion increases during puberty in normal girls and boys. J CIm Endocrinol Metab. 1991;73:428-435.
7. Blench JB. Constitutional delay of growth and adolescent development. In: Bereu BB, ed. Basic and Clinical Aspects of Growth Hormone. New Yotk, NY: Plenum Press; 1988:289-301.
8. Yesalis C. Anabolic Steroids in Exercise and Sport. Champaign, IU: Human Kinetics Publishing. In press.
9. Buckley W, Yesalis C. Friedl K. Anderson W, Streit A. Wright J. Estimated prevalence of anabolic steroid use among male high school seniors. JAMA. 1988;260:3441-3445.
10. Lee PA, Mazur T, Danish R, et al. Micropenis, 1: criteria, etiologies and classification. Johns Hopkins Medical Journal. 1980;146:156-163.
11. KUkland RT, Keenan BS, Holcomb JH, Kirkland JL, Clayton GW. The effect of therapy on mature height in congenital adrenal hyperplasia. J Clin Endocrinol Metab. 1978;47:1320-1324.
12. Bahrke MS, Yesalis CE III, Wright JE. Psychological and behavioral effects of endogenous testosterone levels and anabolic-androgenic steroids among males: a review. Sports Med. 1990;10:303-337.
13. Kashkin KB, Kleber HD. Hooked on hormones? An anabolic steroid addiction hypothesis. JAMA. 1989;262:3166-3170.
14- American Academy of Pediatrics. Anabolic steroids and the adolescent athlete. Pediatrics. 1989:83:127-128.
15. Corder BW, Dezelsky TL, Toohey JV, DiVito CL. Trends in drug use behavior at ten Central Arizona high schools, Arizona Journal of Health, Physical Education, Recreation end Dance. 1975;18:10-11.
16. Newman M. Michigan Consortium of Schools Student Survey. Minneapolis, Minn: Hazelden Research Services; 1986.
17. Polen L, Shnider L1 Sirotowicz A, West J. Teenage drug epidemics: Build up on steroids. Sword and Shield (South Plantation High School newspaper, Broward County. FL), October 1986.
18. Bosworth E, Bents R, Trevisan L, Goldberg L. Anabolic steroids and high school athletes. Med Sci Sports Exerc. 1987;20(suppl 3): 17.
19. Buckley W, Yesalis C, Friedl K, et al. Estimated prevalence of anabolic steroid use among male high school seniors. JAMA. 1988;260:3441-3445.
20. Johnson M, Jay M, Shoup B, et al. Anabolic steroid use in adolescent males. Pediatrics. 1989;83:921-924
21. Windsor R, Dumitru D. Anabolic steroid use by adolescents: Survey. Med Sci Sports Exerc. 1989;21:494-497.
22. Ross J, Winters F, Hartmann K, et al. I988-S9 Survey of Substance Abuse Among Maryland Adolescents. Baltimore, Md: Maryland Department of Health and Mental Hygiene, Alcohol and Drug Abuse Administration; 1989.
23. Ringwalt C. Alcohol ani Other Drug Use Patterns Among Students in Norm Carolina PubL· Schools, Grades 7-12: Results of a 1989 Student Survey. Raleigh, NC: North Carolina Department of Public Instruction, Alcohol and Drug Defense Section, Division of Student Services; 1989.
24. Krowchuk D, Anglin T, Goodfellow D, et al. High school athletes and the use of ergogenic aids. Am J Dis Child. 1989;143:486-489.
25. Hubbeil N. The Use of Steroids by Michigan High School Students and Athletes: An Opinion Research Study of i0th and 12th Grade High School Students and Varsity Athletes, November 1989 through January 1990. Lansing, Mich: Michigan Department of Pubitc Health, Chronic Disease Advisory Committee; 1990.
26. Temey R, McLain L. The use of anabolic steroids in high school students. Am J Dis Child. 1990;144:99-103.
27. Johnston L, Bachman J, O'Malley R Monitoring the Future: Continuing Study of the Lifestyles and Values of Youth. Ann Arbor, Mich: University of Michigan Institute for Social Research; 1990, 1991.
Self-Reported Anabolic Steroid Use Among High School Students*