athletes and steroids research paper

• Subscribe to journal Subscribe
• Get new issue alerts Get alerts

Secondary Logo

Journal logo.

Colleague's E-mail is Invalid

Your message has been successfully sent to your colleague.

Save my selection

Anabolic-Androgenic Steroid Use in Sports, Health, and Society

BHASIN, SHALENDER; HATFIELD, DISA L.; HOFFMAN, JAY R.; KRAEMER, WILLIAM J.; LABOTZ, MICHELE; PHILLIPS, STUART M.; RATAMESS, NICHOLAS A.

1 Department of Medicine, Brigham and Women’s Hospital, Boston, MA

2 Department of Kinesiology, University of Rhode Island, Kingston, RI

3 Department of Physical Therapy, Ariel University, Ariel, Israel

4 Department of Human Sciences, The Ohio State University, Columbus, OH

5 InterMed, P.A., South Portland, ME

6 Department of Pediatrics, Tufts University School of Medicine, Boston, MA

7 Department of Kinesiology, McMaster University, Hamilton, ON

8 Department of Health and Exercise Science, The College of New Jersey, Ewing, NJ

Address for correspondence: Stuart M. Phillips, Ph.D., F.A.C.S.M., Department of Kinesiology, McMaster University Ivor Wynne Centre 1280 Main St, West Hamilton, Ontario, Canada L8S 4K1; E-mail: [email protected] .

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site ( www.acsm-msse.org ).

This consensus statement is an update of the 1987 American College of Sports Medicine (ACSM) position stand on the use of anabolic-androgenic steroids (AAS). Substantial data have been collected since the previous position stand, and AAS use patterns have changed significantly. The ACSM acknowledges that lawful and ethical therapeutic use of AAS is now an accepted mainstream treatment for several clinical disorders; however, there is increased recognition that AAS are commonly used illicitly to enhance performance and appearance in several segments of the population, including competitive athletes. The illicit use of AAS by competitive athletes is contrary to the rules and ethics of many sport governing bodies. Thus, the ACSM deplores the illicit use of AAS for athletic and recreational purposes. This consensus statement provides a brief history of AAS use, an update on the science of how we now understand AAS to be working metabolically/biochemically, potential side effects, the prevalence of use among athletes, and the use of AAS in clinical scenarios.

 This consensus statement is an update of the previous position stand from the American College of Sports Medicine (ACSM), published in 1987 ( 1 ). Since then, a substantial amount of scientific data on anabolic-androgenic steroids (AAS) has emerged and the circumstances of AAS use has evolved in the athletic, recreational, and clinical communities. The objective of this consensus statement is to provide readers with a brief summary of the current evidence and extend the recommendations provided in the 1987 document ( 1 ). Key topics discussed are the brief history of AAS, epidemiology, methods, and patterns of AAS use, androgen physiology and ergogenic effects, side effects of AAS, and clinical uses of AAS (see Box 1). The writing group used the rating system of the National Heart Lung and Blood Institute ( Table 1 ) and a consensus approach to synthesize the available evidence from clinical trials and case reports, narrative and systematic reviews, and meta-analyses ( 3 ). The recommendations represent the consensus of the writing panel, the ACSM, and incorporate guidance from other professional organizations with expertise in the area.

BOX 1. ACSM Consensus Statements and Recommendations Summary.

Consensus Statements and Recommendations

• 1. The administration of AAS in a dose-dependent manner significantly increases muscle strength, lean body mass, endurance, and power. The effects are primarily seen when AAS use is accompanied by a progressive training program. Evidence Category A .
• 2. Historically, AAS use was primarily seen in competitive athletes and aspiring bodybuilders and powerlifters. Recreational AAS use appears to have surpassed athletic AAS use indicated by survey prevalence estimates demonstrating that recreational trainees are the leading consumers of AAS. The ACSM deplores the illicit use of AAS for recreational purposes. Evidence Category C .
• 3. AAS are classified as schedule III drugs, banned by several sport governing bodies, and are illegal to use for athletic purposes. The ACSM deplores the illicit use of AAS for recreational use and performance enhancement in athletes. Evidence Category D .
• 4. Coaches, trainers, and medical staffs should monitor and be cognizant of visible signs of AAS use and abuse. These include (but are not limited to): a substantial increase in muscle mass, strength, and power in a relatively short period of time (or the reverse which could denote AAS withdrawal); acne that is resistant to medical treatment; development of unexplainable rash, gynecomastia, increased body hair, and prominent increases in surface vascularity; changes in temperament, mood, and aggressive behavior (severe depression or suicidality could indicate AAS withdrawal); facial masculinization and fluid retention; and muscle mass that appears disproportionate to body structure or pubertal status in young athletes. In addition, the presence of AAS-related materials (books, articles, websites, dealer information, needles, vials) on the individual could reflect intent and may warrant further dialogue from the coaching, trainer, and medical staffs. Medical staff should be aware of regulations and documentation requirements regarding use of AAS for athletes with medical indications for their use. Evidence Category C .
• 5. Use and abuse of AAS is associated with several notable adverse effects in men and women including (but not limited to) suppression of the hypothalamic-pituitary-gonadal axis, psychological changes, immunosuppression, and unhealthy cardiovascular, hematological, reproductive, hepatic, renal, integumentary, musculoskeletal, and metabolic effects. Several adverse effects may be reversible upon discontinuation but some could pose health risks beyond the duration of AAS use. Evidence Category B .
• 6. Use of AAS in prepubertal and peripubertal children may lead to early virilization, premature growth plate closure, and reduced stature. Evidence Category C .
• 7. Coaches, trainers, and medical staffs should be cognizant of the reasons for AAS use and abuse and deter use when possible. Prevention programs based on education may assist; and providing the individual with scientific nutrition and training advice is a recommended strategy to mitigate the temptation of AAS use. Evidence Category D .
• 8. Androgen replacement therapy is approved for the medical treatment of several clinical diseases and abnormalities. The ACSM acknowledges the lawful and ethical use of AAS for clinical purposes and supports the physicians’ ability to provide androgen therapy to patients when deemed medically necessary. The reader is referred to guidelines established by the Endocrine Society ( 4 ). Evidence Category C .

INTRODUCTION

Anabolic-androgenic steroids are drugs chemically and pharmacologically related to testosterone (T) that promote muscle growth and are not estrogens, progestins, or corticosteroids. An androgen is any natural or synthetic steroid hormone capable of promoting the development of male primary and secondary sexual characteristics via binding to androgen receptors at the tissue level. The term anabolic describes a hormone or other substance capable of enhancing the growth of somatic tissue, such as skeletal muscle and bone. In a sport-related setting, this is typically used to describe the enhancement of skeletal muscle. Table 2 presents nomenclature associated with AAS. In the United States, AAS are classified as Schedule III controlled substances ( 5 ). Although AAS have legitimate medicinal use, nontherapeutic use among athletes and recreationally active young men and women is performed to improve strength, power, increase muscle mass, and improve appearance. Athletic and recreational (i.e., noncompetitive) use of AAS has been widespread over the last 50 yr, creating considerable interest by the scientific and medical communities, as well as sport governing bodies, in examining the potential medical, legal, and ethical issues surrounding the use of these substances. All major national and international sports organizations have banned the illicit use of AAS by athletes.

HISTORICAL PERSPECTIVES

Anabolic-androgenic steroids use has been examined extensively in various chapters, books, meta-analyses, and reviews ( 5–12 ). The effects of testicular extracts and castration on animals and humans have been a source of fascination for thousands of years ( 13,14 ). Suggestions that the consumption of testis tissue could improve impotence were noted ~140 BC ( 13 ). The mid 1700s to late 1800s marked a time where interest in testicular endocrinology increased ( 14 ). Table 3 depicts a brief historical timeline of some key events in AAS use in athletes. Testosterone was synthesized and biochemically described in the late 1920s and 1930s, and a host of different synthetic variations have been developed since ( 5,15,16 ). Testosterone or AAS use by athletes began in the 1940s and 1950s, and increased considerably thereafter, culminating in high usage during the 1968 Olympic Games ( 5,6 ). It has been speculated that the first appearance of AAS use among female athletes dates back to the late 1950s/early 1960s in Soviet track and field athletes ( 17 ).

The sophistication of AAS use by athletes in the late 1960s was characterized by a host of different “stacking routines” (i.e., the consumption of two or more drugs in an attempt to improve the response) using various oral and injectable AAS preparations ( 5 ). Initially, many physicians did not believe AAS improved performance, and the International Olympic Committee (IOC) did not include AAS on the banned substance list. The ACSM adopted this position in their first AAS position stand in 1977 but later corrected in the 1987 publication ( 1 ). Although the 1970s marked a time where AAS use was known mostly among competitive athletes, the 1980s marked a time where AAS use spread well beyond athletics to gyms, health clubs, and public awareness of AAS use increased. The Anti-Drug Abuse Act (1988), Anabolic Steroid Control Act (1990, 2004), and Dietary Supplement Health and Education Act (1994) were enacted, in part, to stem the growing use of AAS. Only a few studies (~17) on AAS use and strength/hypertrophy increases were conducted before the 1980s, and these cumulatively showed minimal effects in untrained men, but significant responses in trained men, despite doses less than that used by many athletes ( 6,7,10 ). The sophisticated protocols and array of drugs used recreationally and by athletes remained a “black box” from a research perspective.

Of current concern is the ease by which AAS users may obtain AAS via the Internet and the proliferation of men’s health clinics. In addition to the use of AAS by competitive athletes, a growing segment of AAS users are nonathletes. Management of men with damaged hypothalamic-pituitary-gonadal regulatory pathways became a new area of medicine resulting in indiscriminate AAS use ( 18 ). Interest in AAS persists as research identifies new information regarding the performance and health aspects of the drugs and through stories of purported use in the sports world. The World Anti-Doping Agency (WADA) has developed new antidoping measures, including blood sampling, guidelines for international information gathering and sharing and revamping their “Athlete Biological Passport” guidelines. While AAS use in sports continues, increases in AAS use in the general population appear to have outpaced athletic use in the last decade ( 19 ).

EPIDEMIOLOGY OF AAS USE

Peer-reviewed studies examining the frequency of illicit AAS use have declined in the past decade despite concern over the growing AAS epidemic in the United States. These studies often rely on self-reports and are fraught with sampling bias, small sample sizes, possible confusion regarding supplement and AAS use, and suboptimal ascertainment ( 5 ). Many AAS users are secretive, with one survey finding that 56% of respondents would not disclose their physicians’ use ( 20 ). Athletes may be unwilling to discuss their use with researchers even when anonymity and confidentially are guaranteed for fear it may jeopardize their career; thus, leading to differences in what athletes reported on surveys versus their actual activities ( 21 ).

In 2014, the National Institute on Drug Abuse estimated that 1.3 million Americans were AAS users, while the Endocrine Society estimated between 2.9 and 4.0 million Americans have used AAS at some point in their lives ( 18,22,23 ). Other reports showed that the number of users might be as high as 4 million men in the United States, with ~100,000 new AAS users annually ( 6,23,24 ). The age of onset of use begins later than most drugs, with only 6% of users starting before 18 ( 23 ).

Although the general public and medical communities attribute AAS use primarily to competitive athletes ( 6 ), research does not support this misperception. Muscle dysmorphia (“megarexia”) is a dominant risk factor for illicit AAS use and indicates that AAS use is often used in pursuit of a more muscular appearance rather than for enhanced athletic performance ( 25 ). Recreationally active individuals age 15 to 24 yr are more likely to use AAS than athletes participating in organized sport ( 26 ). However, reports on the prevalence of illicit AAS use in athlete and nonathlete populations are widely variable. Anabolic-androgenic steroids have been reported in 9% to 67% of elite athletes, while reports of AAS use among gym attendees ranged from 3.5% to 80% ( 27 ). In all areas, men report higher prevalence than women, although the prevalence in women is increasing ( 28 ). Studies in girls have shown prevalence rates between 0.4% and 1.0% in adolescents, ~1.2% in collegiate athletes, and ~10.3% in elite athletes ( 27 ). Others have reported AAS use in young athletes ranging between 0.6% and 6.6% in teenage boys, 0.0% to 3.3% in teenage girls, and between 0.8% and 9.1% for collegiate male athletes ( 29–32 ). Peer-reviewed studies report the highest prevalence of use in weightlifters, powerlifters and bodybuilders, with rates ranging from 33.3% to 79.5% ( 31,33 ).

Several studies have examined sport and activity participation among self-reported AAS users. A survey study of >500 male AAS users (mean age of 29) showed ~70% were recreational exercisers versus 12% competitive bodybuilders, 8% competitive weightlifters, and 9% competitive athletes in other sports ( 34 ). Participation in high school sports was not associated with an increased risk of AAS use ( 34 ). A survey of 12 female AAS users indicated that 33% of the women were recreational users, while 67% participated in competitive bodybuilding and weightlifting. These women used a polypharmacy approach, but their weekly dose was lower than male AAS users ( 35 ). Female users were less likely to stack, more likely to pyramid and less likely to inject AAS than male users ( 35 ).

Rates of AAS use in athletes are sometimes inferred from rates of positive doping tests. However, this data has some inherent limitations, including ongoing updates to banned substances lists, variable drug testing methodologies, and variable lists of targeted substances tested by organizations that do not follow WADA protocols. It has been estimated that drug testing alone may underestimate drug use in elite athletes by 8-fold ( 21 ). The Anti-Doping Administration and Management System maintained by WADA now allows any sports body to share drug testing information. While AAS use in particular divisions, such as men’s vs women’s and underage athletes is still difficult to obtain, the testing databases now include much larger numbers of athletes than in the past. Anabolic agents constitute 87% of atypical findings reported by WADA and 46% of all adverse analytical findings (International Amateur Athletics Federation) ( 36,37 ). Stanazolol and nandrolone have the highest number of AAF at 20% and 14%, respectively, while an “unidentified anabolic agent” (e.g., “designer” AAS) was the third most common at 11% ( 36 ).

The true nature of AAS use and abuse in athletes and recreationally trained individuals is difficult to discern and is often underestimated. In addition to surveys and doping results, other sources of information on AAS use include investigated journalism and government hearings. Unfortunately, all of these methods have significant methodological issues that reduce their estimation accuracy ( 17 ). Journalists have interviewed current and former athletes, coaches, team physicians, and trainers whose estimate of AAS use in sports is much higher than survey reports. There has been an inconsistency between the number of individuals demonstrating signs of AAS use and the statistical prevalence generated via surveys. Drug testing is often limited by circumventing positive tests and has done little to quantify “real-life” use or dissuade AAS use at high levels of competition. Obtaining accurate measures of AAS use in athletes is difficult given the challenges of reducing bias; testing issues, and sincerity needed during interviews and survey completion, for example, fear of accountability, fear of loss of potential income or suspension, or fear of being perceived as a cheater or athlete who needed drugs to be successful.

Attempts have been made to identify the type of individual prone to using AAS ( 38–40 ). Hildebrandt et al. ( 39 ) reported 4 clusters of users from highest to lowest risk, each with different levels of motivation for AAS use: 1) polypharmacy (i.e., use of multiple drugs) approach with high risk (~11%); 2) fat burning (~17%); 3) muscle building (~21%); and 4) low-risk use designed to reduce fat and build muscle (~52%). Others have reported a four-level typology: 1) expert type (exemplifies controlled risk-taking, is knowledgeable about AAS and fascinated with effects on the human body, is scientific and may be focused on muscularity); 2) athlete type (interested in performance enhancement and is competitive); 3) well-being type (interested in looking and feeling good with low risk-taking); and 4) YOLO “You Only Live Once” type (is haphazard using risky behavior, quick improvements, impressing others and peer recognition is important) ( 38,40 ). Despite the typology, athletes’ motivation to use AAS is multi-faceted and influenced by many factors ( Table 4 ).

Several extensive, national studies indicate an overall downward trend in lifetime AAS use among adolescents since peaking in the early 2000s ( 42 ). Monitoring the Future (MTF) is administered annually to a sample of 8th, 10th, and 12th grade students ( 43 ). The MTF reported peak prevalence rates for lifetime AAS use in 2000 to 2002 of 3% to 4% compared with 2018 data in Table 5 (i.e., ~1%–3%). The Youth Risk Behavior Survey (YRBS) is administered annually to a sample of high school students and reports an overall prevalence of 2.9% in 2017 (See Table 6 ), after peaking in 2001 at 5% ( 44 ). Although the YRBS is widely cited, concern has been raised that the term “steroid” is vague and potentially conflated with corticosteroids or steroid-like dietary supplements ( 45 ). Surveys that delineate the type of steroid show usage rates that are markedly lower than those seen in the YRBS data ( 45 ). Although AAS use rates in adolescents are low, ~1 in 8 AAS users initiates their use before age 18 ( 23 ). Several correlates of increased AAS use risk in this group include fitness-related activity ( 46,47 ); weight-related concerns (perceptions of very underweight or overweight status) ( 48,49 ); sexual preference and gender identity ( 25,44 ); and race and ethnicity ( 43,44 ). Some view current AAS use as an epidemic given the emergence of AAS availability through internet/mail order and “backroom” laboratories ( 18,50 ).

METHODS/PATTERNS OF AAS USE

Patterns of AAS use in athletes and resistance-trained populations vary greatly and depend upon: AAS type, self-administration routes, dosages, cycling patterns and durations, and ancillary drugs. A “polypharmacy approach” is commonly used where supraphysiologic doses of injectable and oral AAS are stacked and pyramided progressively in cycles, while ancillary drugs are consumed to minimize side effects, promote other areas of health and fitness, and/or enhance T levels during off-cycles, or periods in between cycles ( Table 7 ). Figure 1 depicts survey results from two studies on usage patterns for >2400 predominately male AAS users ( 34,41 ). These studies indicated that 99.2% of users reported using injectable AAS or a combination of oral and injectable AAS, and >40% used ancillary drugs, such as antiestrogens ( 41 ). Ip et al. ( 34 ) reported that 79% of AAS users “stacked” drugs, 18% used the “pyramid” approach (i.e., where drug intake is progressively increased, plateaus, and then is decreased or tapered until the end of the cycle), and only 9% thought physicians and pharmacists were knowledgeable about AAS. Interestingly, AAS users spent an average of 268 ± 472 h researching AAS prior to use ( 34 ).

ANDROGEN PHYSIOLOGY

Testosterone is the principal androgen and has both androgenic (masculinizing) and anabolic (tissue building) effects. Testosterone is synthesized from cholesterol via the Δ-4 or Δ-5 pathways through the sequential action of several enzymes ( Fig. 2 ). In men, >95% of T is synthesized in the Leydig cells of the testes (with smaller adrenal contributions) under control of the hypothalamic-anterior pituitary-gonadal axis where gonadotropin-releasing hormone stimulates the release of luteinizing hormone (LH). Healthy men produce ~4 to 9 mg of T per day (10–35 nmol·L −1 ) whereas women have approximately 0.5 to 2.3 nmol·L −1 of circulating T in the blood ( 5 ). Gonadotropin-releasing hormone function is under the control of hypothalamic neuropeptides, such as kisspeptins, neurokinin-B, dynorphin-A, and phoenixins ( 51 ). In women, androgens are produced primarily by the ovaries and adrenal glands ( 52 ). Skeletal muscle produces small amounts of androgens ( 53 ). Testosterone circulates in the blood bound to sex hormone-binding globulin (44%–60%), albumin, orosomucoid, and cortisol-binding globulin. Testosterone and other 19-carbon androgens can be converted to 5α-dihydrotestosterone (DHT) by the action of steroid 5α-reductase or converted to estradiol or estrone by the aromatase enzyme. The liver inactivates T, and the resultant metabolites are excreted in the urine.

Androgens perform many ergogenic, anabolic, and anticatabolic functions in skeletal muscle and neuronal tissue, leading to increased muscle strength, power, endurance, and hypertrophy in a dose-dependent manner ( 54 ). A meta-analysis concluded that short-term AAS use increases muscle strength substantially more than placebo and that strength gains and muscle hypertrophy are greater in trained individuals than in nontrained individuals ( 55 ). Gains in body mass and lean body mass (LBM) of ~5% to 20% from AAS use have been reported ( 56 ). Figure 3 depicts some physiological ramifications of androgens that could affect physical performance. However, the findings of controlled clinical trials of T and other AAS may differ from the practical experience of athletes due to the inclusion of mostly untrained subjects in controlled clinical trials; the use of lower doses of T or AAS in clinical trials than those used by many athletes; the use of multiple AAS in stacks with other drugs over long periods; and differences in nutritional patterns, training programs, and study design ( 5,27 ).

Exogenous androgens are often administered orally or parenterally but are also available in cream, nasal spray, buccal, subcutaneous pellets, patches, and gel. Orally administered T is absorbed well but is degraded rapidly. The esterification of the 17-beta-hydroxyl group (e.g., T enanthate, cypionate, decanoate, undecanoate, propionate) makes the androgen more hydrophobic, causing a slow release from the muscle into circulation, increasing the duration of action. When administered intramuscularly, the androgen ester is slowly absorbed into the circulation, where it is then rapidly de-esterified by esterase enzymes to T. Intrinsic potency, bioavailability, and rate of clearance from the circulation are determinants of the biological activity. Other oral and injectable AAS are T, DHT, or 19-nortestosterone derivatives (e.g., methyltestosterone, methandrostenolone, fluoxymesterone, nandrolone decanoate, oxandrolone, trenbolone, stanozolol, and other designer-AAS).

An important and relevant question is how long the effects of a dose of AAS would last in an athlete? That is, how long would potential strength gains or gains in muscle mass persist? The answer to the question is undoubtedly complex and dependent on the AAS being used and their potency (see Fig. 1 ), the history of AAS in the athlete ( 57 ), the athlete’s training age, sex ( 58,59 ), and potentially the developmental stage of the athlete relative to puberty and adulthood (i.e., 18 yr of age). The literature in this area is, unsurprisingly, sparse, but some studies suggest that the effects of AAS persist for weeks after taking the steroids, but at ~12 wk after taking AAS that the effects, at least insofar as strength and muscle mass are concerned, are largely absent ( 55,60 ). For example, Giorgi et al. ( 61 ) showed that testosterone enanthate (TE) (3.5 mg·kg −1 ) administration for 12 wk during training resulted in greater increases in strength, muscle girth, and muscle thickness than a group given a placebo. However, after 12 wk without TE administration, but while still training, there was a reversion of strength and muscle in the TE group to levels no different from the placebo group. In contrast, others have observed preservation of AAS-induced gains in strength and LBM that persist after AAS usage has ceased, at least in the short-term ( 62 ).

Persistent and long-term (at least 5 yr) AAS use in a mixed sample of strength (strongman and powerlifters) and aesthetic sport (bodybuilding) athletes has been reported, in comparison to non-AAS, to result in persistent (i.e., in comparison to a matched group) elevations in LBM, muscle fiber area, capillary density, myonuclei density, and strength that were dose-dependent ( 57 ). The observation that long-term AAS use results in increased myonuclei density ( 57 ) suggesting that a much longer ‘muscle memory’ is perhaps possible in AAS users, particularly those who use AAS early in life. Evidence for such a mechanism comes from preclinical models ( 10 ), where young mice were exposed to AAS and subsequently increased their myonuclear content, resulting in a substantial hypertrophic advantage later in life. The authors of this work ( 63 ) even went so far as to suggest, “… the benefits of even episodic drug [AAS] abuse might be long lasting, if not permanent, in athletes. Our data suggest that the World Anti-Doping Code calling for only 2 yr of ineligibility after… [a doping violation for AAS] use… should be reconsidered.” Support for whether an AAS-induced increase in myonuclear number in humans is lacking; however, if present, then AAS-induced increases in myonuclei are theoretically advantageous to an athlete even if strength and lean mass advantages have been lost.

Residual effects of endogenous testosterone exposure in testosterone-suppressed transgender females are areas of active study and debate. These effects vary greatly depending upon the developmental stage of treatment initiation and will be much less when treatment is initiated before pubertal onset. There is a dichotomy when looking at measures of prepubertal athletic performance. Studies evaluating age-group athletic records report no significant differences in top age-group performances between boys and girls younger than 10 to 12 yr old ( 64–66 ). However, some studies evaluating more specific measures of strength and aerobic capacity reveal an 8% to 10% advantage in prepubertal biologic males relative to females, even after normalizing for body size ( 67,68 ). These performance differences may be residual effects from higher testosterone levels during early infancy (e.g., “mini-puberty”) and/or nonandrogenic genetic factors. Currently, there are no data on the durability of these performance differences in transgender females who start gender-affirming treatment before puberty.

Postpubertal testosterone suppression has variable impacts on performance-related parameters. Within 3 months of starting hormone suppression, hematocrit decreases by 4% to within normal values for cisgender females ( 69 ). A recent systematic review also evaluated evidence to date regarding treatment-related reductions in muscle size, strength, and LBM ( 70 ), summarized in Table 8 . Although the changes documented in Table 8 , along with an increase in fat mass, may contribute to significant reductions in athletic performance, the current lack of data in active or athletic populations makes the magnitude of these changes difficult to assess.

ANDROGEN SIGNALING

Androgen signaling at the tissue level occurs primarily genomically through the classical androgen receptor (AR) with multiple levels of integration with other anabolic/catabolic pathways ( 71 ). Testosterone, DHT, and other AAS bind to cytoplasmic AR ( 72 ). Androgen receptor activity is altered at various sites; phosphorylation may augment androgen/AR transcriptional action (in the presence or absence of androgens) ( 73 ). Androgen receptor signaling is activated primarily by ligand binding, but under some circumstances through ligand-independent mechanisms (e.g., insulin like-growth factor-1 [IGF-1] induced mitogen-activated protein kinase-ERK1/2, p38 and c-Jun N-terminal kinase phosphorylation) ( 74 ) that may sensitize it to anabolic signals in the presence of low androgens ( 75 ). The AR is up-regulated following resistance training and short-term androgen administration ( 54 ).

Upon androgen binding to the ligand-binding domain (LBD) of the AR, the liganded AR undergoes phosphorylation, dimerization, and conformational changes, recruits coregulators, and translocates into the nucleus, where it regulates the transcription of androgen response elements (ARE) of the androgen-responsive genes ( 76 ). Androgen binding activates and stabilizes the AR, which is selectively induced by T, DHT, and AAS ( 77 ). Greater stability is seen with DHT than T ( 78 ). Binding affinity for the AR varies between androgens. Nandrolone and metenolone have a higher binding affinity than T, while stanozolol, methandienone, and fluoxymesterone have a lower binding affinity than T; and oxymetholone has a minimal binding affinity ( 79 ). Androgen binding to the AR completes the pocket that serves as a recruiting surface for co-activators ( 80 ). Some co-activators include BAF57 and 60a, SRC1 and 3, and ARA50 and 74. The activity of these co-regulators and the role of T in ribosome biogenesis may be important in mediating the anabolic effects of AAS on skeletal muscle.

Androgen/AR binding activates signaling through the Wnt-β-catenin pathway. The presence of T (in a dose-dependent manner) increases AR-β-catenin interaction and transcriptional capacity ( 81 ). Androgens promote myogenesis via multiple pathways. Satellite cells and myoblasts express AR and androgen binding, increasing satellite cell activation, proliferation, mobilization, differentiation, and incorporation into skeletal muscle ( 82 ). Androgens increase myogenesis via increased Notch signaling of satellite cells ( 83 ) and increased expression of IGF-1 ( 84 ). Androgen binding to AR on pluripotent mesenchymal cells increases their commitment to myogenesis and inhibits adipogenic differentiation via β-catenin signaling ( 85,86 ). Testosterone upregulates follistatin expression (which blocks signaling through the TGFβ-SMAD 2/3) and increases myogenic differentiation ( 82,84,86–88 ). Androgens may be anticatabolic by decreasing glucocorticoid receptor (GR) expression, interfering with cortisol binding, or the AR-T complex may compete with the cortisol-GR complex for cis -element binding sites on DNA ( 88–91 ).

Nongenomic AR signaling is rapid, with short latency periods acting independently of nuclear receptors ( 92 ). Nongenomic effects are thought to be mediated by direct binding to a target molecule, through intracellular AR activation (i.e., Src kinase), through a transmembrane AR receptor, or via changes in membrane fluidity ( 92 ). Nongenomic signaling involving G-protein 2nd messenger system and may either increase intracellular calcium concentrations via PI3K, phospholipase C, and IP 3 signaling ( 93 ), stimulate the activation of mitogen-activated protein kinase signaling ( 94 ), and mammalian target of rapamycin pathway signaling ( 95 ). Cross-talk between IGF-1 signaling and nongenomic AR signaling appears critical to mediating some anabolic effects ( 96 ). Nongenomic signaling occurs rapidly within seconds to minutes, much faster than classic genomic signaling, which takes hours and requires the constant presence of androgens to maintain intracellular signaling.

SIDE EFFECTS ASSOCIATED WITH ANDROGEN USE AND ABUSE

Investigations examining the safety of androgen use in various populations have been largely inadequate as there is tremendous variability in androgen dosages and patterns of use, including stacking of multiple AAS and concurrent use of accessory drugs ( 5 ). Figure 4 depicts the variety of adverse physiological and psychological effects associated with AAS use. These include relatively rare effects and those that are commonly expected, particularly with long-term AAS abuse ( 30 ).

A survey of 500 AAS users (99% male) who had extensive experience (8 wk to 25 yr with 95% having >1–3 yr of AAS use) with high doses showed that 23% to 64% of respondents reported minor side effects (e.g., testicular atrophy, acne, fluid retention, insomnia, sexual dysfunction, gynecomastia) ( 97 ). Other common effects of AAS use include deleterious changes in cardiovascular (CV) risk factors: decreased plasma high-density lipoprotein (HDL) cholesterol ( 98 ), changes in clotting factors ( 99 ), and mood or psychiatric disturbances ( 79 ). Suppression of the hypothalamic-pituitary-testicular axis and spermatogenesis may result in infertility, while elevations in liver enzymes may reflect liver dysfunction ( 100–102 ). In one study, competitive athletes who used AAS during their competitive careers were more likely to die prematurely than athletes who did not ( 103 ). The use of nonsterile needles and needle sharing practices for intramuscular injections increase the risk for infection, muscle abscess, sepsis, and communicable diseases, such as human immunodeficiency virus (HIV) and hepatitis B and C ( 5 ).

Although CV effects are commonly reported with AAS use, based on an extensive review, the FDA concluded that “... the studies have significant limitations that weaken their evidentiary value for confirming a causal relationship between testosterone and adverse cardiovascular outcomes ” ( 104 ). Part of the difficulty in studying the effects of AAS on CV health is that the impacts of androgens on CV function vary with dose, method of administration, and aromatization potential ( 5 ). Parenteral administration of physiologic T replacement doses are associated with CV function and vary with dose, method of administration, and aromatization potential ( 5 ) with small decreases in plasma HDL, with little or no effect on total cholesterol, low-density lipoprotein (LDL) or triglycerides ( 105–107 ). However, supraphysiologic T doses are associated with significant reductions in HDL ( 108,109 ). Orally administered 17-alpha-alkylated, nonaromatizing AAS produce greater reductions in HDL and increases in LDL than when AAS are administered parenterally ( 110 ). Angell et al. ( 111 ) reported that self-administering AAS (median daily dose = 228 mg) for >2 yr was associated with smaller longitudinal LV strain, right ventricular (RV) ejection fraction, and altered diastolic function compared with nonusers. Others showed impaired RV free wall strain and strain rate associations with AAS abuse in competitive bodybuilders ( 112 ). D’Andrea et al. ( 113 ) showed associations between AAS use (~31 wk; weekly dose = 525 mg) and left atrial impairment (a marker of diastolic burden) in elite bodybuilders compared with nonusers. An increase in left ventricular (LV) mass occurs during resistance training ( 114–116 ); however, potential additional effects from AAS use in humans are unclear. In rats, only high T doses (up to 20 mg per kg body mass) induced cardiac hypertrophy with an impaired contractile process ( 117 ).

Deceased men who had used AAS showed greater cardiac mass than nonusers ( 118 ). Multivariate analysis indicated that increases in heart size were explained by increased body mass and by AAS use. Risk for adverse cardiac events associated with LV mass is supported by case reports detailing sudden death among power athletes who self-administered AAS ( 100,119–122 ). Case reports are largely anecdotal, and a causal relationship between AAS use and risk of sudden death has not been established. Strength/power athletes self-administering AAS have short QT intervals but increased QT dispersion compared with endurance athletes with similar LV mass who have long QT intervals but do not have increased QT dispersion ( 123 ). The interval from the peak to the end of the ECG T wave (Tp-e), Tp-e/QT ratio, and Tp-e/QTc ratio increases in AAS users, suggesting a link between AAS and ventricular arrhythmias, which may increase the risk for sudden death ( 124 ).

Increases in liver enzymes, cholestatic jaundice, hepatic neoplasms, and peliosis hepatis are associated with the use of oral, 17-alpha alkylated AAS ( 102,125,126 ), but not with parenterally administered T or its esters ( 127 ). The association between liver toxicity and AAS use is based on increases in AST and ALT. These enzymes are not liver-specific and are often elevated from muscle damage after resistance exercise ( 101,128 ); thus, possibly overstating the risk of hepatic dysfunction ( 128,129 ).

Endogenous LH and follicle stimulating hormone secretion are suppressed during AAS use, with subsequent effects on testicular T secretion and sperm count ( 130,131 ). Depending on the dose and duration of AAS use, endogenous T, LH, and follicle stimulating hormone may take weeks to months to return to homeostatic levels ( 132 ), and the long-term effects are not well understood. High-dose androgen administration in men is associated with breast tenderness and enlargement, for example, gynecomastia ( 5,133 ), thought to result from peripheral conversion of androgens to estrogens in men administering aromatizable AAS ( 134 ). The prevalence of gynecomastia is unknown, but prevalence rates as high as 54% were reported in AAS users ( 5 ). The use of nonsterile needles and needle-sharing practices for intramuscular injections increases the risk for infection, muscle abscess, sepsis, and communicable diseases, such as HIV and hepatitis B and C ( 5 ).

There is no evidence that T causes prostate cancer, but testosterone replacement therapy (TRT) is associated with a small increase in prostate specific antigen levels in older men with low T, which increases the risk of urological referral for prostate biopsy ( 5 ). Because many older men harbor subclinical prostate cancer, a prostate biopsy may lead to subclinical low-grade prostate cancer detection. Notably, however, TRT increases the risk of prostate biopsy.

The psychological effects of AAS use have garnered much publicity, especially on issues of aggression and suicide. However, the evidence is inconclusive due to the lack of sensitivity of the research instruments used to measure aggressive behavior, large variability in RT programs, preexisting personality or psychiatric disorders, and prevalence of multiple high-risk behaviors and use of other substances, such as alcohol, psychoactive drugs, and dietary supplements ( 5 ). Interestingly, physiologic T replacement in hypogonadal men may improve mood and attenuate negative aspects of mood ( 4 ). Morrison et al. ( 135 ) reported that the aggression and anxiety-provoking influences of androgens in animals are likely a developmental phenomenon and that adult exposure may be anxiolytic over the long term. However, underlying psychological dysfunction may cause a greater susceptibility to AAS use, and high doses of AAS may provoke a “rage” reaction in some individuals with preexisting psychopathology ( 136,137 ). Self-administration of AAS may increase the risk for mood disorders, such as mania, hypomania and depression ( 136,138 ). Resting T concentrations are related to posttraumatic stress (PTSD), in which higher T is associated with a lower risk for PTSD ( 139 ). Further, long-term use of AAS in former weightlifters was associated with poor cognitive function and negative changes in brain morphology ( 140,141 ). Approximately 30% of illicit AAS users will develop AAS dependence, and there is some overlap between AAS dependence and the mechanisms and risk for opioid dependence ( 142,143 ). Sudden discontinuation of exogenous AAS use in those who are dependent or have suppressed endogenous production may result in severe depression and suicidality ( 142,143 ). A multidisciplinary and medically supervised treatment program is indicated for individuals with AAS dependence.

Women self-administering AAS may undergo masculinization and experience hirsutism, deepening of the voice, enlargement of the clitoris, widening of the upper torso, decreased breast size, menstrual irregularities, and male pattern baldness ( 144 ). Some of these adverse effects may not be reversible ( 5 ).

Many of the side effects in adults may be seen in adolescents, but information on use in children is scant. Exogenous AAS exposure in preadolescence triggers pubertal onset and may result in early epiphyseal maturation and closure, leading to loss of ultimate height potential ( 40 ). Although mild acne is common during adolescence ( 40 ), AAS use may result in severe nodular acne, particularly on the back and shoulders, which is often resistant to treatment.

CLINICAL USES OF ANDROGEN THERAPY

Although athletes and recreational trainees have reported obtaining AAS from physicians for illicit purposes ( 26,33,50 ), several clinically approved uses of T exist. Of concern are potential illicit use stemming from a clinical prescription of T given the increased number of antiaging and wellness clinics. The sale of therapeutic T preparations in the United States quadrupled between 2001 and 2011 ( 145 ), and an estimated >2.3 million men received physician-prescribed T therapy as of 2013 ( 146 ). In military treatment facilities, the number of androgen prescriptions increased > twofold (23% per year) from 2007 to 2011, mainly in 35- to-44-yr-old men ( 147 ). Currently, therapeutic T is mostly used to treat primary (i.e., testicular failure) and secondary (i.e., reduced LH) hypogonadism ( 148 ). Androgen therapy has numerous clinical uses outlined in Table 9 ( 145,146 ). A substantial fraction of young men receiving T prescriptions are former AAS users trying to restore endogenous T production ( 149–151 ). The Endocrine Society Clinical Practice Guideline ( 148 ) details decision making regarding androgen therapy and the reader is referred to their specific guidelines on the diagnosis, treatment, and monitoring of hypogonadism in men ( 134 ).

Testosterone replacement therapy has been shown to improve sexual activity ( 152–155 ), vertebral and femoral bone mineral density (BMD) and microarchitecture ( 156,157 ), hemoglobin content ( 158,159 ), LBM, maximal voluntary strength and physical function ( 160–164 ), and reduces body fat and BMI ( 162,165,166 ). There have also been reports of TRT reducing neuroinflammation and depressive symptoms ( 167–169 ), reducing blood pressure and improving lipid profiles ( 166 ), and neuronal regeneration ( 154,156,170–177 ), and may not change or improve cognitive function in older men ( 174,178,179 ). There is a low frequency of adverse events associated with TRT ( 2,148,153,180–190 ). However, all TRT should be accompanied by a structured monitoring plan ( 148 ). The Endocrine Society recommends evaluating symptoms, adverse events, lower urinary tract symptoms, and measurements of T levels, hematocrit, and prostate specific antigen at baseline, 3 to 6 months after starting treatment, and annually thereafter ( 148 ).

Testosterone and free T levels decline with advancing age after peaking in the second and third decades of life ( 191–194 ), leading to increased risk of sexual dysfunction; decreased muscle mass and strength, BMD, mobility; increased falls and fractures, late-life low grade persistent depressive disorder (dysthymia), and CV mortality ( 148,195 ). Low T is associated with an increased risk of diabetes, metabolic syndrome, and increased carotid artery intima-media thickness ( 196,197 ). Whether older men with age-related T decline should receive TRT remains a matter of debate. The Endocrine Society Guideline for TRT of hypogonadal men recommends against routinely prescribing T to all men, 65 yr or older, with low T levels ( 148 ). Decisions regarding TRT should be individualized after discussing potential risks and benefits in men with both symptoms suggestive of consistent T deficiency and burden of symptoms (e.g., low libido, unexplained anemia, osteoporosis) and presence of other co-morbid conditions that increase the risk of T treatment ( 148 ). The shared decision making should weigh the patient’s and clinician’s values. In male children, physiologic doses of T are used for brief periods to initiate pubertal development in those with constitutional delay of growth and puberty. Testosterone is needed permanently for children with congenital or acquired hypogonadism.

Recent interest has focused on the role of T in athletic performance in transgender and sexual developmentally distinct athletes. Individuals transitioning to females may require a therapeutic-use exemption for spironolactone, which is often used to block the androgen receptor and lower overall testosterone levels. Currently, trans female athletes subject to WADA testing must document subthreshold T levels for at least 12 months before being allowed to compete as a female. The IOC sets this threshold at <10 nM, and World Athletics (formerly the International Amateur Athletics Federation) at <5 nM. Interested readers can obtain a much deeper discussion of this topic in several reviews ( 198–200 ).

CONCLUSIONS

Anabolic-androgenic steroids include a wide spectrum of compounds that exert their effects through various mechanisms. Anabolic-androgenic steroid use is advantageous in athletic performance predominantly through enhancements in strength, power, increases in muscle mass, reduced recovery time, and other factors. Major competitive sporting bodies ban the use of AAS; however, the predominant area of AAS usage has now expanded into clinical scenarios, persons undergoing sexual reassignment, and by those interested in AAS for purely aesthetic enhancement. Thus, it is not only athletes who are using AAS to gain performance advantages but also other individuals for various reasons. Use for AAS to enhance athletic performance is banned, and coaches, trainers, and medical staff should monitor for signs of use. The use/abuse of AAS has several notable side effects with various consequences that are, in some cases, reversible. Coaches, parents, trainers, and medical staff need to understand why athletes might use AAS and provide educational programming in a preventive capacity. The position of the ACSM is that the illicit use of AAS for athletic and recreational purposes is, in many cases, illegal, unethical and also poses a substantial health risk. Nonetheless, TRT is used in treating various conditions, and clinicians may elect to use this therapy when medically necessary. The ACSM acknowledges the lawful and ethical use of AAS for clinical purposes and supports the physicians’ ability to provide androgen therapy to patients when deemed medically necessary.

This article is published as an official pronouncement of the American College of Sports Medicine and is an update of the 1987 ACSM position stand on the use of anabolic-androgenic steroids. Click here https://links.lww.com/MSS/C362 to download a slide deck that summarizes this ACSM pronouncement on anabolic-androgenic steroid use. This pronouncement was reviewed for the American College of Sports Medicine by members-at-large and the Pronouncements Committee.

Care has been taken to confirm the accuracy of the information present and to describe generally accepted practices. However, the authors, editors, and publisher are not responsible for errors or omissions or for any consequences from the application of the information in this publication and make no warranty, expressed or implied, with respect to the currency, completeness, or accuracy of the contents of the publication. The application of this information in a particular situation remains the professional responsibility of the practitioner; the clinical treatments described and recommended may not be considered absolute and universal recommendations.

• Cited Here |
• Google Scholar

TESTOSTERONE; HYPERTROPHY; SKELETAL MUSCLE; ANDROGEN; STRENGTH; PERFORMANCE

Supplemental Digital Content

• MSS_2022_10_31_PHILLIPS_20-01062_SDC1.pptx; [PowerPoint] (4.24 MB)
• + Favorites
• View in Gallery

Readers Of this Article Also Read

Nutrition and athletic performance, quantity and quality of exercise for developing and maintaining..., the female athlete triad, exercise and physical activity for older adults, exercise/physical activity in individuals with type 2 diabetes: a consensus....

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 01 June 2022

Anabolic–androgenic steroid use is associated with psychopathy, risk-taking, anger, and physical problems

• Bryan S. Nelson 1 ,
• Tom Hildebrandt 2 &
• Pascal Wallisch 1  

Scientific Reports volume  12 , Article number:  9133 ( 2022 ) Cite this article

• Human behaviour

Previous research has uncovered medical and psychological effects of anabolic–androgenic steroid (AAS) use, but the specific relationship between AAS use and risk-taking behaviors as well as between AAS use and psychopathic tendencies remains understudied. To explore these potential relationships, we anonymously recruited 492 biologically male, self-identified bodybuilders (median age 22; range 18–47 years) from online bodybuilding fora to complete an online survey on Appearance and Performance Enhancing Drug (APED) use, psychological traits, lifestyle choices, and health behaviors. We computed odds ratios and 95% confidence intervals using logistic regression, adjusting for age, race, education, exercise frequency, caloric intake, and lean BMI. Bodybuilders with a prior history of AAS use exhibited heightened odds of psychopathic traits, sexual and substance use risk-taking behaviors, anger problems, and physical problems compared to those with no prior history of AAS use. This study is among the first to directly assess psychopathy within AAS users. Our results on risk-taking, anger problems, and physical problems are consistent with prior AAS research as well as with existing frameworks of AAS use as a risk behavior. Future research should focus on ascertaining causality, specifically whether psychopathy is a risk associated with or a result of AAS use.

Similar content being viewed by others

Adverse effects and potential benefits among selective androgen receptor modulators users: a cross-sectional survey

The effect of GLP-1 receptor agonist use on negative evaluations of women with higher and lower body weight

The effects of acute social ostracism on subsequent snacking behavior and future body mass index in children

Introduction.

An estimated 6% of males globally 1 (including 2.9–4 million Americans 2 ) have used anabolic–androgenic steroids (AAS) such as methyltestosterone, danazol, and oxandrolone, which are a series of synthetic variants of the male sex hormone testosterone that increase lean muscle protein synthesis without increasing fat mass 3 , 4 . Although there are medical uses such as for AIDS-related wasting syndrome 5 , AAS are commonly used by individuals for the purposes of bodybuilding and appearance modification 2 , 3 , 6 . In these cases, doses are commonly 10 to 100 times higher than clinical doses and are typically “cycled” intermittently (i.e., used for a few months, stopped to minimize the stress that AAS impart on the body, then resumed shortly thereafter) 3 , 7 . AAS have a 30% dependence rate among long-term users, higher than many other prescription or illicit drugs such as cocaine and have been linked to medical issues such as liver and kidney damage, cardiovascular problems, testicular atrophy, infertility, hair loss, and gynecomastia 2 , 3 , 7 , 8 , 9 , 10 . AAS use is strongly associated with other substance abuse 8 , 9 , 11 , 12 , and users often exhibit negative, although idiosyncratic, psychological issues 8 , 13 , 14 , 15 , 16 , 17 . Some users report delusions of grandeur and invincibility, while others experience depression and various mood disturbances 8 , 18 , 19 , 20 . As dosage increases, AAS users may become impulsive, moody, aggressive, or even violent 9 , 18 , 19 , 21 , 22 , 23 , 24 , 25 , 26 , 27 . Recent neurobiological studies have focused on effects of AAS on central nervous system functions such as cognition, anxiety, depression, and aggression 10 , 28 , 29 . In recent imaging studies, AAS use was associated with cortex thinning as well as decreased gray matter and increased right amygdala volume 30 , 31 , 32 . AAS use seems to accelerate brain aging through oxidative stress and apoptosis 33 , 34 , 35 , is associated with lower cognitive function 36 , 37 , and may disrupt normal neuronal function in the forebrain, which can increase anxiety and aggressiveness and diminish inhibitory control 10 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 . Increased depression has been frequently observed during AAS withdrawal 32 , 46 .

One area that remains understudied among AAS users is psychopathy, a personality disorder characterized by shallow emotional affect, lack of empathy, and antisocial behavior 47 , 48 , 49 . Psychopathy research has frequently associated psychopathy with violence, repeated imprisonment, disrespect for authority, and substance misuse/abuse 48 , 49 , 50 , 51 , 52 , 53 , 54 , 55 . There is growing evidence that AAS use may be associated with psychopathy, including a direct association between AAS and psychopathy in an Iranian sample 56 as well as numerous reports of associations between AAS use and violent crime or “roid rage” 19 , 21 , 22 , 23 , 25 , 27 , 57 . Prior studies examining AAS use and elements of the “Dark Triad” and “Big Five” personality traits suggest that the relationship between AAS use and both violence and risk-behaviors may be due to self-regulatory deficits and low conscientiousness, and that AAS use is predicted by narcissism, low agreeableness, neuroticism, impulsivity, and inability to delay gratification 56 , 58 . Hauger et al. 28 recently identified significantly lower emotion recognition in AAS dependent users compared to AAS non-using weightlifters, suggesting that this lower emotion recognition may contribute to the higher frequencies of antisocial traits that AAS users have previously reported 59 , 60 . Antisocial personality disorder, which is characterized by the disregard for laws and norms, irritability, and the failure to regard the safety of self and others 61 has been suggested as the mechanism that underlies the link between AAS use and aggression 3 , 9 , 60 , 62 , 63 . Conceptually, there are overlaps between antisocial personality disorder and psychopathy 64 . We therefore argue that psychopathic traits among AAS users are worth exploring.

Thus, the present study assessed whether AAS users were more likely than nonusers to exhibit psychopathic traits, risk-taking behaviors such as sharing needles, anger problems such as getting into altercations, emotional problems such as panic attacks and depression, cognitive problems such as difficulty remembering, and physical problems such as hair loss. We hypothesized that AAS users would display heightened odds of psychopathic traits, substance use risk-taking behaviors, sexual risk-taking behaviors, anger problems, emotional stability problems, cognitive problems, depressive symptoms, anxiety symptoms, impulsivity symptoms, and physical problems, although we recognize that many of these traits are highly idiosyncratic in nature. Finally, we hypothesized there is a dose-dependent relationship between these traits and the variety of substances used as well as the number of cycles.

Participants and procedure

This study was approved by the NYU Committee on Activities Involving Human Subjects and we conducted in accordance with the Declaration of Helsinki principles. We anonymously recruited a large online sample of 492 (Mean age = 22.9, SD age = 4.3) adult biologically male bodybuilders and asked them questions about their Appearance and Performance Enhancing Drug (APED) use (if any), exercise and dietary habits, psychological states, risk-taking behaviors, and any physical problems they might have experienced. The anonymous internet survey was posted to online fitness fora in fall 2015. All participants provided informed consent prior to their participation. Participants had the option to enter an online raffle for one of twenty $50 Amazon gift cards, which were distributed via email.

The following subsections are presented in the same order as the online survey.

Diet and exercise

Participants reported how often they had exercised in the past month (every day, most days, some days, very rarely/never) and rated their caloric intake in the past month on a 5-point ordinal scale (1 = extreme restriction of calories, 5 = extreme over-consumption of calories). We measured caloric intake in terms of restriction, maintenance, or surplus rather than total calories per day because participants likely vary in caloric requirements (i.e., 3000 cal/day may be a surplus for some but a deficit for others).

Appearance and performance enhancing drugs

Each participant indicated whether he had ever used oral, injectable, or topical AAS (“yes, currently,” “yes, formerly,” “no, but considered taking,” “no, never considered taking” for each). Additionally, participants reported how many AAS cycles they had completed and responded whether they had ever used the following APEDs (each with “yes”/”no” options): Testosterone, Dianabol (Methandrostenolone), Deca Durabolin (Nandrolone Decanoate), Winstrol (Stanozolol), Anadrol (Oxymetholone), Human Growth Hormone (Somatropin), Synthol, Anti-Estrogens, Fat Burners (Insulin, Clenbuterol, Cytomel, Cynomel), Trenbolone, or Anavar.

Self-reported events

Participants rated each of the following items as “yes, currently,” “yes, formerly,” or “no, never”.

General events Participants self-reported whether they experienced the following events: depression, increased number of mood swings, getting into altercations, panic attacks, irritability, lack of frustration tolerance, aggression, difficulty focusing, racing thoughts, difficulty making decisions, difficulty remembering, suicidal thoughts, acne, trouble sleeping, water retention, hair loss, changes in appetite, and heart problems.

Risk-taking behavior Participants indicated whether they had engaged in or experienced the following: unprotected sex, sex with multiple partners, sexually transmitted disease or infection (STD), sharing needles, reusing needles, using stimulants without prescription (such as crack, powdered cocaine, methamphetamine, amphetamine, or ecstasy [MDMA]), using opiates without prescription (such as heroin, morphine, codeine, or Oxycontin), using hallucinogens without prescription (such as LSD, mescaline, and psilocybin), using depressants without prescription (such as Valium, Xanax, Librium, and barbiturates), drinking alcohol, smoking tobacco, and smoking marijuana.

Impulsivity

We used the Barratt Impulsiveness Scale to quantify impulsivity (BIS-11) 65 . Participants responded to 30 statements such as “I often have extraneous thoughts” using a 4-point ordinal rating scale (1 = rarely/never, 4 = almost always/always). The BIS-11 displayed strong reliability in this sample (Cronbach’s α = 0.84).

Psychopathic traits

We employed the Levenson Self-Report Psychopathy Scale (LSRP) to assess psychopathy 66 . The scale has 26 items graded on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree) and was strongly reliable in this sample (Cronbach’s α = 0.88).

We assessed anxiety with the Generalized Anxiety Disorder 7-item Scale (GAD-7) 67 . Participants responded to each of the seven items such as “being so restless it is hard to sit still” on a 4-point ordinal rating scale (0 = not at all, 3 = nearly every day). The GAD-7 displayed excellent internal consistency (Cronbach’s α = 0.89). Possible scores range from 0 to 21.

We included the 10-item Center for Epidemiologic Studies Short Depression Scale (CES-D 10) 68 to measure depression. Participants rated statements such as “I felt lonely” on a 4-point ordinal rating scale (0 = rarely or none of the time, 3 = all the time). The CES-D 10 was highly reliable (Cronbach’s α = 0.82), with possible scores ranging from 0 to 30.

Aggravation

Participants responded to the 7-item aggravation subscale of the State Hostility Scale 69 , 70 . In the subscale, participants rate possible descriptions of their current mood (e.g., “stormy” or “vexed”) on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree). The aggravation subscale of the State Hostility Scale had strong reliability (Cronbach’s α = 0.90).

Demographic questions

Lastly, participants reported their age (years), height (inches), weight (pounds), body fat percentage, racial background, and level of education.

Statistical analysis

The survey was convenience sampled, with no pre-specified sample size or power calculation. For our primary analysis, we grouped participants who responded “yes, currently” or “yes, formerly” to having used AAS (oral, injectable, or topical) as AAS users (n = 154, 31.3%). We considered those who responded “no, but considered taking” or “no, never considered taking” to be AAS nonusers (n = 338, 68.7%). We also conducted a secondary analysis using all four categories (current AAS users (n = 121, 24.6%); former AAS users (n = 33, 6.7%); AAS nonuser, considered using (n = 200, 40.7%); AAS nonuser, never considered using (n = 138, 28.0%)).

Both AAS cycle experience and APED variety were self-reported. APED variety was the number of different APED types used (the number each participant responded “yes” to taking of Testosterone, Dianabol (Methandrostenolone), Deca Durabolin (Nandrolone Decanoate), Winstrol (Stanozolol), Anadrol (Oxymetholone), Human Growth Hormone (Somatropin), Synthol, Anti-Estrogens, Fat Burners (Insulin, Clenbuterol, Cytomel, Cynomel), Trenbolone, and Anavar). AAS cycle experience was the number of AAS cycles participants reported. If the participant was an AAS nonuser, then both APED variety and AAS cycle experience were scored as 0.

We grouped traits of interest into the following categories: psychopathic traits, substance use risk-taking behavior, sexual risk-taking behavior, anger problems, emotional stability problems, cognitive problems, depressive symptoms, anxiety symptoms, impulsivity symptoms, and physical problems. Following Brinkley et al. 71 , we considered participants in the top third of the LSRP distribution to have psychopathic traits. We considered any participant that reported sharing needles, reusing needles, hallucinogen use, stimulant use, depressant use, or opiate use as engaging in substance use risk-taking. Similarly, any participant that reported an STD, engaging in unprotected sex, or having multiple sexual partners was categorized as having sexual risk-taking behavior. Any participant scoring in the top half of the aggravation subscale of the State Hostility Scale, reporting physical altercations, or reporting increased aggression was categorized as having anger problems. Participants who reported mood swings, lower frustration tolerance, or irritability were considered to have emotional stability problems while participants with difficulty remembering, difficulty focusing, or trouble making decisions were considered to have cognitive problems. We considered participants with depressive symptoms as those that reported suicidal thoughts, reported increased depression, or had a CES-D 10 score greater than 10 (the established cut point 68 ). Those with anxiety symptoms either had a GAD-7 score greater than the established cut point 67 of 8 or reported panic attacks. A participant who reported racing thoughts or who scored in the top half of the Barratt Impulsiveness Scale was considered to have impulsivity symptoms. Finally, we considered participants to have physical problems if they reported heart problems, appetite changes, water retention, acne, or hair loss.

We used logistic regression to assess possible associations between these traits of interest and AAS use, number of AAS cycles, and variety of APEDs used. We computed odds ratios (OR) with 95% confidence intervals (CI). All analyses adjusted for age, race, education, exercise frequency, caloric intake, and lean BMI. Age, race, and education were included as basic demographic variables, while exercise frequency, caloric intake, and lean BMI were included to account for differences in bodybuilding goals, success, and dedication. We chose to calculate lean BMI to assess how muscular participants were. We used the standard (kg/m 2 ) BMI formula but used each participant’s lean bodyweight instead of his total bodyweight. Lean body weight was calculated by using each participant’s self-reported body fat percentage to determine how much he weighed excluding his body fat (weight in kg * (100%-bodyfat%)). Given that both psychopathy and AAS use are associated with illicit drug use 21 , we conducted a post hoc subgroup analysis among participants without history of polysubstance use (3 or more different drug classes) to ensure any association between AAS use and psychopathic traits was not confounded by polysubstance use. All analyses were conducted in R (version 3.5.1).

Ethics approval

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of New York University.

Consent to participate

Participants provided informed consent prior to their participation in this anonymous internet survey.

Participant characteristics are listed in Table 1 . Most participants were younger than 25 years old (56.5% of AAS users; 79.0% of AAS nonusers), white (85.7% of AAS users; 77.5% of AAS nonusers), and had education beyond high school (75.3% of AAS users; 59.1% of AAS nonusers). The majority in each group exercised most days of the week (79.2% of AAS users; 74.8% of AAS nonusers) and were attempting to gain weight (51.3% of AAS users; 51.2% of AAS nonusers). For AAS users and nonusers, the median (Q1-Q3) lean BMI was 23.6 (22.3–25.4) and 21.6 (20.3–23.3) kg/m 2 . AAS users began use at a median (Q1-Q3) of 21 (20–24) years, had completed 2 (1–3) AAS cycles, and used 4 (2–5) different APED types; 78.6% (121/154) were current AAS users. Among AAS nonusers, 59.2% (200/338) had considered using AAS.

Tables 2 and 3 summarize traits of interest and specific substance use risk-taking behaviors by AAS use status; 25.8% (39/154) of AAS users and 10.2% (34/338) of AAS nonusers had a history of polysubstance use. AAS users had over twice the odds of exhibiting psychopathic traits (OR = 2.50, 95% CI 1.52–4.15), over three times the odds of engaging in substance use risk-taking behaviors (OR = 3.10, 95% CI 1.97–4.93), nearly twice the odds of engaging in sexual risk-taking behaviors (OR = 1.79, 95% CI 1.01–3.26), nearly twice the odds of experiencing anger problems (OR = 1.71, 95% CI 1.02–2.95), and over twice the odds of exhibiting physical problems (OR = 2.23, 95% CI 1.16–4.51) compared to AAS nonusers (Table 4 ). In a post hoc subgroup analysis, AAS users without history of polysubstance use had higher odds of psychopathic traits compared to nonusers without history of polysubstance use (OR = 2.73, 95% CI 1.54–4.90).

In secondary analyses with four levels of AAS use, AAS nonusers who considered using had higher odds of psychopathic traits (OR = 2.19, 95% CI 1.27–3.87), substance use risk-taking (OR = 3.51, 95% CI 2.06–6.14), sexual risk-taking (OR = 3.38, 95% CI 2.00–5.78), anger problems (OR = 3.16, 95% CI 1.86–5.42), emotional stability problems (OR = 1.87, 95% CI 1.16–3.01), depressive symptoms (OR = 2.12, 95% CI 1.32–3.44), and impulsivity symptoms (OR = 2.17, 95% CI 1.31–3.61) compared to AAS nonusers who never considered using; former AAS users had lower odds of both anxiety symptoms (OR = 0.30, 95% CI 0.08–0.84) and impulsivity symptoms (OR = 0.33, 95% CI 0.14–0.74) compared to AAS nonusers who considered using; and current AAS users had higher odds of both impulsivity symptoms (OR = 2.92, 95% CI 1.27–6.84) and physical problems (OR = 5.86, 95% CI 1.83–19.74) compared to former AAS users.

Lastly, we assessed possible relationships between (i) the number of different APED types used and (ii) the number of AAS cycles with the same traits of interest as before. Each additional type of APED used was associated with a 19% increase in the odds of psychopathic traits (OR = 1.19, 95% CI 1.07–1.33), a 24% increase in the odds of substance use risk-taking (OR = 1.24, 95% CI 1.12–1.38), an 18% increase in the odds of sexual risk-taking (OR = 1.18, 95% CI 1.02–1.38), a 15% increase in the odds of emotional stability problems (OR = 1.15, 95% CI 1.04–1.27), and a 33% increase in the odds of physical problems (OR = 1.33, 95% CI 1.12–1.66). For every one-unit increase in the number of AAS cycles, there was a 26% increase in the odds of substance use risk-taking (OR = 1.26, 95% CI 1.10–1.46) and an 85% increase in the odds of physical problems (OR = 1.85, 95% CI 1.29–3.01).

In our online survey of adult biologically male bodybuilders, we found AAS use was associated with higher odds of psychopathic traits, both for AAS users compared to nonusers as well as for increased APED variety. Importantly, this association was also present among participants with no history of polysubstance use. It is not certain whether AAS use predicts psychopathic traits or if the existence of psychopathic traits may actually be a risk factor for AAS use. We note that AAS nonusers who considered AAS use had over twice the odds of psychopathic traits compared to AAS nonusers who never considered AAS use. A recent study of 285 competitive athletes reported that Machiavellianism and psychopathy explained 29% of the variance in positive attitude toward AAS 72 . This is supported generally by the well-established association between psychopathic traits and risk-taking behaviors such as substance abuse 48 . In that case, a large proportion of bodybuilders willing to make the jump to using AAS may already have pre-existing psychopathic traits. Psychopathy is related to both antisocial personality disorder and conduct disorder, each of which is associated with AAS use 9 , 60 . Conduct disorder in particular is a major risk factor for AAS use 9 that cannot be entirely explained by use of other drugs 59 . The relationship may be dynamic; bodybuilders with psychopathic tendencies may be more willing to begin AAS in the first place. Subsequently, these traits might be amplified either chemically by AAS use or psychologically by the environment; prior work has shown the difference between psychopaths and non-psychopaths in emotional-regulatory activity in the aPFC is modified by endogenous testosterone level 73 . With this in mind, longitudinal research is needed to further explore the causal nature of this relationship.

Our study is one of many to link AAS use substance use risk-taking behaviors 74 , 75 and sexual risk-taking behaviors 59 , 76 . It is difficult to ascertain the specific relationship between AAS use and risk-taking. Unlike physical, psychological, cognitive, and anger problems, which have all had experimental and translational research done to strengthen causal interpretations of such links 16 , 77 , there has not been experimental work to test whether risk-taking behaviors are caused by AAS use. In fact, it is important to consider that AAS use is itself a risk behavior, and another form of substance use, so AAS users may already engage in many other risk-taking behaviors prior to their first use. This may be especially true in light of our findings that AAS nonusers who considered AAS use had over three-times the odds of both substance use and sexual risk-taking behaviors compared to AAS nonusers who never considered AAS use, as well as our results regarding APED variety and AAS cycle experience. AAS users willing to try more types of APEDs or willing to undergo more AAS cycles may be more likely to also engage in risk-taking behaviors. Perhaps the relationship between AAS and risk-taking behaviors is bidirectional and interactive, where athletes that engage in these risk behaviors such as illicit drug use experiment with AAS, which may lower their inhibitions to take further risks.

Our finding that AAS users have higher odds of experiencing anger problems is in line with prior research 16 , 19 , 20 . Notably, anger has been previously reported as both a potential risk factor 78 as well as a potential outcome 27 . We did not observe associations between AAS use and emotional stability problems, cognitive problems, depressive symptoms, anxiety symptoms, or impulsivity symptoms. Prior research has identified various psychological and cognitive traits among AAS users such as depression, impulsivity, and mania 18 , 19 , 20 , but they are generally idiosyncratic in nature 8 , 79 , 80 , 81 . We do note that AAS nonusers who considered AAS use had higher odds of emotional stability problems, depressive symptoms, and impulsivity symptoms compared to AAS nonusers who never considered AAS use, former AAS users had lower odds of anxiety symptoms and impulsivity symptoms compared to AAS nonusers who considered AAS use, and current AAS users had higher odds of impulsivity symptoms compared to former AAS users. These findings comparing AAS nonusers who considered vs. never considered AAS use are consistent with prior research about factors relating to the decision to use AAS, including research on the “Big Five” personality traits 58 . Additionally, we observed increased odds of emotional stability problems with increased APED variety. Lastly, our hypothesis about physical problems was supported for AAS users compared to nonusers as well as the dose dependent response in relation to increased APED variety and increased AAS cycle experience. These findings are consistent with prior studies 3 , 8 , 10 , 32 .

There are several limitations. Although we successfully elicited responses from real-world users of AAS, there remain questions about how representative our sample is. AAS users in our sample were relatively new users (median of 2 prior cycles). Our findings may have been different with a group of more experienced users. It is also possible that our online survey was more likely to attract individuals with psychopathic traits or that AAS users with psychopathic traits are more willing to take an online survey than other users. We note that > 50% of AAS users and nonusers were considered to have substance use risk-taking, sexual risk-taking, anger problems, emotional stability problems, cognitive problems, depressive symptoms, impulsivity symptoms, and physical problems. Lastly, this cross-sectional study is entirely correlational and any attempts to speculate about causality should be made with extreme caution. Further prospective or experimental studies are needed. In light of the findings on Machiavellianism and psychopathy in relation to willingness to use AAS 72 , it would be interesting to also examine the link to narcissism and self-esteem/insecurity 82 . We wonder whether self-esteem or narcissistic traits could play an additional role in the motivation to begin AAS use, given the known downsides.

This study is among the first to directly assess psychopathy within AAS users. Our results on risk-taking, anger problems, and physical problems are consistent with prior AAS research as well as with existing frameworks of AAS use as a risk behavior. Increased psychopathic traits in AAS users may serve as the underlying mechanism to predict increased anger problems (see 60 regarding antisocial personality disorder as a mechanism between AAS and aggression). Although the present study highlights the relationship between AAS use and psychopathic traits, future research should emphasize possible causal explanations and try to elucidate the directionality of this relationship. Additionally, the mechanisms between AAS use and risk and violent behaviors should be further explored.

Data availability

All data generated or analyzed during this study are included in this published article’s supplementary information files. R code used in data analysis can be made available upon reasonable request to the corresponding author.

Sagoe, D., Molde, H., Andreassen, C. S., Torsheim, T. & Pallesen, S. The global epidemiology of anabolic-androgenic steroid use: A meta-analysis and meta-regression analysis. Ann. Epidemiol. 24 , 383–398 (2014).

Article   PubMed   Google Scholar  

Pope, H. G. Jr. et al. The lifetime prevalence of anabolic-androgenic steroid use and dependence in Americans: Current best estimates. Am. J. Addict. 23 , 371–377 (2014).

Kanayama, G., Brower, K. J., Wood, R. I., Hudson, J. I. & Pope, H. G. Jr. Anabolic-androgenic steroid dependence: An emerging disorder. Addiction 104 , 1966–1978 (2009).

Article   PubMed   PubMed Central   Google Scholar  

Kanayama, G., Hudson, J. I. & Pope, H. G. Features of men with anabolic-androgenic steroid dependence: A comparison with nondependent AAS users and with AAS nonusers. Drug Alcohol Depend. 102 , 130–137 (2009).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Woerdeman, J. & de Ronde, W. Therapeutic effects of anabolic androgenic steroids on chronic diseases associated with muscle wasting. Expert Opin. Investig. Drugs 20 , 87–97 (2011).

Article   CAS   PubMed   Google Scholar  

Buckley, W. E. et al. Estimated prevalence of anabolic steroid use among male high school seniors. J. Am. Med. Assoc. 260 , 3441–3445 (1988).

Article   CAS   Google Scholar  

National Institute on Drug Abuse. Anabolic Steroids DrugFacts. National Institute on Drug Abuse https://www.drugabuse.gov/publications/drugfacts/anabolic-steroids (2018).

Pope, H. G. et al. Adverse health consequences of performance-enhancing drugs: An endocrine society scientific statement. Endocr. Rev. 35 , 341–375 (2014).

Kanayama, G., Pope, H. G. & Hudson, J. I. Associations of anabolic-androgenic steroid use with other behavioral disorders: an analysis using directed acyclic graphs. Psychol. Med. 48 , 2601–2608 (2018).

Albano, G. D. et al. Adverse effects of anabolic-androgenic steroids: A literature review. Healthc. Basel 9 , 97 (2021).

Article   Google Scholar  

DuRant, R. H., Rickert, V. I., Ashworth, C. S., Newman, C. & Slavens, G. Use of multiple drugs among adolescents who use anabolic steroids. N. Engl. J. Med. 328 , 922–926 (1993).

Melia, P., Pipe, A. & Greenberg, L. The use of anabolic-androgenic steroids by Canadian students. Clin. J. Sport Med. 6 , 9–14 (1996).

Hartgens, F. & Kuipers, H. Effects of androgenic-anabolic steroids in athletes. Sports Med. 34 , 513–554 (2004).

Lood, Y., Eklund, A., Garle, M. & Ahlner, J. Anabolic androgenic steroids in police cases in Sweden 1999–2009. Forensic Sci. Int. 219 , 199–204 (2012).

Pope Jr, H. G. & Kanayama, G. Anabolic-androgenic steroids. In Drug Abuse and Addiction in Medical Illness: Causes, Consequences and Treatment (Springer, 2012). https://doi.org/10.1007/978-1-4614-3375-0 .

Su, T.-P. et al. Neuropsychiatric effects of anabolic steroids in male normal volunteers. J. Am. Med. Assoc. 269 , 2760–2764 (1993).

Scarth, M. et al. Severity of anabolic steroid dependence, executive function, and personality traits in substance use disorder patients in Norway. Drug Alcohol Depend. 231 , 109275–109275 (2022).

Corrigan, B. Anabolic steroids and the mind. Med. J. Aust. 165 , 222–226 (1996).

Pope, H. G. Jr. & Katz, D. L. Homicide and near-homicide by abanolic steroid users. J. Clin. Psychiatry 51 , 28–31 (1990).

PubMed   Google Scholar  

Pope, H. G. & Katz, D. L. Psychiatric and medical effects of anabolic-androgenic steroid use: A controlled study of 160 athletes. Arch. Gen. Psychiatry 51 , 375–382 (1994).

Lundholm, L., Frisell, T., Lichtenstein, P. & Langstrom, N. Anabolic androgenic steroids and violent offending: Confounding by polysubstance abuse among 10365 general population men. Addiction 110 , 100–108 (2015).

Skarberg, K., Nyberg, F. & Engstrom, I. Is there an association between the use of anabolic-androgenic steroids and criminality?. Eur. Addict. Res. 16 , 213–219 (2010).

Klötz, F., Garle, M., Granath, F. & Thiblin, I. Criminality among individuals testing positive for the presence of anabolic androgenic steroids. Arch. Gen. Psychiatry 63 , 1274–1279 (2006).

Perry, P. et al. Measures of aggression and mood changes in male weightlifters with and without androgenic anabolic steroid use. J. Forensic Sci. 48 , 646–651 (2003).

Thiblin, I. & Parlklo, T. Anabolic androgenic steroids and violence. Acta Psychiatr. Scand. 106 , 125–128 (2002).

Williamson, D. J. & Young, A. H. Psychiatric effects of androgenic and anabolic-androgenic steroid abuse in men: A brief review of the literature. J. Psychopharmacol. 6 , 20–26 (1992).

Thiblin, I., Kristiansson, M. & Rajs, J. Anabolic androgenic steroids and behavioural patterns among violent offenders. J. Forensic Psychiatry 8 , 299–310 (1997).

Hauger, L. E. et al. Anabolic androgenic steroid dependence is associated with impaired emotion recognition. Psychopharmacology 236 , 2667–2676 (2019).

Almeida, O. P., Yeap, B. B., Hankey, G. J., Jamrozik, K. & Flicker, L. Low free testosterone concentration as a potentially treatable cause of depressive symptoms in older men. Arch. Gen. Psychiatry 65 , 283–289 (2008).

Bjørnebekk, A. et al. Structural brain imaging of long-term anabolic-androgenic steroid users and nonusing weightlifters. Biol. Psychiatry 1969 (82), 294–302 (2016).

Google Scholar  

Kaufman, M. J. et al. Brain and cognition abnormalities in long-term anabolic-androgenic steroid users. Drug Alcohol Depend. 152 , 47–56 (2015).

Grönbladh, A., Nylander, E. & Hallberg, M. The neurobiology and addiction potential of anabolic androgenic steroids and the effects of growth hormone. Brain Res. Bull. 126 , 127–137 (2016).

Article   PubMed   CAS   Google Scholar  

Bjørnebekk, A. et al. Long-term anabolic-androgenic steroid use is associated with deviant brain aging. Biol. Psychiatry Cogn. Neurosci. Neuroimaging 6 , 579–589 (2021).

Kaufman, M. J., Kanayama, G., Hudson, J. I. & Pope, H. G. Supraphysiologic-dose anabolic–androgenic steroid use: A risk factor for dementia?. Neurosci. Biobehav. Rev. 100 , 180–207 (2019).

Piacentino, D. et al. Anabolic-androgenic Steroid use and Psychopathology in Athletes. A systematic review. Curr. Neuropharmacol. 13 , 101–121 (2015).

Kanayama, G., Kean, J., Hudson, J. I. & Pope, H. G. Cognitive deficits in long-term anabolic-androgenic steroid users. Drug Alcohol Depend. 130 , 208–214 (2013).

Bjørnebekk, A. et al. Cognitive performance and structural brain correlates in long-term anabolic-androgenic steroid exposed and nonexposed weightlifters. Neuropsychology 33 , 547–559 (2019).

Caraci, F. et al. Neurotoxic properties of the anabolic androgenic steroids nandrolone and methandrostenolone in primary neuronal cultures. J. Neurosci. Res. 89 , 592–600 (2011).

Penatti, C. A. A., Porter, D. M. & Henderson, L. P. Chronic exposure to anabolic androgenic steroids alters neuronal function in the mammalian forebrain via androgen receptor- and estrogen receptor-mediated mechanisms. J. Neurosci. 29 , 12484–12496 (2009).

Robinson, S., Penatti, C. A. A. & Clark, A. S. The role of the androgen receptor in anabolic androgenic steroid-induced aggressive behavior in C57BL/6J and Tfm mice. Horm. Behav. 61 , 67–75 (2012).

Bueno, A., Carvalho, F. B., Gutierres, J. A. M., Lhamas, C. & Andrade, C. M. A comparative study of the effect of the dose and exposure duration of anabolic androgenic steroids on behavior, cholinergic regulation, and oxidative stress in rats. PLoS ONE 12 , e0177623–e0177623 (2017).

Article   PubMed   PubMed Central   CAS   Google Scholar  

Henderson, L. P., Penatti, C. A. A., Jones, B. L., Yang, P. & Clark, A. S. Anabolic androgenic steroids and forebrain GABAergic transmission. Neuroact. Steroids Old Play. New Game 138 , 793–799 (2006).

CAS   Google Scholar  

Melloni, R. H. & Ricci, L. A. Adolescent exposure to anabolic/androgenic steroids and the neurobiology of offensive aggression: A hypothalamic neural model based on findings in pubertal Syrian hamsters. Sex Drugs Sex Differ. Horm Eff. Drug Abuse. 58 , 177–191 (2010).

Quaglio, G. et al. Anabolic steroids: Dependence and complications of chronic use. Intern. Emerg. Med. 4 , 289–296 (2009).

Bertozzi, G. et al. The role of anabolic androgenic steroids in disruption of the physiological function in discrete areas of the central nervous system. Mol. Neurobiol. 55 , 5548–5556 (2018).

Brady, K., Levin, F. R., Galanter, M., Kleber, H. D. & American Psychiatric Association Publishing, issuing body. In The American Psychiatric Association Publishing Textbook of Substance Use Disorder Treatment . (American Psychiatric Association Publishing, 2021).

Hare, R. D. Manual for the Revised Psychopathy Checklist (Multi-Health Systems, 1991).

Patrick, C. J. Handbook of Psychopathy (The Guilford Press, 2005).

Blair, J. The Psychopath: Emotion and the Brain (Blackwell, 2005).

Coid, J. et al. Psychopathy among prisoners in England and Wales. Int. J. Law Psychiatry 32 , 134–141 (2009).

Hildebrand, M. & de Ruiter, C. Psychopathic traits and change on indicators of dynamic risk factors during inpatient forensic psychiatric treatment. Int. J. Law Psychiatry 35 , 276–288 (2012).

Kantor, M. The Psychopathy of Everyday Life: How Antisocial Personality Disorder Affects All of Us (Praeger, 2006).

Neumann, C. S. & Hare, R. D. Psychopathic traits in a large community sample: Links to violence, alcohol use, and intelligence. J. Consult. Clin. Psychol. 76 , 893–899 (2008).

Smith, S. S. & Newman, J. P. Alcohol and drug abuse-dependence disorders in psychopathic and nonpsychopathic criminal offenders. J. Abnorm. Psychol. 99 , 430–439 (1990).

Woodworth, M. & Porter, S. In cold blood: Characteristics of criminal homicides as a function of psychopathy. J. Abnorm. Psychol. 111 , 436–445 (2002).

Chegeni, R., Sagoe, D., Bergen, S. P. & Maleki, A. The dark triad and big five personality traits in anabolic steroid users (2019).

Pope, H. G., Kanayama, G., Hudson, J. I. & Kaufman, M. J. Review article: Anabolic-androgenic steroids, violence, and crime: Two cases and literature review. Am. J. Addict. 30 , 423–432 (2021).

Garcia-Argibay, M. The relationship between the big five personality traits, impulsivity, and anabolic steroid use. Subst. Use Misuse 54 , 236–246 (2019).

Hallgren, M. et al. Anti-social behaviors associated with anabolic-androgenic steroid use among male adolescents. Eur. Addict. Res. 21 , 321–326 (2015).

Yates, W. R., Perry, P. J. & Andersen, K. H. Illicit anabolic steroid use: A controlled personality study. Acta Psychiatr. Scand. 81 , 548–550 (1990).

Diagnostic and statistical manual of mental disorders : DSM-5 . (American Psychiatric Association, 2013).

Hauger, L. E., Havnes, I. A., Jørstad, M. L. & Bjørnebekk, A. Anabolic androgenic steroids, antisocial personality traits, aggression and violence. Drug Alcohol Depend. 221 , 108604–108604 (2021).

Borjesson, A. et al. Male anabolic androgenic steroid users with personality disorders report more aggressive feelings, suicidal thoughts, and criminality. Med. Kaunas Lith. 56 , 265 (2020).

Warren, J. I. & South, S. C. Comparing the constructs of antisocial personality disorder and psychopathy in a sample of incarcerated women. Behav. Sci. Law 24 , 1–20 (2006).

Patton, J. H., Stanford, M. S. & Barratt, E. S. Factor structure of the Barratt Impulsiveness Scale. J. Clin. Psychol. 51 , 768–774 (1995).

Levenson, M. R., Kiehl, K. A. & Fitzpatrick, C. M. Assessing psychopathic attributes in a noninstitutionalized population. J. Pers. Soc. Psychol. 68 , 151–158 (1995).

Spitzer, R. L., Kroenke, K., Williams, J. B. W. & Löwe, B. A brief measure for assessing generalized anxiety disorder: The GAD-7. Arch. Intern. Med. 166 , 1092–1097 (2006).

Radloff, L. S. The CES-D Scale: A self-report depression scale for research in the general population. Appl. Psychol. Meas. 1 , 385–401 (1977).

Anderson, C. A., Deuser, W. E. & DeNeve, K. M. Hot temperatures, hostile affect, hostile cognition, and arousal: Tests of a general model of affective aggression. Pers. Soc. Psychol. Bull. 21 , 434–448 (1995).

Anderson, C. A. & Carnagey, N. L. Causal effects of violent sports video games on aggression: Is it competitiveness or violent content?. J. Exp. Soc. Psychol. 45 , 731–739 (2009).

Brinkley, C. A., Schmitt, W. A., Smith, S. S. & Newman, J. P. Construct validation of a self-report psychopathy scale: Does Levenson’s self-report psychopathy scale measure the same constructs as Hare’s psychopathy checklist-revised?. Personal. Individ. Differ. 31 , 1021–1038 (2001).

Nicholls, A. R., Madigan, D. J., Backhouse, S. H. & Levy, A. R. Personality traits and performance enhancing drugs: The Dark Triad and doping attitudes among competitive athletes. Personal. Individ. Differ. 112 , 113–116 (2017).

Volman, I. A. C. et al. Testosterone modulates altered prefrontal control of emotional actions in psychopathic offenders. eNeuro 3 , ENEURO.0107–15.2016 (2016).

Dodge, T. & Hoagland, M. F. The use of anabolic androgenic steroids and polypharmacy: A review of the literature. Drug Alcohol Depend. 114 , 100–109 (2010).

Sagoe, D. et al. Polypharmacy among anabolic-androgenic steroid users: A descriptive metasynthesis. Subst. Abuse Treat. Prev. Policy 10 , 12–12 (2015).

Ganson, K. T., Jackson, D. B., Testa, A., Murnane, P. M. & Nagata, J. M. Performance-enhancing substance use and sexual risk behaviors among U.S. Men: Results from a prospective cohort study. J. Sex Res. 1–7. ahead-of-print.

Benson, E. More male than male. Monit. Psychol. 33 , 49 (2002).

Sagoe, D., Mentzoni, R. A., Hanss, D. & Pallesen, S. Aggression is associated with increased anabolic-androgenic steroid use contemplation among adolescents. Subst. Use Misuse 51 , 1462–1469 (2016).

Kanayama, G., Hudson, J. I. & Pope, H. G. Illicit anabolic–androgenic steroid use. Horm. Behav. 58 , 111–121 (2010).

Chegeni, R., Notelaers, G., Pallesen, S. & Sagoe, D. Aggression and psychological distress in male and female anabolic-androgenic steroid users: A multigroup latent class analysis. Front. Psychiatry 12 , 629428–629428 (2021).

Hildebrandt, T., Langenbucher, J. W., Carr, S. J. & Sanjuan, P. Modeling population heterogeneity in appearance- and performance-enhancing drug (APED) use: Applications of mixture modeling in 400 regular APED users. J. Abnorm. Psychol. 1965 (116), 717–733 (2007).

Kowalchyk, M., Palmieri, H., Conte, E. & Wallisch, P. Narcissism through the lens of performative self-elevation. Personal. Individ. Differ. 177 , 110780 (2021).

Download references

Acknowledgements

We thank Ward Pettibone and Andre Nakkab for administrative assistance. This work was supported by the New York University Dean’s Undergraduate Research Fund.

This work was supported by the New York University Dean’s Undergraduate Research Fund.

Author information

Authors and affiliations.

Department of Psychology, New York University, New York, NY, USA

Bryan S. Nelson & Pascal Wallisch

Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, NY, USA

Tom Hildebrandt

You can also search for this author in PubMed   Google Scholar

Contributions

All authors contributed to the study conception and design. Material preparation was performed by BSN, TH, and PW. Data recording and analysis was performed by BSN. The first draft of the manuscript was written by BSN and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Bryan S. Nelson .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Supplementary information 1., supplementary information 2., rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Nelson, B.S., Hildebrandt, T. & Wallisch, P. Anabolic–androgenic steroid use is associated with psychopathy, risk-taking, anger, and physical problems. Sci Rep 12 , 9133 (2022). https://doi.org/10.1038/s41598-022-13048-w

Download citation

Received : 05 October 2021

Accepted : 25 April 2022

Published : 01 June 2022

DOI : https://doi.org/10.1038/s41598-022-13048-w

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Anabolic–androgenic steroids: How do they work and what are the risks?

• Android Health Clinic
• This person is not on ResearchGate, or hasn't claimed this research yet.

Abstract and Figures

Discover the world's research

• 25+ million members
• 160+ million publication pages
• 2.3+ billion citations

• Anderson Carlos Marçal
• Beyza Suvarikli Alan
• Avni Camgoz
• Mustafa Orhun Dayan
• Nida Karakaya

• Per Wiik Johansen
• Anders Palmstrøm Jørgensen

• Enrico Fulco

• Francesco Giuseppe Foschi

• Dominique Poulin

• Pauline Couacault
• Dennisse Avella
• Sara Londoño‐Osorio
• Michael Witting
• J APPL TOXICOL

• Ingra Tais Malacarne
• Wilton Mitsunari Takeshita
• Daniel Araki Ribeiro

• M. Z. Stevanović
• D. S. Jakimov
• Marija N. Sakač
• SUBST USE MISUSE
• Paulo Marcelo de Andrade Lima
• Ycaro C Barros
• Ana B N Barros
• Letícia M Farias

• Eugene Braunwald
• James A de Lemos
• Robert Byington
• Jane J. Lee
• Clara Fitzgerald

• Diederik L. Smit

• Willem de Ronde
• Atherosclerosis
• Gregory G. Schwartz
• Christie M Ballantyne
• J Am Med Assoc

• Michael J. Barry
• Carol M Mangione

• Mathias Tischler
• Christof Mrstik

• EUR HEART J
• Frank L J Visseren

• Yvo M Smulders

• Madelon M. Buijs
• NEW ENGL J MED
• A. Michael Lincoff
• Stephen J. Nicholls

• Recruit researchers
• Join for free
• Login Email Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google Welcome back! Please log in. Email · Hint Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google No account? Sign up

• Search Menu
• Sign in through your institution
• Advance articles
• Author Guidelines
• Open Access
• About Integrative and Comparative Biology
• About the Society for Integrative and Comparative Biology
• Editorial Board
• Advertising and Corporate Services
• Journals Career Network
• Self-Archiving Policy
• Dispatch Dates
• Journals on Oxford Academic
• Books on Oxford Academic

Article Contents

Introduction, implications of studies of humans for studies of nonhuman animals, conclusions, acknowledgments.

• < Previous

Steroid use and human performance: Lessons for integrative biologists

From the symposium “Hormonal Regulation of Whole-Animal Performance: Implications for Selection” presented at the annual meeting of the Society for Integrative and Comparative Biology, January 3–7, 2009, at Boston, Massachusetts.

2 Present address: Department of Biology, University of South Dakota, Vermillion, SD 57069, USA

• Article contents
• Figures & tables
• Supplementary Data

Jerry F. Husak, Duncan J. Irschick, Steroid use and human performance: Lessons for integrative biologists, Integrative and Comparative Biology , Volume 49, Issue 4, October 2009, Pages 354–364, https://doi.org/10.1093/icb/icp015

• Permissions Icon Permissions

While recent studies have begun to address how hormones mediate whole-animal performance traits, the field conspicuously lags behind research conducted on humans. Recent studies of human steroid use have revealed that steroid use increases muscle cross-sectional area and mass, largely due to increases in protein synthesis, and muscle fiber hypertrophy attributable to an increased number of satellite cells and myonuclei per unit area. These biochemical and cellular effects on skeletal muscle morphology translate into increased power and work during weight-lifting and enhanced performance in burst, sprinting activities. However, there are no unequivocal data that human steroid use enhances endurance performance or muscle fatigability or recovery. The effects of steroids on human morphology and performance are in general consistent with results found for nonhuman animals, though there are notable discrepancies. However, some of the discrepancies may be due to a paucity of comparative data on how testosterone affects muscle physiology and subsequent performance across different regions of the body and across vertebrate taxa. Therefore, we advocate more research on the basic relationships among hormones, morphology, and performance. Based on results from human studies, we recommend that integrative biologists interested in studying hormone regulation of performance should take into account training, timing of administration, and dosage administered when designing experiments or field studies. We also argue that more information is needed on the long-term effects of hormone manipulation on performance and fitness.

One of the most widely discussed and controversial arenas of human performance concerns the use of steroid supplements to enhance athletic ability for a variety of sports, ranging from bicycling to baseball. There is strong evidence that human athletes have attempted to enhance their athletic performance using steroids since the 1950s, but whether, and in which sports, steroids are actually effective remains controversial (reviewed by Ryan 1981 ; George 2003 ; Hartgens and Kuipers 2004 ). In general, steroids used by athletes encompass a wide variety of forms of the androgen testosterone (George 2003 ), and most seem to have the classical androgenic and anabolic effects on men, although steroid use by women cannot be ignored (Malarkey et al. 1991 ; Gruber and Pope 2000 ). Alternative forms of testosterone (e.g., testosterone enanthate, methandrostenolone) are typically used by those desiring enhanced performance because ingested or injected testosterone is quickly metabolized into inactive forms (Wilson 1988 ). Thus, studies of humans that we cite involve testosterone derivatives. Early studies of the effects of steroids on human performance, however, had major flaws in design, such as lack of control groups and a double-blind procedure, the presence of confounding factors (e.g., differences in level of exercise and in motivation), and inappropriate statistical techniques (reviewed by Bhasin et al. 2001 ; George 2003 ). These problems left open for many years the question of whether, and in what capacity, steroids actually enhance athletic performance, until more recent studies conclusively showed significant effects of steroids.

The topic of steroid effects on human athletic performance is germane to an emerging field of research investigating hormonal effects on animals’ performance (e.g., sprint speed, endurance capacity, bite-force capacity) (Husak et al. 2009a ), as testosterone may exert general effects on performance across widely divergent vertebrate taxa. Our goal in this review is to interpret the effects of steroids on human performance in this broader context of hormonal effects across a wider range of taxa. We are particularly interested in drawing lessons and potential avenues of research for animal biologists from published research on humans. We have performed a selective review of studies examining how humans' use of steroids affects skeletal muscle physiology and subsequent athletic performance. While studies of performance on nonhumans have dealt extensively with the effects of morphological traits on performance and the impact of performance on individual fitness (Arnold 1983 ; Garland and Losos 1994 ; Irschick and Garland 2001 ; Irschick et al. 2007 , 2008 ; Husak et al. 2009a ), there has been relatively little synthetic discussion of how hormones affect performance in non-human animals. We also point the reader towards several recent reviews of steroid use and performance by humans for details not discussed in our review (Bhasin et al. 2001 ; George 2003 ; Hartgens and Kuipers 2004 ).

General effects of testosterone on the phenotype of males

The development of primary and secondary sexual characteristics is stimulated by testosterone in vertebrate males, and these effects can be either organizational or activational in nature (Norris 1997 ; Hadley 2000 ). Organizational effects tend to occur early in development, and during a critical window of time, thereby resulting in permanent effects. On the other hand, activational effects occur in adults, and the effects are typically temporary (Arnold and Breedlove 1985 ). The hypothalamus stimulates production of gonadotropin-releasing hormone, which in turn stimulates production of luetenizing hormone in the anterior pituitary. Luetenizing hormone then stimulates production of testosterone in the Leydig cells of the testes. Testosterone then circulates throughout the body where it exerts effects on multiple target tissues that have the appropriate receptors or appropriate enzymes (e.g., aromatase or 5α-reductase) to convert testosterone for binding to other types of receptors (Kicman 2008 ). The widespread effects of circulating levels of testosterone on aggression, secondary sexual traits, and growth of skeletal muscle in males of many vertebrate species are well-documented (Marler and Moore 1988 ; Wingfield et al. 1990 ; Ketterson and Nolan 1999 ; Sinervo et al. 2000 ; Ketterson et al. 2001 ; Oliveira 2004 ; Adkins-Regan 2005 ; Hau 2007 ; contributions in this issue). In particular, production of testosterone by males has been linked with the expression of color and behavioral display signals, as well as aggression (Marler and Moore 1988 ; Kimball and Ligon 1999 ; Hews and Quinn 2003 ; Adkins-Regan 2005 ; Cox et al. 2008 ) and increased growth (Fennell and Scanes 1992 ; Borski et al. 1996 ; Cox and John-Alder 2005 ), although this latter effect may depend on specific selective pressures on males (Cox and John-Alder 2005 ).

Effects of testosterone on the physiology of human skeletal muscle

Testosterone has multiple effects on skeletal muscle at the biochemical and cellular levels, but the direct cause-and-effect relationships among these effects are still unclear (Sinha-Hikim 2002 ; Hartgens and Kuipers 2004 ). The studies that we discuss here, and throughout the paper are from experiments or correlative studies conducted on adult individuals such that the effects seen are activational in nature, causing rather rapid changes to the phenotype. Increased testosterone causes increased protein synthesis by muscle cells (Griggs et al. 1989 ; Kadi et al. 1999 ; Hartgens and Kuipers 2004 ), which is necessary for anabolic effects and an increase in lean muscle mass. Sinha-Hikim et al. ( 2002 ) found a dose-dependent increase in the mean number of myonuclei found in skeletal muscle fibers ( vastus lateralis muscle) with testosterone supplementation, as well as in the number of myonuclei per fiber (see also Eriksson et al. 2005 ). This increase was also associated with an increase in the number of satellite cells in the muscle tissue (but see Eriksson et al. 2005 ). Satellite cells are progenitor cells found external to muscle fibers that are incorporated into fibers and promote repair and growth of the muscle (Kadi and Thornell 2000 ; Reimann et al. 2000 ). However, the mechanism by which testosterone causes an increase in the number of satellite cells is unknown and could be due to testosterone (1) promoting cell division of satellite cells, (2) inhibiting apoptosis of satellite cells, or (3) causing differentiation of stem cells into satellite cells (Sinha-Hikim 2002 ). In any case, the functional implications for these findings are clear. More satellite cells likely result in more myonuclei per fiber, which, combined with increased protein synthesis, contribute to increases in muscle growth via an increased number and hypertrophy of muscle fibers (Kadi 2000 ; Kadi and Thornell 2000 ).

Testosterone also appears to cause a dose-dependent increase in the cross-sectional area of muscle fibers, although details about which types of fibers are affected and where in the body this occurs remains equivocal. Testosterone may increase the cross-sectional area of both type I (oxidative “slow twitch”) and type II (glycolytic “fast twitch”) fibers simultaneously after administration (Sinha-Hikim 2002 ; Eriksson et al. 2005 ), but other studies have shown greater increases in type I than in type II fibers (Hartgens et al. 1996 ; Kadi et al. 1999 ; also in growing rats, Ustunelet al. 2003 ), increased size in only type I fibers (Alén et al. 1984 ; Kuipers et al. 1991 , 1993 ), or increased size in only type II fibers (Hartgens et al. 2002 ). These mixed results are intriguing, because they suggest that different parts of the body, and, hence, different performance traits, may be affected differently by elevated testosterone levels. The likely mechanism for these differences is variation in density of receptors within the myonuclei of muscle fibers in different regions of the body (Kadi 2000 ; Kadi et al. 2000 ). An alternative hypothesis is that different types of fiber have differing relationships between the number of internal myonuclei and muscle cross-sectional area during hypertrophy (Bruusgaard et al. 2003 ). That is, some types of fibers may have internal myonuclei that can serve larger “nuclear domains” than can other types of fibers (reviewed by Gundersen and Bruusgaard 2008 ). If either of these hypothesized mechanisms is correct, then circulating levels of testosterone may only explain a portion of inter-individual (or interspecific) variation in performance. Testosterone may also stimulate changes in the proportions of types of fibers in muscles (Holmang et al. 1990 ; Pette and Staron 1997 ), although evidence for this effect in humans is mixed. For example, Sinha-Hikim et al. ( 2002 ) did not observe a change in the proportions of type I and type II fibers after administration of testosterone.

Changes in lower-level traits (e.g., protein synthesis, number of satellite cells, cross sectional area of muscle fibers) after testosterone supplementation, as described above, thus, result in changes at the whole-muscle level and explain many of the classic effects of testosterone that are desired by humans using steroids. That is, increasing testosterone via steroid use increases body weight, lean body mass, as well as cross-sectional area, circumference, and mass of individual muscles (i.e., “body dimensions”); however, there are numerous studies with contradictory results, finding no change in one, or all, of these traits, depending on the drug used, the dose taken, and the duration of use (reviewed by Bhasin et al. 2001 ; Hartgens and Kuipers 2004 ). The finding that testosterone can change muscle physiology and increase whole-muscle size and/or body mass is consistent with results in nonhuman animals. For example, testosterone implants increased size and number of fibers in the sonic muscles of male plainfin midshipman fish ( Porichthys notatus ) (Brantley et al. 1993 ). Similarly, testosterone supplementation increased muscle mass and changed contractile properties of trunk muscles of male grey treefrogs ( Hyla chrysoscelis ) (Girgenrath and Marsh 2003) and of forelimb muscles of male frogs ( Xenopus laevis , Regnier and Herrera 1993 ; Rana pipiens , Sidor and Blackburn 1998 ).

Effects of testosterone on humans’ performance

Whether steroids actually enhance performance of athletes was a subject of great controversy throughout the 1980s and 1990s (Ryan 1981 ; Haupt and Rovere 1984 ; Cowart 1987 ; Wilson 1988 ; Elashoff et al. 1991 ; Strauss and Yesalis 1991 ; Hartgens and Kuipers 2004 ), largely due to flaws in design of early studies (see above). However, the past decade has seen a surge in more carefully designed studies that have convincingly tested whether, all else equal, steroids increase performance. Hartgens and Kuipers ( 2004 ) found that 21 out of 29 studies they reviewed found an increase in humans’ strength after steroid use, with improvements in strength ranging from 5% to 20%. Storer et al. ( 2003 ) found that testosterone caused a dose-dependent increase in maximal voluntary strength of the leg (i.e., amount of weight lifted in a leg press), as well as in leg power (i.e., the rate of force generation). They further tested whether increased muscle strength was due simply to increased muscle mass or to changes in the contractile quality of muscle affected by testosterone, but they found no change in specific tension, or in the amount of force generated per unit volume of muscle. This latter result suggests that, at least for leg-press performance, testosterone increases strength by increasing muscle mass and not by changing contractile properties. Rogerson et al. ( 2007 ) found that supraphysiological doses of testosterone increased maximal voluntary strength during bench presses (see also Giorgi et al. 1999 ) and increased output of work and output of power during cycle sprinting compared to placebo control subjects. Thus, “burst” or “sprint” performance traits appear to be enhanced by increased testosterone, and this is in general agreement with studies of nonhuman animals (John-Alder et al. 1996 , 1997 ; Klukowski et al. 1998 ; Husak et al. 2007 ). For example, experimentally elevated levels of testosterone caused increased sprint speed, relative to sham-implanted individuals, in northern fence lizards ( Sceloporus undulatus ) (Klukowski et al. 1998 ). These findings contrast with results for endurance events, in which no increase in performance has been detected experimentally in humans (reviewed in George 2003 ; Hartgens and Kuipers 2004 ). The finding that endurance by humans is not enhanced by testosterone is unexpected since testosterone may increase hemoglobin concentrations and hematocrit (Alén 1985 , but see Hartgens and Kuipers 2004 ) and exogenous testosterone increases endurance in rats (Tamaki et al. 2001 ) and male side-blotched lizards ( Uta stansburiana ) (Sinervo et al. 2000 ). More studies of the effects of increased testosterone on endurance would help to better clarify these seemingly paradoxical findings. One possibility that might explain species’ differences in endurance is the relative proportion of type I fibers available for enhancement, which likely varies across species (Bonine et al. 2005 ), although this hypothesis needs explicit testing. Steroid use does not seem to consistently enhance recovery time after strenuous exercise (reviewed in Hartgens and Kuipers 2004 ), although it may in non-human animals (Tamaki et al. 2001 ). Storer et al. ( 2003 ) also found no change in fatigability (i.e., the ability of a muscle to persist in performing a task) of muscle during exercise, which is consistent with other studies (George 2003 ).

One of the problems in early studies of steroid effects was that the participants’ history of training and exercise while taking steroids was not taken into account or controlled (Bhasin et al. 2001 ; George 2003 ; Hartgens and Kuipers 2004 ). Recent studies have shown that the presence or absence of exercise training during testosterone supplementation can have a marked impact on how much performance is enhanced, thus complicating results when training is not controlled. Bhasin et al. ( 2001 ) reviewed several examples of such results. They pointed out that testosterone supplementation alone may increase strength from baseline levels, but so will exercise alone with a placebo, such that strength levels with exercise alone are comparable to those with testosterone addition alone (Bhasin et al. 1996 ). Testosterone supplementation while undergoing exercise training typically has the greatest increase in strength compared to exercise only or testosterone only (Bhasin et al. 1996 , 2001 ). These findings are consistent with those of others (reviewed by George 2003 ). Indeed, George ( 2003 ) suggested that steroids will only consistently enhance strength if three conditions are met: (1) steroids are given to individuals who have been training and who continue to train while taking steroids, (2) the experimental subjects have a high protein diet throughout the experiment, and (3) changes in performance are measured by the technique with which the individuals were training while taking steroids. That is, one may, or may not, find a change in bench-press performance if individuals trained with leg presses, and not bench presses, while taking steroids. We note that the confounding effect of training is a rather intuitive finding, but it does point out potential problems in studies of non-human animals, specifically laboratory studies, which we address below.

Given the effects of steroids on physiology and performance of human muscle, what can integrative biologists take away from these findings? We suggest that they can provide some valuable insights into the mechanisms of how hormones might regulate whole-animal performance traits in nonhuman animals. The most obvious lesson is that manipulating the circulating levels of testosterone, or its derivatives, increases overall strength, which has apparent benefits for performance in bursts, such as sprint speed. In contrast, there is little evidence from studies on humans for a positive effect on capacity for endurance, which is counter-intuitive, given the known effect of testosterone on hemoglobin concentrations and hematocrit. However, these same studies of humans also raise a host of issues that merit special consideration by researchers interested in hormonal effects on nonhuman animals, including effect of training, timing of administration, and dosage administered. We also argue that more information is needed on the long-term effects of hormonal manipulation on performance and fitness. Although recent studies suggest that increasing testosterone levels can enhance certain types of performance, we are not advocating or justifying the use of steroids by humans. There are numerous side effects of prolonged steroid use in humans, including cardiovascular problems, impaired reproductive function, altered behavior, increased risk of certain tumors and cancers, and decreased immune function, among others (reviewed by Pärssinen and Seppälä 2002 ; George 2003 ). These “side-effects” are in accordance with studies of nonhuman animals where higher testosterone levels are associated with such detrimental effects as increased loads of parasites, reduced immunocompetence, decreased body condition, reduced growth, and increased use of energy, ultimately resulting in reduced survival (Marler and Moore 1988 ; Folstad and Karter 1992 ; Salvador et al. 1996 ; Wikelski et al. 1999 , 2004 ; Moore et al. 2000 ; Peters 2000 ; Klukowski and Nelson 2001 ; Wingfield et al. 2001 ; Hau et al. 2004 ). Indeed, it is the presence of these very “side-effects” that has driven a great deal of research on behavioral and life-history tradeoffs mediated by testosterone (Ketterson and Nolan 1999 ; Ketterson et al. 2001 ). Higher levels of testosterone may enhance performance and increase success at some tasks, but its widespread “pleiotropic” effects on other aspects of the phenotype may result in a net detriment to fitness (Raouf et al. 1997 ; Reed et al. 2006 ; Ketterson et al. 2009).

We encourage researchers to complete more detailed studies of the interactions among hormones, morphology, and performance, especially across different types of performance traits (dynamic versus regulatory, see Husak et al. 2009a ). Comparative data on whether the same, or different, hormones affect the same performance traits in different taxa (e.g., burst speed in fish, sprint speed in lizards) would be useful for understanding how different species have evolved unique, or conserved, endocrine control of morphology and function. A comparative approach is important, as other studies have shown different effects of testosterone on performance in different taxa (e.g., an increase in endurance for rats and lizards, but none for humans), and more research is needed to determine whether such differences are valid or purely methodological. Even though testosterone is confined to vertebrates, it is possible that studies with invertebrates may reveal similar effects on performance via different hormones, e.g., recent work showing a seemingly similar role of juvenile hormone for invertebrates as testosterone has for vertebrates (Contreras-Garduno et al. 2009 ; see also Zera 2006 ; Zera et al. 2007 ; Lorenz and Gäde 2009).

Correlative studies relating endogenous circulating hormone levels to natural variation in performance traits can provide valuable insight into potential mechanistic regulators of performance, but manipulations allow a more detailed examination of cause-and-effect relationships. Whether performance can be manipulated by reduction (castration) or supplementation (implants) of testosterone in nonhuman animals will depend on the type of performance and how it is affected by circulating levels of the androgen. Many dynamic performance traits, especially maximal performance, may show different responses to exogenous hormone in the laboratory versus field, compared to coloration or “behavioral” traits. For example, supplementation with testosterone may rapidly increase display behavior or aggression in the laboratory (Lovern et al. 2001 ; Hews and Quinn 2003 ) compared to control animals, or corticosterone supplementation may decrease sexually selected color patterns (reviewed by Husak and Moore 2008 ). These examples are in contrast to supplementing testosterone in the laboratory and testing for an effect on performance. Aggression and coloration will not likely require training of the target trait to reveal an observed effect, whereas some performance traits may require training. Furthermore, regulatory performance traits (e.g., regulation of ions in seawater), on the other hand, may respond more directly to hormonal manipulation (see McCormick 2009), and will likely not require any training, but more empirical data are necessary to make generalizations.

It is also important to more closely inspect those traits that show no significant effect of testosterone on dynamic performance after manipulation in the laboratory. Such a “noneffect” may be due to numerous possibilities, the most obvious of which is that testosterone simply has no effect on a particular type of performance. However, a second possibility is that muscles involved in performance were not adequately trained during administration of supplemental testosterone, or there was no control of exercise during the period of testosterone administration. As an hypothetical example, one might not expect to see a large increase in the maximal flight speed of birds that were never allowed to fly following administration of exogenous testosterone. Indeed, Gallotia galloti lizards given exogenous testosterone were compared to lizards given sham implants and there was no difference in maximal bite force at the end of the experiment (K. Huyghe, J.F. Husak, R. Van Damme, M. Molina-Borja, A. Herrel, in review), despite increases in mass of the jaw muscles in testosterone-supplemented males. One possible explanation for this result is that these lizards did not “train” their jaw muscles enough while in captivity to increase muscle mass sufficiently to result in a measurable enhancement of performance. It is also possible that receptor density is very low or becomes low in trained muscles. Nevertheless, while training in animals seems straightforward in principle, in practice it is far trickier, and there also appear to be striking differences among species in the effects of training. Whereas some studies of mammals have successfully increased performance through training in a laboratory (Brooks and Fahey 1984 ; Astrand and Rodahl 1986 ), similar studies with lizards have found no effect (Gleeson 1979 ; Garland et al. 1987 ). In addition, while training might be successful with animals acclimated to a laboratory setting, inducement of stress, with a concomitant effect on corticosterone (Moore and Jessop 2003 ), and potentially circulating testosterone levels, is a significant confounding factor. Another complementary option is to use field studies, where experimental groups are released into the wild to “train” themselves while accomplishing their day-to-day tasks and performing naturally. Of course, this approach also cannot take into account variation in “training” within experimental groups, as individuals will likely use their performance traits in different ways when left to their own devices. Consequently, this approach could result in unpredictable results in how hormones impact performance, unless one accepts the unlikely assumption that all experimental animals are performing in the same ways. Further, a field approach also does not take into account other “pleiotropic” effects of increased (or decreased) testosterone on the phenotype (e.g., increased activity or conspicuousness to predators), which can eliminate potential benefits to fitness from enhanced performance due to testosterone supplementation.

Studies seeking to manipulate performance with testosterone supplementation should also consider the timing of experiments. For example, testosterone should ideally be increased or decreased during times when the hypothalamic–pituitary–gonad (HPG) axis is responsive and receptors are expressed in the appropriate target tissues. Seasonal sensitivity of the male HPG axis is well documented (Fusani et al. 2000 ; Jawor et al. 2006 ; Ball and Ketterson 2008 ), and such effects should be considered. For example, male green anoles ( Anolis carolinensis ) given exogenous testosterone after the end of the breeding season in a laboratory setting did not increase head size or bite-force performance (J. Henningsen, J. Husak, D. Irschick, and I. Moore, unpublished data), presumably because some or all of the relevant target tissues were no longer sensitive to androgens. On the other hand, male brown anoles ( Anolis sagrei ) did show enhanced maximal bite force when testosterone was supplemented at the beginning of the breeding season when the target tissues are presumably sensitive to androgens (Cox et al., in press). Timing of experimentation is thus critical for designing studies examining hormonal effects, and the interaction between timing and training should also be considered, as training effects may be relevant for some seasonal periods, but not for others.

A related issue concerns how much hormone to administer to experimental subjects. Studies of human steroid use typically involve supraphysiological doses of testosterone, as this is the typical regimen for steroid-abusing athletes (George 2003 ; Hartgens and Kuipers 2004 ). Indeed, many studies of steroid use by humans have been criticized for having experimental groups using physiological doses of testosterone. However, such criticism of seemingly unrealistic dosages highlights the differing goals of studies on human and non-human animals. Whereas studies of humans are focused on the role of supraphysiological doses on performance, those of nonhuman animals are more broadly interested in whether circulating testosterone affects performance within more natural bounds of variation (reviewed by Fusani et al. 2005 ; Fusani 2008 ). Supraphysiological doses can result in unexpected, or even counterintuitive, effects because endocrine systems tend to be homeostatic and compensatory after disruption via up- or down-regulation of various components within the system (Brown and Follett 1977 ).

There are few data on how testosterone affects dynamic performance during different stages of development, either in humans or in non-human animals. Practically all studies examining the effects of exogenous testosterone on humans have been on adults (reviewed by Hartgens and Kuipers 2004 ), but an increasing area of concern is steroid use by teenagers (Johnston et al. 2005 ). Because they are still developing physically, steroids may have dramatically different effects on dynamic performance in developing juveniles versus older adults. For example, steroid use is known to cause closure of growth plates of long bones (George 2003 ), potentially preventing growth to full height. Any manipulative hormone study examining effects on dynamic performance should also take baseline circulating levels into account, as there may be striking differences among age groups. For example, among sexually mature male green anole lizards in a well-studied New Orleans, Louisiana (USA) population, smaller “lightweight” males have lower circulating testosterone levels (Husak et al. 2007 , 2009b ), relatively smaller heads, and lower bite forces than do larger “heavyweight” males (see Lailvaux et al., 2004; Vanhooydonck et al., 2005a), with the difference apparently due to age (Irschick and Lailvaux 2006). Smaller males with low testosterone levels seem unable to produce higher levels (Husak et al. 2009b ), suggesting that testosterone levels are likely suppressed until a critical body size when the individuals become competitive with larger males. At this body size, elevated testosterone levels may accelerate growth of the head and increase bite force, although more data are needed to test this hypothesis. This ontogenetic increase in testosterone levels suggests that exogenous administration will have quite different effects on different age groups. For example, many hormones exert threshold effects (reviewed in Hews and Moore 1997 ) in which increased amounts above a threshold level produce little noticeable effect, suggesting that exogenous administration may accomplish little for larger lizards already with high testosterone levels, but may have substantial effects on smaller lizards with low testosterone levels.

In this context, long-term studies in animal species that focus on younger individuals (see Cox and John-Alder 2005 and references therein) might be useful for understanding the potential costs and benefits of hormones in improving or decreasing dynamic performance. Scientists are well-aware of some of the short-term activational effects of testosterone in humans and nonhuman animals, but while some long-term effects of supraphysiological doses on human health are recognized (see Hartgens and Kuipers 2004 ), we know far less about long-term effects of elevated (but not supraphysiological) testosterone levels on longevity and lifetime reproductive success of nonhuman animals. Ethical considerations may preclude long-term hormone implantation in humans and nonhuman animals, but correlating natural variation in testosterone levels both with performance traits and with other demographic features, such as longevity and lifetime reproductive success, would be useful for understanding chronic effects. Elegant studies with the dark-eyed junco ( Junco hyemalis ) (Ketterson et al. 2001 ; Reed et al. 2006 ) show complex trade-offs between different components of reproductive success (e.g., investment in extra-pair fertilizations versus parental care) as a result of testosterone supplementation; other similar trade-offs might be occurring over longer time spans in other animal species.

Despite popular interest in steroids and their effects on human athletic performance, we still lack a broad understanding of the effects of testosterone on performance in different animal species.

Our review of the literature on human steroids highlights several issues that could prove useful for integrative biologists interested in determining links among hormones, morphology, performance, and fitness in nonhuman animal species. First, studies of steroid use by humans reveal many caveats related to experimental design and interpretation that should be considered by those studying nonhuman animals (e.g., training, diet, dosage effects). Second, because of conflicting results of testosterone on different performance traits (e.g., burst performance versus endurance), more data are needed for such biomechanically opposing performance traits; testosterone may enhance multiple kinds of performance in some species, and only one kind in another. Third, while testosterone may have some general effects on dynamic performance in vertebrates, are there other hormones (e.g., juvenile hormone) that play a similar role in invertebrates? Finally, human steroid abusers often use various systems of “stacking”, where multiple drugs are taken in a specific order (George 2003 ), and such regimens are believed, by those who use them, to markedly increase dynamic performance. However, few studies have specifically examined how these regimes affect performance, or how the different regimes may be more, or less, effective in enhancing performance, either in humans or in non-human animal species. Furthermore, such practices are not restricted to multiple androgens, but may also include other hormones, such as growth hormone and insulin-like growth factor-I, which may, when taken exogenously, also enhance athletic performance and other aspects of the phenotype (Gibney et al. 2007 ). In this manner, the interactive effects of different hormone regimens for increasing animal performance are highly understudied. In conclusion, we have advocated an integrative approach for studying the evolution of morphology, function, and endocrine systems, and increased collaboration between researchers interested in human and in other animal systems may prove fruitful for both groups.

Financial support was provided by the National Science Foundation (IOS 0421917 to DJI and IOS 0852821 to I. T. Moore, JFH and DJI).

We are thankful to the symposium participants for fruitful discussions about hormones and performance. We thank the Society for Integrative and Comparative Biology, especially the Divisions of Animal Behavior, Comparative Endocrinology, and Vertebrate Morphology, for providing logistical and financial support.

Google Scholar

Google Preview

Author notes

Email alerts

Citing articles via.

• Recommend to your Library

Affiliations

• Online ISSN 1557-7023
• Print ISSN 1540-7063
• Copyright © 2024 The Society for Integrative and Comparative Biology
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

• Access Member Benefits

In This Section:

Anti-doping efforts in paris 2024.

After years of waiting, the summer Olympics are finally here. All attention will turn to the amazing feats of athletes from all over the world. Spectators, media and fans will witness the highs and lows as dreams are made (and broken) against the spectacular backdrop of Paris. Many friends and family will debrief after the performances they have witnessed and inevitably the topic of performance enhancing drugs will be raised by someone. It’s usually a topic only talked about if there is a big scandal but anti-doping is a global sporting movement with not only a crucial role at the Games but every day of the Olympic cycle.

Olympic and paralympic hopefuls are bound by the World Anti-doping Code which sets out rules to ensure a level playing field across all sports and countries. A number of important areas of the Code relate to Testing and Whereabouts. Day to day, athletes may be tested by their National Anti-doping organization of their International Sport Federation both in or out of competition tests. Many of us associate drug testing with those that occur immediately after a sporting event, known as in-competition or IC testing. These tests are of course very important to verify a “clean podium” at the time of the event, but the majority of anti-doping tests that are conducted around the world actually occur during an athlete’s everyday life – known as out-of-competition (OOC) testing. Elite athletes who are part of a registered testing pool are required to submit “Whereabouts” information to anti-doping authorities which provides information on where they are staying as well as known locations and times for training, work and study. On top of this, athletes must provide a daily one-hour window where they can be located. Failure to update or be where the athlete says they will be in the one-hour window can have significant consequences for athletes under the Code, but it is obvious that this is a big responsibility and a fair burden on athletes, especially those with busy travel and competition schedules.

Athletes may be required to provide any or all of the following approved sample types – urine, blood or dried blood spots (DBS). Sample must be collected by an accredited Doping Control Officer (DCO) who must follow strict international standards for sample collection and transport, not only to ensure the integrity of the sample but also to protect the Athlete’s rights. Samples must analyzed at a World Anti-doping Agency (WADA) Accredited Laboratory, again following the International Standard for analysis. Anti-doping authorities will determine the sample type and analysis according to the sport and type of athlete being tested. Very simply, specialized testing for anabolic steroids may be more common in sports where muscle size and strength is advantageous, whereas testing for erythropoietic stimulating agents such as EPO may be more common in endurance sports. However, since the substances named on WADA’s Prohibited List of substances apply to all sports (with a few exceptions), any substances can be tested for regardless of the athlete’s sport. In addition to direct testing, longitudinal monitoring of certain biomarkers in blood and urine has also been adopted in anti-doping, resulting in the concept of the Athlete’s Biological Passport (ABP).

Anti-doping testing has been “part of the job” for the most of the athletes here at the Olympics for the past three years, or more for the seasoned campaigners, with out of competition testing a necessary feature as well as IC testing at trials and qualifying events. Here in Paris, this doesn’t stop! Under the jurisdiction of the International Olympic Committee, anti-doping testing continues from the opening of the Olympic Village and at every competition venue. All being well, anti-doping will continue its hard work in the background without scandal and the dinner conversations can be more focused on the athlete’s performances.

Laura Lewis, Ph.D. , Born in the UK, Laura moved to Australia in 2005 to begin a graduate training position in the Physiology department at the Australian Institute of Sport. She began a sports-based Ph.D. in 2007, examining in the role of hemoglobin mass on cycling performance, with her research taking her to Europe and the US with the Australian National cycling team. Graduating in 2011, Laura continued to work as an applied sports scientist with Australia’s aspiring Olympians from a range of sports including water polo, soccer and track and field. She maintained her research interests, particularly in Altitude and anti-doping, combining the two in 2014 during a challenging project at the Tour of Qinghai Lake in northern China. Laura worked as a Research Fellow at the Australian Institute of Sport and Australian Catholic University and was awarded a PCC grant in 2015 to study the combined effects of altitude and iron supplementation on the Athlete Biological Passport. In 2020, Laura joined the U.S. Anti-Doping Agency as Director of Science, where she works alongside the testing, education and legal teams, advocating for clean sport and Athlete’s rights to fair competition.

• 2024 Election Results
• News Archive
• Certification Task Force
• Advertise with ACSM
• Current Partners
• Exercise is Medicine
• Rankings Archive
• Blogs and Resources
• National Youth Sports Health & Safety Institute
• Annual Report 2022
• Annual Report 2023
• Annual Report 2024
• Honor & Citation Awards
• Strategic Plan
• Student Membership
• Alliance Membership
• Professional-in-Training Membership
• Professional Membership
• Renew Membership
• Regional Chapters
• Member Code of Ethics
• ACSM Member Spotlight
• Group Exercise Instructor
• ACSM Personal Trainer Prep
• ACSM Exercise Physiologist Prep
• ACSM Clinical Exercise Physiologist Prep
• Beijing Institute of Sports Medicine
• Wellness Academy
• Frequently Asked Questions
• Recertification FAQs
• Find an ACSM Certified Professional
• Certified Professional of the Year
• Wellcoaches
• Certified Professional Discounts
• Hire ACSM Certified Professionals
• Specialty Certificate Programs
• ceOnline FAQs
• Approved Providers
• Current Sports Medicine Reports
• Exercise and Sport Sciences Reviews
• Exercise, Sport, and Movement
• Health & Fitness Journal
• Medicine & Science in Sports & Exercise
• Translational Journal
• Paper of the Year Awards
• ACSM's Guidelines for Exercise Testing and Prescription
• ACSM's Resources for the Personal Trainer
• ACSM's Resources for the Exercise Physiologist
• ACSM's Clinical Exercise Physiology
• ACSM's Resources for the Group Exercise Instructor
• ACSM's Certification Review
• ACSM's Foundations of Strength Training and Conditioning
• ACSM's Nutrition Exercise Science
• ACSM's Essentials of Youth Fitness
• ACSM's Introduction to Exercise Science
• ACSM's Health/Fitness Facility Standards and Guidelines
• ACSM’s Body Composition Assessment
• ACSM's Complete Guide to Fitness and Health
• Preparticipation Physical Evaluation (PPE) Monograph, 5th Edition
• ACSM's Fitness Assessment Manual
• ACSMs Exercise Testing and Prescription
• Textbook Adoption
• Translated Position Stands
• ACSM Official Statements
• Team Physician Consensus Statements
• Resource Library
• ACSM Fitness Trends
• Autism and Exercise
• Sudden Cardiac Arrest
• Mental Health
• Physical Activity Guidelines
• Reducing Sedentary Behavior
• Sex Differences and Transgender Athlete Care
• Faculty Resources
• EIM Clinical Resources
• Black History Month
• ACSM's Brown Bag Series (archived)
• Emerging Physician Leaders Pilot Program
• Annual Meeting
• IDEA & ACSM Health & Fitness Summit
• Advanced Team Physician Course
• Sports Medicine Essentials
• Integrative Physiology of Exercise Conference
• International Team Physician Course
• Regional Chapter Meetings
• Meeting Exhibits and Sponsors
• Research & Program Grants
• Howard G. "Skip" Knuttgen Scholar Award
• Travel and Research Awards
• ACSM Research Grant Recipients
• Dedicated Endowments & Funds
• Planned Giving / Discovery Society

Olympic officials address gender eligibility as boxers prepare to fight

PARIS – The case of two Olympic boxers has drawn attention to a thorny issue: Who and what determines which female athletes can compete.

Algerian boxer Imane Khelif and Lin Yu-ting of Taiwan both were disqualified from the 2023 women’s boxing world championships when they reportedly failed gender eligibility tests.

But this week, the International Olympic Committee confirmed the two boxers have been cleared to compete here at the Paris Games , as they both did at the Tokyo Games in 2021. The issues of so-called gender verification or sex testing have fueled discussion at the Olympics as the fighters prepare to enter the ring at North Paris Arena.

Khelif, a silver medalist at the 2022 world championships, is scheduled to fight Thursday against Angela Carini of Italy in the welterweight division at 146 pounds. Lin, a two-time world champion, is scheduled to fight Sitora Turdibekova of Uzbekistan in the featherweight division at 126 pounds.

“Yeah, it’s really tricky," Australian boxer Tiana Echegaray told reporters Tuesday when asked about the situation. "I don’t know exactly what their circumstances are."

IOC spokesman Mark Adams indicated Tuesday no personal information about the boxers' medical histories would be disclosed. "They've been competing in boxing for a very long time," Adams told reporters. “They've achieved all the eligibility requirements in terms of sex and age. We're following the rules in place in Tokyo."

Who's in charge of boxing?

At the Summer Olympics, when it comes to gender eligibility, the IOC defers to the international federations that govern each of the 32 sports.

The IOC does provide a framework to the international federations . But it's “nonbinding."

In other words, it’s not up to the IOC. And the situation has grown especially complicated with boxing.

Last year the IOC banished the International Boxing Association (IBA), long plagued with scandal and controversy that jeopardized the future of Olympic boxing. In fact, the IOC denied IBA the right to run Olympic boxing during the Tokyo Games in 2021 and instead turned over control to an ad-hoc unit.

Opinion: Olympic female boxers are being attacked. Let's just slow down and look at the facts

With that ad-hoc unit in charge, Kehlif and Lin both competed at the Tokyo Olympics. Neither won a medal.

But the IBA has maintained control of the world championships and gender eligibility rules. And after Lin won gold and Kehlif won bronze at the event in March 2023, officials announced the boxers had failed medical eligibility tests and stripped them of the medals.

IBA president Umar Kremlev said DNA tests “proved they had XY chromosomes and were thus excluded."

What's the eligibility criteria?

A passport could be key, based on comments from Adams, the IOC spokesman.

“I would just say that everyone competing in the women’s category is complying with the competition eligibility rules," he said. “They are women in their passports and it is stated that is the case.”

Thursday Adams added that the issues with the previous tests for the boxers "was not a transgender issue, there's been some misreporting on that in press. ... These women have been competing as women for many years.

"What I would say just quickly on testosterone is, the testosterone (test) is not a perfect test. Many women can have testosterone, even what would be called 'male levels' and still be women and still compete as women. So this is not a panacea − this idea that suddenly you test, do one test for testosterone. Each sport needs to deal with this issue but I think we agreed, I hope we're agreed, we're not going to go back to the bad old days of 'sex testing'. That would be a bad idea."

In the past, other eligibility standards have hinged on science.

Caster Semenya, a two-time Olympic gold medalist in track and field in 2012 and 2016, was forced to give up competing in the 800 meters because her testosterone levels were too high based on tests administered by World Athletics, the sport’s international federation previously known as the IAAF.

Semenya was assigned female at birth. She said she was told at age 18 that she has XY chromosomes and naturally had high levels of testosterone.

Khelif and Lin have not publicly addressed details of their medical histories regarding the tests.

The issue of eligibility surfaced as a source of controversy in the United States in 2022 when swimmer Lia Thomas became the first openly transgender athlete to win an NCAA championship.

At the time, the NCAA required transgender female athletes to have undergone one year of testosterone suppression treatment to be eligible to compete on a women's team in any sport. The NCAA has been under pressure to update its guidelines after the NAIA banned all transgender athletes from competing in women's sports.

The Court of Arbitration for Sport upheld a decision in June by World Aquatics, the international federation for swimming, that prevented Thomas from competing in elite competitions through World Aquatics or USA Swimming.

Who are these two boxers?

Lin, 28, has been fighting as an amateur for more than a decade, according to BoxRec, a widely regarded boxing site.

She made her official amateur debut about three months shy of her 18th birthday, winning at the 2013 AIBA World Women’s Championships. She won gold medals at the world championships in 2019 and 2022.

At 5-foot-9, she often has enjoyed a height advantage while amassing a record of 40-14 with one knockout. The record does not reflect the four fights she won at the 2023 world championships before her disqualification, which resulted in the outcome of the fights being changed to “no contest.’’

She lost her last fight – a split-decision defeat against Brazil’s Jucielen Cerqueira Romeu in April at the 2024 USA Boxing International Invitational in Pueblo, Colorado.

Khelif, 25, made her amateur debut at the 2018 Balkan Women's Tournament. She won a silver medal at the 2022 world championships.

At 5-foot-10, she also has enjoyed a height advantage while amassing a record of 36-9 with four knockouts, according to BoxRec. That does not include the three fights she won at the 2023 world championships before her disqualification resulted in the fights being changed to “no contest.’’

In one of those fights, Khelif stopped her opponent by TKO.

Contributing: Kim Hjelmgaard

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Curr Neuropharmacol
• v.13(1); 2015 Jan

Anabolic-androgenic Steroid use and Psychopathology in Athletes. A Systematic Review

Daria piacentino.

1 NESMOS (Neurosciences, Mental Health, and Sensory Organs) Department, School of Medicine and Psychology, Sapienza University–Rome, Italy; UOC Psychiatry, Sant’Andrea Hospital, Rome, Italy;

Georgios D. Kotzalidis

Antonio del casale.

6 Department of Psychiatric Rehabilitation, P. Alberto Mileno Onlus Foundation, San Francesco Institute, Vasto, Italy;

Maria Rosaria Aromatario

2 Department of Anatomical, Histological, Forensic Medicine, And Orthopedic Sciences. Sapienza University–Rome, Italy;

Cristoforo Pomara

3 Department of Forensic Pathology, University of Foggia; Ospedale Colonnello D'Avanzo, Foggia, Italy;

Paolo Girardi

5 Centro Lucio Bini, Rome, Italy;

Gabriele Sani

5 IRCCS Santa Lucia Foundation, Department of Clinical and Behavioral Neurology, Neuropsychiatry Laboratory, Rome, Italy;;

The use of anabolic-androgenic steroids (AASs) by professional and recreational athletes is increasing worldwide. The underlying motivations are mainly performance enhancement and body image improvement. AAS abuse and dependence, which are specifically classified and coded by the DSM-5, are not uncommon. AAS-using athletes are frequently present with psychiatric symptoms and disorders, mainly somatoform and eating, but also mood, and schizophrenia-related disorders. Some psychiatric disorders are typical of athletes, like muscle dysmorphia. This raises the issue of whether AAS use causes these disorders in athletes, by determining neuroadaptive changes in the reward neural circuit or by exacerbating stress vulnerability, or rather these are athletes with premorbid abnormal personalities or a history of psychiatric disorders who are attracted to AAS use, prompted by the desire to improve their appearance and control their weights. This may predispose to eating disorders, but AASs also show mood destabilizing effects, with longterm use inducing depression and short-term hypomania; withdrawal/discontinuation may be accompanied by depression. The effects of AASs on anxiety behavior are unclear and studies are inconsistent. AASs are also linked to psychotic behavior. The psychological characteristics that could prompt athletes to use AASs have not been elucidated.

INTRODUCTION

Professional and recreational athletes commonly use anabolic-androgenic steroids (AASs) to enhance performance or improve their physical appearance; the impact of this practice on psychopathology is unknown and so is the presence of psychopathology in those who will make later use of AASs. AASs include testosterone and its numerous synthetic analogs that have been modified to boost their anabolic, rather than their androgenic effects. The higher the anabolic:androgenic ratio, the higher the anabolic effect of a given AAS. Anabolic effects consist in protein synthesis, muscle growth, and erythropoiesis [ 1 - 3 ]. Therefore, they allow athletes to increase muscle size and reduce body fat. AAS users find that their muscles recover faster from intense strain and muscle injury, allowing them to train longer and harder [ 4 ]. However, Imanipour et al . [ 5 ] have shown AASs to increase serum creatine kinase and muscle damage. AASs produce their anabolic effects through binding to steroid receptors; they activate androgen receptors, thus controlling the transcription of target genes that regulate DNA accumulation required for muscle growth. When AASs bind to skeletal muscle androgen receptor, they cause an increase in muscular mass and strength, since amino acids are used more effectively for protein synthesis [ 1 , 6 ]. They also reduce glucocorticoid-dependent metabolic breakdown by binding competitively to glucocorticoid receptors [ 7 ].

AASs, due to their diverse biological actions, have shown benefit in a variety of conditions, including HIV-related muscle wasting, muscle dystrophies, severe burn injuries, bone marrow failure, hereditary angioedema, and growth retardation in children [ 8 - 3 ]. However, AAS use is associated with various dose-related side-effects. High doses of AASs can lead to serious physical and psychological complications, such as hypertension, atherosclerosis, myocardial hypertrophy and infarction, abnormal blood clotting, hepatotoxicity and hepatic tumors, tendon damage, reduced libido, and psychiatric/behavioral symptoms like aggressiveness and irritability [ 14 - 22 ]. In addition, AASs are related to hypofertility and gynecomastia in men [ 12 , 23 ] and to virilization in women, with hirsutism, male-pattern baldness, irregular menses, and lower-pitched voice [ 24 ].

AAS use to improve performance and acquire more muscular bodies is on the rise worldwide. In the US alone, at least two million individuals use or have used AASs [ 25 , 26 ] and epidemiologic data suggest that there are millions of other AAS users worldwide, from the UK [ 27 ] to Sweden [ 28 ], to other European countries [ 29 ], to Canada [ 30 ], Australia [ 31 ], and Brazil [ 32 ]. Economic costs vary, with annual per capita expense for AASs ranging from $90 to $6780. Men use AASs significantly more than women, even if the latter are using them increasingly [ 33 ]. AAS users frequently have at least one friend or acquaintance that uses or has used AASs. AAS use can also be associated with or predict future consumption of other types of psychoactive substances and might be part of a general pattern of poly-substance use and risk-taking behavior, especially among adolescents [ 34 - 37 ]. In fact, AAS use is no longer limited to athletes, bodybuilders, or weightlifters, but is currently extending to the general population, including young people, probably because of the highly competitive nature of high school and college varsity sports [ 38 - 41 ]. Welder and Melchert [ 42 ] reported that in the US over half a million high school students have taken AASs for nonmedical purposes. This raises serious concerns among physicians regarding the numerous adverse effects of these substances.

AASs have been added to the International Olympic Committee (IOC)’s list of banned substances in 1975; their use without physician’s prescription and supervision is illegal in the US and Canada and is termed “doping”. The World Anti-Doping Agency (WADA), which works for the IOC, has published in 2013 a list of prohibited substances, both in- and out-competition [ 43 ]; the list includes substances lacking governmental regulatory health authorities’ approval for human therapeutic use (e.g., discontinued drugs or still under preclinical or clinical development, designer drugs, substances approved only for veterinary use). It contains over 100 different AASs, both synthetic and natural. The most common are androstenediol, androstenedione, boldenone, danazol, dehydroepiandrosterone, dihydrotestosterone, methandienone, mesterolone, methenolone, nandrolone, oxandrolone, and tetrahydrogestrinone. These substances are formulated to be administered either orally, or parenterally by intramuscular injections, or subcutaneously by implantable pellets, or transdermally by patches or topical gels. It must be noted that some athletes also use legal “dietary supplements” that are not regulated in both US and Europe. These substances are not subjected to the same testing for safety and effectiveness as all over-the-counter and prescription medications. They are legal substances, sold in the form of pills, which the body converts into testosterone. How efficiently this converted testosterone might work is unknown, as is the extent to which it may obtain desirable effects, and it is suspected that these pills may be as harmful as AASs in other formulations.

The aim of this review is to investigate the relationship between AAS use and psychopathology in athletes and to identify possible prevention and treatment methods. We may suspect that exogenously administered testosterone and its synthetic analogs may alter the developmental trajectory of the brain in adolescents and young adults and its pattern of adaptation to environmental stimuli in middle-aged or older adults. AASs have numerous central nervous system effects, the extent of which varies with several factors, such as the athlete’s background resilience, the duration of AAS use and dose, concurrent organic diseases, and use of other medications, alcohol or illegal substances [ 44 - 47 ].

Eligibility Criteria

This systematic review was conducted according to the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) Statement [ 48 ]. Retrospective and prospective studies examining the relationship between AAS use and psychopathology in athletes were included. Psychiatric diagnoses had to be based on the Diagnostic and Statistical Manual 5 th Edition (DSM-5) criteria [ 49 ] or its predecessors or on the International Classification of Diseases 10 th revision (ICD-10) [ 50 ] criteria or its predecessor. Studies involving animals only or not including athletes were excluded.

Search Criteria and Critical Appraisal

The search strategy differed according to the database explored. For PubMed we used anabolic steroid users AND "Mental Disorders" [MeSH], limiting to humans, as a search strategy after trying several variations that produced an excess of irrelevant papers. The search yielded 151 papers as of July 27, 2014. Subsequent appraisal based on titles and abstracts left 55 papers. The same strategy with no time limits and without using the MeSH function in the other databases yielded 388 papers in ScienceDirect Scopus and four additional relevant papers, 263 papers and one additional with respect to the other two in the INFORMA healthcare database, 2953 papers in Excerpta Medica Database (EMBASE, limited to July 24, 2014) adding further two relevant papers, 5195 papers in PsycINFO (limited to the fourth week of July), adding further three relevant articles, and 18 papers searching for all anabolic steroid users in the Cochrane Central Register of Controlled Trials (CENTRAL), which added no new relevant paper (Fig. ​ 1 1 ). Reference lists of all located articles were further searched to detect still unidentified literature, but yielded no new articles. Methodological appraisal of each study was conducted according to PRISMA standards, including bias evaluation.

PRISMA flowchart of systematic review inclusion and exclusion.

Search Results and Included Studies

From each electronic database we read all titles and selected those promising to be relevant, which were 64. Subsequently, we read abstracts and downloaded the full text of those actually relevant, which amounted to 60 articles. The reviews were downloaded to further search their reference lists similarly to other papers, but they yielded no other potentially eligible article. Reviews were subsequently excluded. All reasons for exclusion of studies are shown in Fig. 1 . The final number of included papers was 37. Sample sizes for AAS users in included studies ranged from 1 (i.e., eight case reports) to 550, with a mean of 93 and a median of 45, indicating skewness towards smaller samples.

Study Characteristics

The following data were extracted from included studies: study source; type of study; total number of participants in the study; number of participants with AAS use/abuse/dependence; number of participants without AAS use (controls); study duration; criteria used for psycho-pathological diagnosis and any rating scale used for psycho-pathological screening; key findings. A data collection form was developed and used to extract data from the included studies. Data items extracted are listed in Table ​ 1 1 .

Data items extracted from the selected studies

Risk of Bias

No evidence of language bias was found, as the search was not limited to English language studies. This reduced the possibility of missing relevant studies. Moreover, no evidence of significant publication bias was found. The Cochrane database search produced no unpublished study, initiating, ongoing, or finished.

AASs and Mood Disorders

Bidirectional relationships between AAS use and mood disorders have been described, with physiological AAS doses affecting minimally mood and even showing beneficial effects (at least with regard to testosterone) in dysthymia and refractory depression [ 22 , 46 , 51 , 52 ] and supraphysiological doses being associated with depression, hypomania, or mania. Pope and Katz [ 53 ] interviewed bodybuilders and football players using AASs and found that 22% qualified for mood disorders. In a subsequent study, Pope and Katz [ 54 ] showed that 23% of athletes who abused AASs met DSM-III criteria for mood disorders, from major depressive disorder to type I and II bipolar disorders. Athletes met these criteria during AAS consumption significantly more than in the absence of it and significantly more than AAS nonusers (6%). The higher the AAS dose, the more severe the psychopathological symptoms. A controlled, retrospective study [ 55 ], comparing weightlifters with and without AAS use, found that AAS users lamented significantly more depressive symptoms when they were using AASs than when they were not, and also compared to nonusers. However, no difference in the frequency of major mood disorders was found between the current and past AAS users and the nonusers. The authors concluded that “the organic mood changes associated with AAS abuse usually present as a subsyndromal depressive disorder of insufficient severity to be classified as a psychiatric disorder”. Depressive symptomatology was prominent in a sample of power athletes who were former AAS users [ 56 ], implying that AAS use may have long-lasting effects on mood. Subsequently, the same group of investigators [ 57 ] compared mood alterations between male weightlifters with and without AAS use. AAS users scored higher on the Hamilton Rating Scale for Depression and on the Modified Mania Rating Scale than nonusers; however, no participant met formal criteria for any mood disorder. A retrospective study by Malone et al . [ 58 ] examined weightlifters and bodybuilders who used AASs and found hypomania in approximately 10% of the sample. Depression occurred in another 10% of the sample who discontinued AASs.

A study that recruited exclusively female athletes [ 59 ] found a 33% prevalence of current or past AAS use. AAS users reported poly-substance use more frequently than nonusers and showed hypomanic symptoms during AAS use (56%) and depressive symptoms during AAS withdrawal (40%), even if they did not meet full DSM-IV criteria for a hypomanic or a major depressive episode. Ip et al . [ 60 ] found that 15% of male AAS-dependent users reported a history of major depressive disorder, compared to only 7% of AAS-nondependent users. In addition, Pope and Katz [ 53 ] found AAS users to be at significant risk for major depressive episodes during the first months following AAS discontinuation. Similar results were obtained by Brower [ 61 , 62 ], who observed that AAS withdrawal usually presented with depressive symptoms, such as apathy, anhedonia, concentration issues, sleep changes, and decreased libido, especially after intense abuse. However, another study [ 4 ] failed to find differences on the Profile of Mood States Questionnaire between male weightlifters with AAS use vs . without AAS use; both samples did not differ from the general population.

Brower et al . [ 63 ] presented a case report of a 24 year-old weightlifter with AAS dependence and revealed how the uncontrolled pattern of AAS use persisted despite adverse effects, such as severe mood disorders – more specifically, depression –, marital conflict, and deterioration of the individual’s moral values. Subsequently, the same group of investigators [ 64 ] found prominent depressive symptoms in eight AAS-using weightlifters who met criteria for DSM-III-R substance dependence. A cross-sectional study of 130 bodybuilders showed how 38% of them had used AASs for a median length of time of two years [ 65 ]. Half of the AAS users reported unspecified mood disturbances.

Tennant et al . [ 66 ] authored a case report regarding a 23-year-old bodybuilder who admitted to being “addicted” to AASs and not being able to stop taking them without experiencing severe withdrawal symptoms, including low mood, feelings of hopelessness, loss of energy, nausea, and dizziness. Thus, a diagnosis of AAS dependence was hypothesized and the resemblance with opioid dependence was underlined.

Another case report described the effects of long-term AAS use [ 67 ]. A 25-year-old man, who had been involved in bodybuilding, martial arts, and illegal boxing matches or other fights and had used AASs since the early adolescence, underwent an involuntary psychiatric hospital admission for a manic episode. He had a prior history of psychotic depression and was diagnosed with substance-related bipolar disorder, attributed to AAS use.

Animal models seem to support the association between AAS abuse and mood disorders. Matrisciano et al . [ 68 ] injected subcutaneously 5 mg/kg of body weight nandrolone or stanozolol for four weeks (i.e., a dose considered to the one abused by athletes) in male adult rats; they found that AASs decreased hippocampal and prefronto-cortical brain-derived neurotrophic factor levels and hippocampal low-affinity glucocorticoid receptor expression, while increasing morning trough plasma corticosterone. Intrestingly, major depressive disorder has been variously associated with all these alterations. Furthermore, both nandrolone and stanozolol increased immobility of rats tested on the widely used antidepressant drug screening test, i.e., the forced swimming test. Co-administration of clomipramine, a tricyclic antidepressant drug, prevented all AAS-produced effects. The fact that supraphysiological AAS doses per se may be associated with changes indicating depression in normal rats, prompts us to consider that AAS abuse in humans may produce similar states without needing stress- or other risk factor-exposure. The effects of stanozolol in rats were also tested by Tucci et al . [ 69 ], who injected subcutaneously 5 mg/kg/day stanozolol vs . vehicle for four weeks in young male rats. Dopamine increased in the hippocampus and decreased in the prefrontal cortex, serotonin decreased in the prefrontal cortex, nucleus accumbens, striatum, and hippocampus, and norepinephrine decreased in the nucleus accumbens 24 hours after the last injection. All these changes are compatible with the neurochemistry of depression, hence, subchronic use of stanozolol is associated with rat brain biogenic amine changes that in humans could possibly lead to depression. Recently, Zotti et al . [ 70 ] injected subchronically 5 mg/kg/day nandrolone for 4 weeks in male rats, showing a depressive, but not anxious profile in animals, accompanied by brain region-dependent dopaminergic, serotonergic, and noradrenergic changes.

AASs and Suicide

Substance abuse in general is a risk factor for suicide [ 71 , 72 ]. Alink between AAS use and suicide in athletes was firstly suspected by Brower et al . [ 63 ], who reviewed case reports and newspaper articles and found that suicidal ideation and completed suicide were not infrequent among AAS users. They described a case of a 24 year-old weightlifter with AAS dependence and severe depression, who developed suicidal ideation after AAS withdrawal [ 73 ]. This user had no personal or family history of depression or suicidality.

In a case series of eight young men who committed suicide and were taking ( N =5) or had discontinued AASs ( N =2) (in one it was not possible to establish whether taking or discontinued), family members had noticed depressive symptoms associated with AAS discontinuation in five cases. After protracted AAS use, four had developed major depressive disorder. Two had manifested hypomanic symptoms before committing suicide. Only one had manifested suicidal ideation before starting AAS use [ 74 ]. The authors speculated that long-term AAS use may induce psychopathology, which may contribute to suicidality in predisposed individuals. The same group described a series of 34 violent deaths occurring in AAS users, which were due to suicide ( N =11), homicide ( N =9), accidents ( N= 12), or undetermined cause (N=2) [ 75 ]. In suicides, forensic history and significant others’ reports highlighted AAS-related impulsive behavior characterized by violent rage, mood swings, and propensity to depression.

Pärssinen et al . [ 76 ] studied prospectively for 12 years 62 male elite Finnish weightlifters with suspected AAS use ( N= 62), comparing them to the general population ( N= 1094). The athletes showed a 4.6 times higher mortality rate that was mainly due to suicide (5%) and myocardial infarction (5%). Two individuals belonging to the population control group committed suicide, compared to three belonging to the athlete group. Suicide committed in unspecified conditions was described in a case report of AAS-induced mood disorder with manic features occurring in a young bodybuilder/martial artist who was briefly hospitalized for a manic episode manifesting itself with psychomotor agitation, euphoria, irritability, and aggressiveness [ 67 ].

An autoptic study [ 77 ] compared toxicological findings and death modality in 52 deceased AAS users vs. 68 deceased heroin and/or amphetamine users who were negative for AASs. About 60% of AAS users were taking illicit substances. Compared to heroin/amphetamine related-deaths, AAS-related ones occurred earlier and more frequently by violence, suffered (23% AASs vs . 0% heroin/amphetamine; Fisher’s exact test, p <0.0001) or self-inflicted (21% AASs vs . 7% heroin/amphetamine, χ 2 , p <0.0001), suggesting that AAS users are highly suicidal and engage frequently in risky behaviors. AAS users’ suicides were caused by voluntary intoxication ( N =4), explosion/shooting ( N =4), hanging ( N =1), car crashing ( N =1), or jumping in front of a train ( N =1). No cases of defenestration occurred. A recent 30-year follow-back study [ 78 ] compared the rates and causes of death in 181 male former élite athletes of power sports “like wrestling, weightlifting, and some track and field disciplines” who were suspected for AAS use, with those of age- and gender-matched healthy controls. Although overall mortality was not increased in the athletes’ sample, individuals in the 20-50 age range in the former athlete group died 45% more often than people in the general population, and those in the 30-50 age range died from suicide 30% more often than the general population. Suicide was committed by 11.6% of the former athletes.

Darke et al . [ 79 ] studied 24 cases of abrupt or unnatural death among male AAS users, who were all in their early 30s. Deaths were due to inadvertent drug lethality (62.5%), suicide by gunshot or hanging (16.7%), and homicide (12.5%). In all individuals, except for one, substances other than AASs, mainly psychostimulants, opioids (e.g., heroin, morphine), and benzodiazepines, were detected. This case series study underlines how deaths among AAS users usually concern young male poly-substance users, whose AAS- or poly-substance-related aggressive, disinhibited, and depressive behavior may increase the risk of premature death.

As for the pathophysiological basis of the relationship between AASs and suicidal behavior, conflicting data on plasma testosterone levels in suicide attempters have been published. Low testosterone levels after suicide attempts were detected by two studies [ 80 , 81 ]; a study of male veterans with posttraumatic stress disorder showed that there was no association between testosterone levels and a history of suicide attempt [ 82 ]; a recent study reported no difference in testosterone levels between male suicide attempters and healthy controls [ 83 ]. Differently, another recent study of suicide attempters with bipolar disorder [ 84 ] found plasma testosterone to correlate with the number of manic episodes and suicide attempts, suggesting that testosterone could represent a pathophysiological link between mood disorders and suicidal behavior.

AASs and Anxiety Disorders

AAS use by adults, but even more by adolescents – at a time when affect-regulating brain areas are still developing and highly hormone/neuromodulator-sensitive – is associated with increased anxiety and altered responses to stress [ 85 ]. We already mentioned that 22% of 41 AAS-using bodybuilders and football players met DSM III-R criteria for at least one mood or anxiety disorder [ 53 ]. In a survey of 479 athletes recruited from Internet forums of several fitness, bodybuilding, weightlifting, and steroid websites, 16% of male AAS-dependent users had a history of DSM-IV-TR anxiety disorders (generalized anxiety disorder, panic disorder, posttraumatic stress disorder, obsessive-compulsive disorder, or social phobia), compared to only 8% of AAS-nondependent users [ 86 ]. A further study by the same research group [ 60 ] examined the characteristics of 67 male AAS users and 76 male nonusers, all aged 40 and older, who were recruited through fitness, weightlifting, bodybuilding, and steroid websites. Anxiety disorders were present in more AAS users than nonusers (12.0% vs . 2.6%; p <0.05).

The most recent animal studies, using a variety of experimental paradigms, support that both male and female animals are sensitive to the anxiogenic effects of AASs [ 87 - 89 ], the latter more so [ 90 ]. These studies, performed on rodents, used subchronic administrations of either 5 mg/kg nandrolone decanoate or 7.5 mg/kg of a mixture containing testosterone cypionate, nandrolone decanoate, and methandrostenolone. However, studies administering 3.5-5 mg/kg testosterone in rodents found anxiolytic effects [ 91 - 93 ]. These studies used in addition to the elevated plus maze test, the open-field and the Vogel conflict paradigms [ 94 ]. Another study that found anxiolytic effects for AASs employed 17-β-hydroxy-1-methyl-5α-androstan-3-one in male rats [ 95 ].

Finally, studies using 17α-methyltestosterone in mice found no influence of AASs on anxiety [ 96 - 97 ]. Summarizing, the inconsistency of animal studies focusing on the effects of AASs on anxiety-like behavior reflect species, sex, design, paradigm, and drug heterogeneity.

AASs, Somatoform and Eating Disorders

It is not uncommon for athletes to be affected by somatoform and/or eating disorders. The underlying psychological basis may share several features. Physical appearance and eating patterns are interrelated; lately, there is evidence that young boys and men are becoming as concerned about these aspects as young girls and women. However, women point at thinness, while men strive to increase muscular mass and body size, in line with media-endorsed prototypes. The latter have encountered major changes in the last 50 years and more and more individuals are using AASs simply to look good. The pressure to look good may be one of the main triggers for the above mentioned disorders.

Several studies point to the relationship between body image and eating-related attitudes and disorders. A study compared the psychological characteristics of females with anorexia nervosa and males taking part in bodybuilding competitions, who had both adopted strict standards of body perfection and displayed unhealthy behaviors, such as rigorous food restriction, extreme exercise, and AAS use to pursue their goals [ 98 ]. Findings confirmed that anorexic women’s and bodybuilders’ psychological profile was extremely similar. Both groups were significantly more obsessive, perfectionistic, anhedonic, and narcissistic than the general population. Yet, bodybuilders reported positive perceptions of themselves, as opposed to anorexic women. These results may be interpreted in the context of a theoretical model of anorexia nervosa and bodybuilding, which focuses on the role of personality in the onset and maintenance of excessive behaviors.

Another study compared exercise, body shape, eating habits, and weight-related symptomatology in a sample of 15 male gym-users, 21 males with muscle dysmorphia, and 24 males with anorexia nervosa [ 99 ]. Muscle dysmorphia, also known as reverse anorexia, or bigorexia, or Adonis complex, is a subtype of body dysmorphic disorder generally affecting men, with its onset in adolescence or early adulthood, characterized by obsessiveness and compulsivity directed towards achieving a lean and muscular physique, even at the expense of health. The study used various questionnaires and a measure of appearance- and performance-enhancing drugs (APED) use. Similarities in the domains of altered body image, disordered eating, and exercise behavior between men with muscle dysmorphia and men with anorexia nervosa were found, however the two groups differed in their pursued goals, which were opposite. Furthermore, significant correlations between muscle dysmorphia and eating disorder measures were observed. These findings provide support for the hypothesis that muscle dysmorphia and anorexia nervosa may have a nosological similarity. A study of male and female university students, dichotomized them into high (HSPA) and low (LSPA) social physique anxiety groups on the basis of their median scores on the Social Physique Anxiety Scale; students also filled out the Eating Attitudes Test and the Physical Activity Assessment Questionnaire [ 100 ]. Young men had healthier eating attitudes and greater physical activity levels than women. HSPA participants showed unhealthier eating attitudes and greater physical activity levels than LSPA participants. A group × gender interaction was found for eating attitudes, but not for physical activity. HSPA women scored higher on the Eating Attitudes test than HSPA men and LSPA men and women. Swami et al . [ 101 ] compared body size ideals, dissatisfaction with body image, and media influence among 88 female athletes – in particular, 41 track-and-field athletes and 47 martial artists – and 44 female nonathletes. There were no significant between-group differences in ideal body size after controlling for body mass index (BMI). By contrast, track-and-field athletes reported the highest dissatisfaction with body image and internalization of athletic media messages. Participants’ BMI and internalization of athletic media messages predicted dissatisfaction with body image for each sport type and for all sports. These results suggest that women participating in leanness-promoting sports experience greater dissatisfaction with body image than those in other sports or nonathletes.

The media have a relevant role in inducing body dissatisfaction, weight control, and muscle-development. They promote a body stereotype that emphasizes strength and muscularity for men and thinness for women, leading to poorer satisfaction with physical attractiveness and body dimensions and, ultimately, to related disorders [ 102 ]. A study investigating the interactions between media use and eating disorders in young adults found that media exposure significantly influenced men’s, but not women’s endorsement of personal thinness and dieting [ 103 ]. A study exploring the role of media in triggering weight concerns among preadolescent/early adolescent children, found that boys and girls who strived to resemble same-sex media icons were more likely than their peers to develop preoccupation with weight and become constant dieters [ 104 ].

As for the features of the relationship between AAS use and somatoform and/or eating disorders, Pope and Katz [ 54 ] showed that 18.2% of 88 male weightlifters who abused AASs reported a history of muscle dysmorphia, compared to none of the 68 male weightlifters who did not use AASs (controls). Blouin and Goldfield [ 105 ] examined the relationship between body image disturbances, eating attitudes, and AAS use in 43 male bodybuilders vs. 48 runners and 48 martial artists of the same sex, all recruited from fitness centers. Bodybuilders showed significantly greater body dissatisfaction, with a high tendency to bulk and thinness, and increased inclinations towards bulimia than the other two groups. Additionally, they reported higher perfectionism and ineffectiveness, as well as lower self-esteem. They also consumed more AASs and had freer attitudes towards AAS use. The main reason for taking AASs, according to AAS users, was physical improvement: AAS users reported a stronger drive to put on muscle mass in the form of bulk, more maturity fears, and greater tendencies towards bulimia than AAS nonusers. Thus, male bodybuilders seem to be at risk for body image disturbances and the associated psychopathological characteristics that have been commonly observed in patients with eating disorders. These psycho-pathological characteristics also appear to predict AAS use in this group of men.

One study examined body dissatisfaction, social anxiety, social physique anxiety, and upper body esteem among 135 male athletes and 50 male nonathletes [ 106 ]. The athletes were represented by 35 bodybuilders with AAS use, 50 bodybuilders without AAS use, and 50 athletically active exercisers (involved in aerobics, jogging, basketball, or racquetball) without AAS use. Results indicated that the AAS-using bodybuilder group was characterized by lower levels of social physique anxiety, higher upper body strength perception, and a trend towards higher body dissatisfaction than the nonuser groups (AAS-nonusing bodybuilders, athletically active exercisers, and nonexercisers). No difference in terms of social anxiety was found among the four groups.

Kanayama et al . [ 107 ] assessed 89 weightlifters, 48 of whom AAS current and past users and 41 AAS nonusers, on measures of self-esteem, attitudes towards male roles, body image, eating attitudes and disorders, and muscle dysmorphia. Current and past AAS users showed no significant differences from nonusers on most measures, although they did show a marginally higher total score on the six subscales of the Eating Disorders Inventory, and, most importantly, they showed more muscle dysmorphia. Furthermore, a distinction was made among AAS users between short-term AAS “experimenters”, i.e., those who reported lifetime use of AASs for 2-5 months, and long-term AAS “heavy users”, i.e., those who reported lifetime AAS use for 6-150 months. The former group was almost indistinguishable from nonusers, but the latter showed significant differences from nonusers on many measures, including stronger endorsement of conventional male roles and marked symptoms of muscle dysmorphia. Thus, both pathology of body image and narrow views of masculinity seam to be frequent among men with long-term AAS use and could contribute to the onset of AAS use disorders. A study by Goldfield et al . [ 108 ] compared 27 competitive male bodybuilders, 25 recreational male bodybuilders, and 22 men with bulimia nervosa on a wide variety of eating attitudes and disorders, body image, weight and shape preoccupation, weight loss practices, and AAS use. Competitive bodybuilders reported more body dissatisfaction, bulimia nervosa, binge eating, and AAS use than recreational ones, but less eating-related and general psychopathology than bulimic men. It remains unknown whether men with a history of eating disorders are attracted to competitive bodybuilding or, vice versa , competitive bodybuilding triggers disordered eating and AAS use. Another study by Goldfield and collaborators [ 109 ] assessed body image, eating disorders, general psychopathological characteristics, and AAS use in 20 female competitive bodybuilders compared to 25 female recreational weighttrainers (controls), all recruited by posting advertisements in local gymnasia. Competitive bodybuilders had a higher incidence of binge eating, excessive concern with body weight or shape, strict dieting, and extreme exercise for weight control compared to recreational weighttrainers. The rate of AAS use was higher in the former group compared to the latter (40% vs. 0%).

Another study compared psychological status and AAS use in 51 male weightlifters, 15 of whom meeting current criteria for muscle dysmorphia, 8 reporting past muscle dysmorphia, and 28 with no history of muscle dysmorphia (controls), recruited through advertisements placed in gymnasia and nutrition stores [ 110 ]. Men with current muscle dysmorphia experienced more aversive symptoms regarding body image, including frequent thoughts about muscularity, appearance dissatisfaction, body checking, bodybuilding dependence, and functional impairment than controls. Men with past muscle dysmorphia were characterized by a higher prevalence of mood and anxiety disorders than controls. No significant difference in AAS use, abuse, or dependence was found among the three groups.

Walker et al . [ 111 ] examined the association between body checking, importance of shape and weight, symptoms of muscle dysmorphia, mood disorders, and use of APEDs, such as AASs, in 550 undergraduate males taking part in college sports. In men, body checking correlated with weight and shape concern, symptoms of muscle dysmorphia, depression, negative affect, and APED use. Overall, 85.8% of men were APED nonusers, 14.2% past or current APED user. Among the latter, 2.9% used AASs, 3.6% illegal ergo/thermogenics (i.e., fat burners), 3.8% nonsteroidal anabolics, and 10% over-the-counter ergo/thermogenics. Moreover, 69% used only one type of APED, 19% two types, 10% three types, and only 1% all four types. It is of interest that when interviewed, most men who had used APEDs reported past use, rather than current use. When asked how many years they planned to use AASs or licit and illicit ergo/thermogenics, even if occasionally, 98% reported that they did not plan to use them, the remaining 12% that they planned to use them for a mean of another 3.5 years. More than a quarter of those who had ever used APEDs (28.4%) reported that they would continue to use them even if it was to be proved beyond any doubt that they caused severe health problems. Approximately one-fourth (26.4%) reported that if they could rapidly meet all of their physical training goals through the use of APEDs, they would be willing to decrease their life duration by an average of 4.8 years. Of all the study variables, only body checking predicted APED use and accounted alone for the largest amount of variance in the Muscle Dysmorphic Disorder Inventory scores (16%). This result was confirmed by the significant difference observed on the Muscle Dysmorphic Disorder Inventory scores of APED users and nonusers ( p <0.001). Overall, the results of all the described studies support the view that body checking is an important construct in male body image, muscle dysmorphia, and body change strategies and is the best predictor of APED use.

AASs and Behavioral Disorders

AAS use can lead to changes in behavior, such as increased aggression, hostility, and unprovoked rage attacks. It has been shown that AAS users have higher levels of alertness, lower tolerance to frustration or poor performance, and loss of impulse control. The typical sudden and exaggerated aggressive AAS-induced response to minimal provocations has been termed as “roid rage” [ 54 , 112 ]. A significant incidence of violent crimes and physical partner abuse during AAS use has also been reported. To explain the role of AASs in the control of aggression in males, the “challenge hypothesis” has been formulated. It was initially suggested to account for testosterone-aggression associations in monogamous birds. Testosterone levels were postulated to increase and reach moderate levels at puberty, hence supporting reproductive physiology and behavior. During sexual arousal and challenges to fertile-age males, testosterone levels would rise further. Moreover, this would facilitate competitive behavior, such as aggression. When males are asked to care for offspring, testosterone levels will decrease. The challenge hypothesis was then extended to humans, requiring some modifications: in fact, testosterone levels in men are associated with different behavioral profiles, in relation to life history strategies involving emphasis on either mating or parenting [ 113 ]. However, since athletes who take AASs generally train hard, it is also possible that changes attributed to AAS use may partly reflect exercising. A triad may exist between behavioral changes, AAS use, and exercise. Weighttraining, bodybuilding, and other practices should be considered as potential confounding factors in studies designed to examine the behavioral effects of AASs [ 114 ].

A case report by Conacher and Workman [ 112 ] studied the association between AAS use and violent crime in a 32-year-old amateur bodybuilder who had been convicted of his wife’s murder. He had no psychiatric or criminal history. Three months before the crime, he had started taking AASs. About four weeks before the crime, he had become irritable, quarreling, and sleepless. Thus, AASs may be involved in personality changes and in the increase of violent behavior. Lefavi et al . [ 115 ] examined AAS-induced psychological alterations in 45 male bodybuilders, in particular, 13 with current AAS use, 14 with past AAS use, and 18 who never used AASs. Bodybuilders completed a psychological-profile questionnaire. AASs were associated with more frequent and intense episodes of anger, with some AAS users reporting episodes of lack of control and violence. No dose-response or duration-response relationships were found and no significant difference was detected in the ways that current or past AAS users expressed their anger in comparison with nonusers. In a prospective study, six male strength athletes, three of whom were AAS users and three were not, were monitored over several months as they underwent normal training and competitions [ 116 ]. They filled out the Profile of Mood States Questionnaire, the Buss-Durkee Hostility Inventory, and the Rosenweig Picture Frustration Test on four occasions: two on-drug periods and two off-drug periods for AAS users and equivalent test periods for AAS nonusers. AAS presence was assessed by gas chromatography and mass spectrometry. While those AASs openly reported by the athletes were confirmed in the on-drug samples and most of the off-drug samples, AAS traces were detectable in some supposed off-drug periods. This could explain why AAS users were always more aggressive compared to nonusers. At any rate, self-rated aggression increased significantly in AAS users during their acknowledged on-drug periods. In particular, multiple AAS use (a practice known as “stacking”) determined severe aggression and hostility; one AAS user admitted to be guilty of attempted murder during a previous on-drug period. Brower et al . [ 117 ] investigated addictive patterns of AAS use and related symptomatology in 49 AAS using male weightlifters through an anonymous, self-administered questionnaire. At least one DSM-III-R symptom of dependence was reported by 94% of the sample; three or more symptoms, consistent with a diagnosis of dependence, were reported by 57%. Dependent users could be distinguished from nondependent ones by higher doses, more cycles, more dissatisfaction with body size, and more aggressive symptoms. Doses and dissatisfaction with body size were the best predictors of dependence. Bahrke et al . [ 4 ] conducted a study of 50 male weightlifters to assess physiological and psychological features accompanying AAS use. Participants were 12 current AAS users, 14 past AAS users, and 24 nonusers. They underwent physical examination, blood chemistry assessment, and urinalysis, were interviewed with a semi-structured interview concerning their physical training and the effects of any substance use, and completed the Profile of Mood States questionnaire and the Buss-Durkee Hostility Inventory. Current and past AAS users reported, concurrently with AAS use, increased aggression and irritability, insomnia or oversleeping, changes in muscle size and muscle strength, faster recovery from workouts and injuries, and loss of interest in sexual activity. No significant differences in mood disturbances and in hostility scores were found among the three groups and in comparison with normative values in the general population.

AAS abuse by a bodybuilder has been reported to lead to psychiatric symptoms and violent outbursts [ 118 ]. Choy and Pope [ 119 ] investigated whether athletes’ increased aggression during AAS use could lead to wife battering. For this purpose, 23 strength athletes with AAS use and 14 without AAS use were assessed with the Dyadic Adjustment and Conflict Tactics Scales. Questions focused on relationship with partner, with AAS users invited to respond about their last AAS use cycle and their last AAS-free period and nonusers required to report on their relationship in the preceding three months. AAS users differed from nonusers for an increased number of fights, verbal aggression episodes, and violent episodes against their wives/girlfriends during their on-drug periods, but not during their off-drug periods. Furthermore, these episodes were significantly more frequent in AAS users during the on-drug, compared to the off-drug periods. These findings point to a serious risk of abuse of AAS users’ partners when the user is in an on-AAS period. These results were confirmed by a subsequent study of the same group of investigators. Bond et al . [ 120 ] carried out a study on 46 male strength athletes to measure the effects of AASs on attentional bias to aggressive cues. They were 16 current AAS users, 16 former AAS users, and 14 nonusers. Testosterone, nandrolone, and oxymetholone were the most commonly consumed AASs during the last cycle and an average of 2-3 AASs were used during each cycle (“stacking”). Participants completed an Aggression Rating Scale to assess current feelings of anger and hostility and were presented with a modified Stroop Color Word Conflict Task containing sets of neutral (e.g., shelves, broom), verbally aggressive (e.g., hostile, ridicule), and physically aggressive (e.g., attack, violence) words. Current AAS users rated themselves more impatient ( p < 0.10) and belligerent ( p < 0.10) than past users and took longer than past users and nonusers to name the colors of all word sets, but there were no significant differences between word sets. Therefore, there were no attentional bias differences among and betwixt the three groups, but current AAS use produced subtle mood changes and slowed performance compared to past AAS users and AAS nonusers. These findings draw attention to the fact that AAS users’ partners may be at risk of serious abuse during the on-drug periods. A study by Perry et al . [ 57 ] attempted to clarify the supposed relationship between supraphysiological doses of AASs and increased aggression, which is believed to be associated with high plasma testosterone levels. Ten weightlifters with AAS use and 18 without were interviewed using the Modified Mania Rating Scale, the Hamilton Depression Rating Scale, the Buss-Durkee Hostility Inventory, the Point Subtraction Aggression Paradigm, and the Personality Disorder Questionnaire. Higher aggression levels were found in AAS users than in nonusers with both self-rated and clinician-rated measures; aggression levels were associated with higher plasma testosterone concentrations. However, the higher prevalence of DSM-IV cluster B Personality Disorders, such as antisocial, borderline, and histrionic ones, in AAS users than in nonusers, as assessed with the Personality Disorder Questionnaire, could have mediated these differences.

Animal models seem to support the relationship between AASs and behavioral disorders. Steensland et al . [ 121 ] examined the effect of the chronic administration of nandrolone on dominant and subordinate male rats in a pair-housed condition and found that during allowed social interactions, dominant nandrolone-pretreated rats had highly aggressive behaviors more often than dominant placebo-treated rats. Furthermore, the probability for highly aggressive behaviors was constantly present for the nandrolone-treated rats throughout the study, whereas it decreased in the placebo-treated group. Nandrolone-treated rats showed less fear in case of potentially threatening events, compared to placebo group. These findings point out to the relatively long-term behavioral alterations due to AAS abuse that have been noted in human beings. A study of male rats tested the hypothesis that behavioral alterations associated with AAS use during the adolescence may be carried over to adulthood even after discontinuing AASs [ 122 ]. Prepubertal rats were treated with five weekly intramuscular injections of 5 mg/kg testosterone propionate vs. vehicle for five weeks. The AAS group presented greater aggressiveness in the pubertal phase and higher levels of horizontal and vertical exploration and anxiety-like behavior in the adult phase than the controls, as shown by aggression, elevated plus maze, and open-field tests. A study with pubertal Syrian hamsters [ 123 ] found aggression and anxiety during early AAS use to predict behavioral responding during withdrawal. The hamsters received daily subcutaneous injections of a “stack” of testosterone cypionate (2 mg/kg), nortestosterone (2 mg/kg), and dihydrotestosterone undecyclate (1 mg/kg), which are three of the most commonly abused AASs, vs. vehicle for one month. Three weeks after the last AAS/vehicle injection, experimental and control animals were tested for anxiety on the elevated plus maze, the dark/light, and the seed finding tests and then examined for differences in serotonin afferent innervation to selected anxiety-related brain areas. To confirm the absence of an aggressive phenotype, on the twentieth day of AAS withdrawal animals were tested for offensive aggression using the resident-intruder test. The study showed that AAS intake elicits aggression in youngsters and AAS discontinuation elicits anxiety.

AASs and Psychosis

In the seventies, the term “steroid psychosis” described a spectrum of psychoses with no precise presentation; symptoms ranged from affective to schizophrenia-like, to organic brain syndrome-related. The most commonly reported were emotional lability, agitation, trouble focusing, increase in the speed of conversation, sensory flooding, sleep changes, auditory and visual hallucinations, intermittent memory impairment, mutism, and delusions. Premorbid personality, history of previous psychopathological disorders, and history of previous steroid psychosis did not clearly increase an individual's risk of developing a psychotic reaction during any given course of AASs [ 124 ]. One of the first papers highlighting the relationship between AASs and psychosis is a case report of an athlete consuming AASs [ 125 ]. The authors temporally related the use of these substances to the development of an acute schizophreniform illness. Recommendation was made to consider this “side-effect” in the differential diagnosis of a schizophrenic episode. A study by Pope Katz [ 53 ] of 41 body-builders and football players taking AASs estimated the prevalence of psychotic symptoms associated to AAS use to be about 12%. Lastly, a 30-year old male athlete who had taken AASs in the last 1.5 years, eight weeks after having received an intramuscular injection of nandrolone developed anxiety, paranoid ideation and other psychotic symptoms leading to a full-blown psychotic syndrome, for which he had to be admitted to an inpatient psychiatric service [ 126 ]. There is dearth of evidence suggesting schizophrenia-like reactions to acute or chronic AAS use, thus the conclusion must be that these reactions may develop on a personal vulnerability basis. An animal study offers some support to this view, as nandrolone pretreatment was found to potentiate amphetamine-mediated aggression in rats [ 121 ]. However, it should be mentioned that there have been early attempts to treat schizophrenia with AASs; in particular, it has been claimed that the anabolic steroid Δ 1 ,17-α-methyl-testosterone ameliorated its symptoms [ 127 ], but another study found no effect [ 128 ].

Throughout their lives, athletes generally consume AASs only for few blocks of time, commonly known as “cycles”, that last 8-16 weeks and are separated by substance-free intervals of months or years [ 35 ]. However, some individuals go on to abuse AASs to improve their athletic performance and muscular appearance, and in 30% of cases they develop dependence [ 129 ]. AAS use becomes abuse when an uncontrolled urge to take AASs arises, even if it proves to be detrimental for health; it becomes dependence when withdrawal symptoms appear upon abrupt discontinuation. Dependence is usually accompanied by tolerance, which fosters the need for dose increase to obtain the same effect. This ensues in habitual, chronic use that renders AASs even more harmful. In AAS dependence, athletes begin to take doses from 10 to 100 times higher than those used in legitimate medical practice (hence the term “supraphysiological”) and often employ a pyramid administration schedule (a practice known as “pyramiding”), progressively increasing doses in a stepwise manner through the first half of a cycle before reducing them symmetrically in the second half, disregarding all safety issues [ 117 ]; they take a combination of two or more different AASs (a practice known as “stacking”), sometimes through different routes of administration, although the combination has not been demonstrated to possess any added desirable effect [ 130 ]; they consume very frequent or lengthy cycles, often despite adverse effects [ 117 , 131 - 132 ]; they use AASs almost continuously for years, reaching an excessive cumulative duration of use [ 130 ].

The psychopathological aspects of AAS dependence are still under investigation. One issue in studying AAS dependence has been the illicit nature of these substances, thus the existing studies, mainly observational, have often found difficulties in verifying their exact nature or amounts taken. Furthermore, many AAS users simultaneously consume other APEDs and dietary supplements that may have psychopathological effects which become indistinguishable from those due to AASs in the context of a complex psychiatric presentation. Moreover, there are no universally accepted appropriate criteria to apply to a psychiatric clinical picture with AAS-related mental symptoms. A “sex steroid hormone dependence disorder” was suggested by Yale psychiatrists [ 133 ], but has still to gain consensus. Over the years, the existing literature has applied DSM-III, DSM-III-R, DSM-IV, and DSM-IV-TR criteria to assess dependence among populations of AAS users. A problem that has arisen is that the standard DSM-IV-TR substance-dependence criteria are difficult to apply to AASs, since they were mainly projected for intoxicating drugs. Even if AAS suspension causes the classic withdrawal syndrome, mediated by neuroendocrine and cortical neurotransmitter systems, AASs are cumulatively acting drugs that produce little or no acute intoxication, hence they do not deliver an immediate reward upon consumption and do not usually impair daily functions like other intoxicating drugs, such as narcotics, stimulants, and hallucinogens. AAS dependence may involve the opioidergic system. In an attempt to address this problem, a group of researchers [ 129 ] suggested a set of diagnostic criteria for AAS dependence, based on the standard substance dependence criteria of DSM-IV-TR, slightly modified and adapted to apply specifically to AASs. The DSM-5 has accepted these suggestions and has recommended, for AAS dependence, the use of the code "other substance use disorder", where the specific substance has to be indicated. This code indicates a mental disorder in which the repeated use of an “other substance” typically continues, despite the individual's knowing that the substance is causing serious problems. The substance cannot be classified within the alcohol, caffeine, cannabis, hallucinogen (phencyclidine and others), inhalant, opioid, sedative, hypnotic, anxiolytic, stimulant, or tobacco categories. DSM-5 diagnostic criteria are shown in Table ​ 2 2 .

DSM-5 diagnostic criteria for AAS dependence [ 49 ]. At least two of the following criteria must be met over a 12-month period..

Evidence of a relationship between AAS use and psychopathology in athletes has been extensively reported in the literature. The first to observe AAS use to enhance athletic performance and its link with psychology has been John Ziegler, the physician for the US men’s weightlifting team, back in the 1950s. He is reported to have stated “What I failed to realize until it was too late was that most of the weightlifters had such obsessive personalities. To them, if two tablets were good, four would be better” [ 134 ]. However, this topic has gained public attention and media exposure in recent years, owing to news of AAS abuse by high-profile athletes in professional associations and Olympic sports [ 13 , 15 ]. The nature of the relationship between AAS use and psychopathology is complex. Individuals who are more vulnerable to the stress associated with improving and maintaining physical functioning, coping with social pressures, and striving for goals, or those with specific “pre-AAS” personality traits, such as antisocial personality disorder and narcissism [ 35 , 135 ], or “pre-AAS” psychopathological disorders, such as somatoform (e.g., body dysmorphic disorder) and eating disorders (e.g., anorexia or bulimia nervosa), are more prone to develop an AAS dependence. A possible explanation is given by motivation to exercise. Although physical activity presents in different forms, most research designed to increase motivation for it, and adherence to it, pay attention to exercise behavior and not to sports participation. When motivation for physical activity is more extrinsic and focused on appearance and weight control (the so-called “self-objectification”), [ 136 ], rather than intrinsic and focused on enjoyment and health improvement, it may facilitate a type of training based on high levels of performance. This may trigger high stress levels and psychopathological symptoms and may herald the onset of AAS use [ 137 - 139 ]. On the other hand, individuals with AAS dependence have been shown to be frequently affected by mood, anxiety, or psychotic disorders, and to present behavioral disturbances some time after AAS use. This may simply reflect the established comorbidity between drug addiction and psychopathological disorders or, alternatively, AASs might cause psychopathological symptoms by determining neuroadaptive changes in the reward neural circuit or by affecting stress vulnerability and neurotrophism that lie in the core of psychopathological disorders [ 140 ].

Thus, the bidirectional comorbidity between AAS use and psychopathology is not so much the expression of a linear relationship, but rather of a circular process with a possible neuroendocrine basis (i.e., central nervous system circuitry and the integrated hypothalamic-pituitary-adrenal axis) [ 141 ], making it difficult to establish a clear cause-effect relationship. AASs act on the central nervous system in several ways. They can influence neuronal function both directly through the modulation of intracellular receptors and indirectly through influencing the function of ligand-gated ion channels and neurotransmitter receptors [ 142 - 145 ]. Furthermore, they can be converted into estrogen derivatives and activate second messenger systems, they may release endogenous opiate peptides, and they can modulate the expression of γ-aminobutyric acid (GABA) inhibitory receptors [ 146 ] and 5-HT 1B and 5-HT 2 serotonin receptors [ 147 ]. These effects involve brain areas associated with depression, stress, anger, and sexual behavior.

The psychopathological complications associated with AAS use, especially manic and mixed symptoms – usually treated with antipsychotic drugs or lithium [ 148 , 149 ], for which a careful therapeutic drug monitoring is recommended [ 150 ] – are a major concern for physicians [ 151 ]. As depression and suicide are other possible complications, a medically supervised detoxification may be useful. Addicted AAS users may possibly benefit from a dose reduction through a tapering course of medically prescribed steroids, as abrupt AAS discontinuation may precipitate severe depression and suicide [ 66 , 73 ]. However, the detoxification of AAS abusers is a poorly studied area, and the prescription of an abused substance to a substance user can be problematic, unless close supervision is provided. Caution in using clonidine for AAS detoxification is recommended, as clonidine itself has been associated with depression. Antidepressants, which have shown promising results during cocaine withdrawal, deserve to receive trials in the treatment of AAS withdrawal.

CONCLUSIONS

AAS use in athletes is associated with mood and anxiety disturbances, as well as reckless behavior, in some predisposed individuals, who are likely to develop various types of psychopathology after long-term exposure to these substances. There is a lack of studies investigating whether the preexistence of psychopathology is likely to induce AAS consumption, but the bulk of available data, combined with animal data, point to the development of specific psycho-pathology, increased aggressiveness, mood destabilization, eating behavior abnormalities, and psychosis after AAS abuse/dependence. The use of AASs in athletes deserves further investigation.

FINANCIAL COMPETING INTERESTS DISCLOSURE

In the past two years, Paolo Girardi has received research support from Lilly, Janssen, and Springer Healthcare, and has participated in Advisory Boards for Lilly, Otsuka, Pfizer, Schering, and Springer Healthcare and received honoraria from Lilly and Springer Healthcare. All other authors of this paper have no relevant affiliations or financial involvement with any organization or entity with a financial interest in, or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

ROLE OF THE FUNDING SOURCE

This study is part of the FIRB project code RBFR12LD0W_002 and has been funded by a grant of the Italian Ministry of Research.

ACKNOWLEDGMENTS

The authors wish to thank Ms Mimma Ariano, Ms Ales Casciaro, Ms Teresa Prioreschi, and Ms Susanna Rospo, Librarians of the Sant’Andrea Hospital, School of Medicine and Psychology, Sapienza University, Rome, for rendering precious bibliographical material accessible, dr. Flavia Napoletano for helping in collecting relevant references, as well as their Secretary Lucilla Martinelli for her assistance during the writing of the manuscript.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• My Bibliography
• Collections
• Citation manager

Save citation to file

Email citation, add to collections.

• Create a new collection
• Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed.

• Search in PubMed
• Search in NLM Catalog
• Add to Search

Anabolic steroids

Affiliation.

• 1 Department of Pharmacology and Cancer Biology, Duke University Medical Center, Durham, North Carolina 27710, USA. [email protected]
• PMID: 12017555
• DOI: 10.1210/rp.57.1.411

The term "anabolic steroids" refers to testosterone derivatives that are used either clinically or by athletes for their anabolic properties. However, scientists have questioned the anabolic effects of testosterone and its derivatives in normal men for decades. Most scientists concluded that anabolic steroids do not increase muscle size or strength in people with normal gonadal function and have discounted positive results as unduly influenced by positive expectations of athletes, inferior experimental design, or poor data analysis. There has been a tremendous disconnect between the conviction of athletes that these drugs are effective and the conviction of scientists that they aren't. In part, this disconnect results from the completely different dose regimens used by scientists to document the correction of deficiency states and by athletes striving to optimize athletic performance. Recently, careful scientific study of suprapharmacologic doses in clinical settings - including aging, human immunodeficiency virus, and other disease states - supports the efficacy of these regimens. However, the mechanism by which these doses act remains unclear. "Anabolism" is defined as any state in which nitrogen is differentially retained in lean body mass, either through stimulation of protein synthesis and/or decreased breakdown of protein anywhere in the body. Testosterone, the main gonadal steroid in males, has marked anabolic effects in addition to its effects on reproduction that are easily observed in developing boys and when hypogonadal men receive testosterone as replacement therapy. However, its efficacy in normal men, as during its use in athletes or in clinical situations in which men are eugonadal, has been debated. A growing literature suggests that use of suprapharmacologic doses can, indeed, be anabolic in certain situations; however, the clear identification of these situations and the mechanism by which anabolic effects occur are unclear. Furthermore, the pharmacology of "anabolism" is in its infancy: no drugs currently available are "purely" anabolic but all possess androgenic properties as well. The present review briefly recapitulates the historic literature about the androgenic/anabolic steroids and describes literature supporting the anabolic activity of these drugs in normal people, focusing on the use of suprapharmacologic doses by athletes and clinicians to achieve anabolic effects in normal humans. We will present the emerging literature that is beginning to explore more specific mechanisms that might mediate the effects of suprapharmacologic regimens. The terms anabolic/androgenic steroids will be used throughout to reflect the combined actions of all drugs that are currently available.

PubMed Disclaimer

Similar articles

• Effects of androgenic-anabolic steroids in athletes. Hartgens F, Kuipers H. Hartgens F, et al. Sports Med. 2004;34(8):513-54. doi: 10.2165/00007256-200434080-00003. Sports Med. 2004. PMID: 15248788 Review.
• Current concepts in anabolic-androgenic steroids. Evans NA. Evans NA. Am J Sports Med. 2004 Mar;32(2):534-42. doi: 10.1177/0363546503262202. Am J Sports Med. 2004. PMID: 14977687 Review.
• Anabolic steroids: a review of their effects on the muscles, of their possible mechanisms of action and of their use in athletics. Celotti F, Negri Cesi P. Celotti F, et al. J Steroid Biochem Mol Biol. 1992 Oct;43(5):469-77. doi: 10.1016/0960-0760(92)90085-w. J Steroid Biochem Mol Biol. 1992. PMID: 1390296 Review.
• Detection of anabolic androgenic steroid use by elite athletes and by members of the general public. Anawalt BD. Anawalt BD. Mol Cell Endocrinol. 2018 Mar 15;464:21-27. doi: 10.1016/j.mce.2017.09.027. Epub 2017 Sep 21. Mol Cell Endocrinol. 2018. PMID: 28943276 Review.
• Position stand on androgen and human growth hormone use. Hoffman JR, Kraemer WJ, Bhasin S, Storer T, Ratamess NA, Haff GG, Willoughby DS, Rogol AD. Hoffman JR, et al. J Strength Cond Res. 2009 Aug;23(5 Suppl):S1-S59. doi: 10.1519/JSC.0b013e31819df2e6. J Strength Cond Res. 2009. PMID: 19620932 Review.
• The dichotomy between health and drug abuse in bodybuilding. Horn J. Horn J. Nordisk Alkohol Nark. 2024 Apr;41(2):212-225. doi: 10.1177/14550725231206011. Epub 2023 Nov 6. Nordisk Alkohol Nark. 2024. PMID: 38645972 Free PMC article.
• Online information and availability of three doping substances (anabolic agents) in sports: role of pharmacies. Garcia JF, Seco-Calvo J, Arribalzaga S, Díez R, Lopez C, Fernandez MN, Garcia JJ, Diez MJ, de la Puente R, Sierra M, Sahagún AM. Garcia JF, et al. Front Pharmacol. 2023 Dec 4;14:1305080. doi: 10.3389/fphar.2023.1305080. eCollection 2023. Front Pharmacol. 2023. PMID: 38111382 Free PMC article.
• Public awareness of side effects of systemic steroids in Asir region, Saudi Arabia. Alhammadi NA, Mohammed Al Oudhah SM, Mofareh Asiri MA, Alshehri MA, Almutairi BAB, Mohammed Abdullah Thalibah A, Asiri FNM, Alshahrani ASA. Alhammadi NA, et al. J Family Med Prim Care. 2023 Sep;12(9):1854-1858. doi: 10.4103/jfmpc.jfmpc_2202_22. Epub 2023 Sep 30. J Family Med Prim Care. 2023. PMID: 38024924 Free PMC article.
• Structural Changes in the Skeletal Muscle of Pigs after Long-Term Administration of Testosterone, Nandrolone and a Combination of the Two. Skoupá K, Bátik A, Št'astný K, Sládek Z. Skoupá K, et al. Animals (Basel). 2023 Jun 28;13(13):2141. doi: 10.3390/ani13132141. Animals (Basel). 2023. PMID: 37443939 Free PMC article.
• Anabolic Steroids in Fattening Food-Producing Animals-A Review. Skoupá K, Šťastný K, Sládek Z. Skoupá K, et al. Animals (Basel). 2022 Aug 18;12(16):2115. doi: 10.3390/ani12162115. Animals (Basel). 2022. PMID: 36009705 Free PMC article. Review.

Publication types

• Search in MeSH

Related information

• PubChem Compound
• PubChem Compound (MeSH Keyword)
• PubChem Substance

LinkOut - more resources

Other literature sources.

• The Lens - Patent Citations
• MedlinePlus Health Information
• Citation Manager

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.