Anabolic-Androgenic Steroid Use: Patterns of Use among a National Sample of Canadian Adolescents and Young Adults

Abstract

Background:

Androgenic-anabolic steroid (AAS) use is relatively common among international recreational and community samples and is associated with several deleterious side effects and health consequences that have not been studied recently among a large Canadian sample. Therefore, the purpose of this study was to assess and describe characteristics of AAS users compared to non-AAS users among a national sample of Canadian adolescents and young adults.

Methods:

Data from 2,774 adolescents and young adults from the Canadian Study of Adolescent Health Behaviors were analyzed. Participants were recruited via social media advertisements and completed the survey online. Survey items measured sociodemographics, AAS use, AAS side effects, motives for AAS use, AAS administration, body satisfaction, other illicit substance use, mental health, and violence involvement.

Results:

Among the overall sample, 1.6% of participants reported lifetime AAS use, with the majority of users being White, heterosexual males. Most users strived to gain weight, with main motives for AAS use including increased muscularity and strength, and improved appearance. Most AAS users were satisfied with their body compared to non-users. Approximately one in five users had AAS dependence and experienced multiple adverse side effects. Few differences between users and non-users in self-reported mental health were found.

Conclusion:

Findings indicate that AAS use among a community sample of Canadian adolescents and young adults is an important health issue, particularly given the relatively high occurrence of AAS dependence and adverse side effects reported by users. Findings support the need for education and intervention efforts from healthcare professionals. Further investigations among Canadian AAS users may provide more insight regarding the unique supports these individuals require.

1. Introduction

Sociocultural body ideals for men and women often influence individual behaviors to strive to achieve these ideals (Ata et al., 2015). Male sociocultural body ideals continue to emphasize a bulky, muscular, and lean physique (Murray et al., 2017). Similarly, female body ideals, once overwhelmingly emphasizing thinness, have recently trended towards curvy, toned, and athletic (Bozsik et al., 2018). In keeping with pressures to adhere to these strict body ideals, and the difficulty in achieving and maintaining them, many adolescents and young adults frequently report body dissatisfaction (Bucchianeri et al., 2013). In some instances, this can lead individuals to rely on the use of appearance- and performing-enhancing drugs and substances (APEDS), such as anabolic-androgenic steroids (AAS), to achieve body ideals. Specifically, empirical research has shown a link between body dissatisfaction and AAS use (Hibbert et al., 2021), muscle dissatisfaction and intentions to use AAS (Griffiths et al., 2017), and drive for muscularity and AAS use (Eik-Nes et al., 2018; Murray et al., 2016), underscoring the strong connection between muscular body ideals and AAS use.

The use of AAS has become common outside sporting and competitive settings. Indeed, most AAS users today are not competitive athletes but recreational users (Sagoe et al., 2014). Although use of AAS without a prescription in Western countries is considered illegal, such drugs are often readily accessible through “underground” means (e.g., at the gym) or from the Internet (Kanayama et al., 2020). The most recent large review by Sagoe and colleagues (2014) estimated an overall global AAS use prevalence of 3.3%, with prevalence differing depending on the country of study. Of AAS users, 98% are male (Pope et al., 2014), as androgenic (masculinizing) properties such as deepening of voice, hair growth, and masculinization of secondary sex characteristics, may deter use among women (Börjesson et al., 2021; Kanayama et al., 2011). However, research on female AAS users has shown that primary motivators for use are body dissatisfaction and attempts to achieve and adhere to specific sociocultural body ideals (Börjesson et al., 2021; Hibbert et al., 2021), which are similar motivators among males. Indeed, extant research has demonstrated that a more muscular, lean, and toned physique is acceptable among women (Rodgers et al., 2018; Tiggemann & Zaccardo, 2018), and among AAS users, women commonly report polysubstance use (Havnes et al., 2020). Despite this, the high prevalence of AAS use among men has resulted in most epidemiological research investigating characteristics of use primarily among men, resulting in a need for more comprehensive research using samples that include multiple gender identities (i.e., cisgender men, cisgender women, and transgender/gender non-confirming [TGNC] individuals).

Several comprehensive studies have been conducted to describe the characteristics of male AAS users (Bonnecaze et al., 2020; Ip et al., 2011; Rowe et al., 2017). The majority of AAS users were recreational exercisers, aligning with the notion that AAS use is more prevalent outside elite athletic settings. AAS users reported unique AAS regimens, AAS dependence complications, research prior to use, and how they acquired AAS given their illicit status. Overall, individuals reported using AAS to increase muscle mass, improve physical appearance, and increase strength (Bonnecaze et al., 2020; Ip et al., 2011; Piatkowski et al., 2020; Piatkowski, White, et al., 2021), which supports the notions of trying to achieve a particular body type and concomitant strength to align with the masculine body ideal (Eik-Nes et al., 2018; Melki et al., 2015; Murray et al., 2016; Zahnow et al., 2018). More recent research among an international sample of over 2,300 men, showed similar findings, including a high proportion of AAS users reporting their engagement based motivations to improve physical appearance, gain muscle, and increase strength (Bonnecaze et al., 2020). However, AAS can have serious physical and psychological consequences. AAS use is correlated with cardiac complications, hypogonadism (testicular atrophy), sexual dysfunction, hormonal imbalances, mood disorders, substance use, aggression and violence, and AAS dependence symptoms (Bonnecaze et al., 2020; Kanayama et al., 2020; Rowe et al., 2017). Indeed, this research on the adverse effects underscores the need for more research and understanding of AAS use, particularly given that characteristics of recreational AAS users may vary depending on the country of investigation and demographic characteristics of the sample.

AAS use has often been studied in the United States and various European countries, yet there is a scarcity of research that has examined use among Canadians. There have been no recent studies (i.e., within the past 10 years) that have examined the prevalence of AAS use among Canadians specifically. Melia et al. in 1996 was the largest most recent study, with a sample of 16,169 students, aged 11 to 18 years old, from across Canada. Overall, the prevalence of using AAS for sports and appearance-based reasons was 2.8% (1.5% sports and 1.3% appearance). McVey et al. (2005) surveyed 1,458 middle school students aged 10 to 14 years old in Canada on whether they were taking supplements (i.e., vitamins, pills, and steroids) to aid with weight loss or to gain muscle. Findings indicated that 8.3% girls and 7.1% boys used supplements to gain muscle, but the study did not distinguish prevalence between types of drugs and supplements, so prevalence of AAS use alone was not reported.

Therefore, a more recent national sample of Canadians describing the patterns of AAS use specifically is warranted, particularly given the relatively high prevalence of weight-gain attempts (likely for muscularity-related purposes) among the Canadian population (Ganson, Nagata, et al., 2021; Minnick et al., 2020). The aim of this study was to examine and describe characteristics of AAS users compared to non-AAS users among a national sample of Canadian adolescents and young adults aged 16 to 30 years. This age range was selected given the trajectory of body dissatisfaction and muscularity-oriented behaviors across adolescence into young adulthood (Bucchianeri et al., 2013; Nagata, Ganson, Griffiths, et al., 2020).

2. Methods

Data from 2,774 adolescent and young adults from the Canadian Study of Adolescent Health Behaviors were analyzed. National advertisements on Instagram and Snapchat from November and December 2021 were used for participant recruitment. No targeting strategy based on user interest was implemented to allow for a broad reach. Eligibility criteria included: anyone living in Canada, between the ages of 16 to 30 years old, and who could understand English. Participants represented all 13 provinces and territories in Canada and completed the survey online using Qualtrics. Participants could voluntarily enter a draw to win one of two Apple iPads or one of 20 Starbucks gift cards. The research ethics board at the host institution approved this study (#41707) and informed written consent (via survey checkbox) was obtained from all participants. Methods and findings from the prior research using the Canadian Study of Adolescent Health Behaviors can be found elsewhere (Ganson, Cunningham, et al., 2022a, 2022b).

2.1. Measures

For a full description of the measures, see Supplement Table 1.

2.1.1. Sociodemographics and descriptive characteristics.

The sociodemographic variables measured included age, height and weight (to calculate body mass index [BMI], kilogram/meters²), current gender identity, race/ethnicity, sexual identity, and highest level of education completed. Relevant descriptive characteristics that were measured included current weight change behaviors, current weight perception, and whether individuals engaged in weight training in the past 12 months.

2.1.2. Androgenic-anabolic steroid use.

Participants were asked whether they engaged in lifetime steroid use or use within the past 12 months. If they indicated “yes”, they received additional questions about whether they received permission for use from a doctor or healthcare professional, how they used AAS (i.e., injection, pill, liquid, or cream), and motives for using AAS.

2.1.3. AAS dependence.

To measure AAS dependence, diagnostic criteria for substance use disorder from the Diagnostic and Statistical Manual of Mental Disorders, 5^th Edition (DSM-5) (American Psychiatric Association, 2013) was adapted, similar to prior research (Ip et al., 2011). Participants indicated “yes” or “no” to whether they experienced dependence symptoms across 13 questions.

2.1.4. AAS use side effects.

Participants were asked, “Since you began using anabolic-androgenic steroids, have you experienced any of the following side effects? Select all that apply.” The options included: abnormal lipids and/or cholesterol levels, acne, clitoral enlargement, edema, gynecomastia, hair loss, heart palpitations or racing heart, increased blood pressure, injection side pain, insomnia or trouble falling or staying asleep, irregular menstrual cycle, mood changes, sexual dysfunction, striae, testicular atrophy, voice chances, or other side effects (Ip et al., 2011).

2.1.5. Body satisfaction.

The body satisfaction items were based on prior research (i.e., Project EAT; Eisenberg et al., 2019; Larson et al., 2013; Miller et al., 2019), assessing satisfaction with a number of aspects of the body. Participants indicated on a scale from 0 to 4 how dissatisfied or satisfied with their height, weight, body shape, waist, hips, thighs, stomach, face, body build, shoulders, muscles, chest, and overall body fat. A higher score indicates greater body dissatisfaction.

2.1.6. Mental health and eating disorder diagnoses.

Participants were asked about lifetime prevalence of a diagnosed mental health disorder, and within the past 12 months, indicating “yes” or “no”. The list of mental disorders and eating disorders included those that appear in the DSM-5 (American Psychiatric Association, 2013) and adapted from prior research (Eisenberg & Lipson, 2022).

2.1.7. Psychoactive drug use and violence history.

Items relating to psychoactive drug use and violence history were adapted from previous community studies (Eisenberg & Lipson, 2022). Psychoactive drug use (i.e., marijuana, cocaine, opioids) assessed any use in the past 12 months. For items related to violence history, participants responded “yes” or “no” to whether they had been raped, sexual assaulted, physically abused, emotionally abused, or perpetrated physical violence in the past 12 months.

2.2. Statistical analysis

The sample designated as AAS users included those who reported lifetime use or use in the past 12 months, as well as users who indicated they had received permission from a doctor or healthcare professional, as they remained AAS users and could have been using the drug illegally (i.e., without a proper prescription for use despite professional oversight [i.e., regular blood tests]). Descriptive statistics using means (M), standard deviations (SD), and frequencies (percentages) were estimated among the sample. Differences between AAS users and non-AAS users were assessed using independent samples t tests for continuous variables and Fisher’s exact tests for categorical variables. Statistical significance was defined as two-sided p < 0.05 and all analyses were conducted using Stata 17.

3. Results

From our sample of 2,774 adolescents and young adults (Mage = 22.9, SD = 3.9), 1.6% (N = 44) report lifetime AAS use and 0.8% (N = 23) reported using AAS in the past 12 months. Of the lifetime users, 18.2% (n = 8) of people received permission from a doctor or health professional (75% of which were men), compared to 26.1% (n = 6) of those who reported AAS use in the past 12 months received permission (50% of which were men). The vast majority of AAS users were men (81.8%), compared to 13.6% of women, and 4.6% of TGNC individuals. There were no significant differences between users and non-users on race/ethnicity. Lifetime AAS users were mainly heterosexual (70.5%) followed by identifying as gay or lesbian (18.2%). Notably, lifetime AAS users were more likely to have completed higher education including college or an undergraduate degree (54.6% versus 42.6% of non-users), or a master’s degree and higher (18.2% versus 11.6% of non-users). Almost half of AAS users reported wanting to gain weight, whereas almost half of non-users wanted to lose weight. No significant differences between AAS users and non-users were noted for engagement in weight training in the past 12 months. The overall estimated mean age of beginning AAS use was 22.1 years (SD = 3.7). See Table 1 for full sample sociodemographics and differences between users and non-users. For full sample sociodemographics and differences between past 12-month AAS users and non-users, see Supplement Table 2.

Table 1.

Sociodemographic characteristics of lifetime AAS users versus non-users in the Canadian Study of Adolescent Health Behaviors.

	AAS non-users (N = 2,730) M (SD) / %	Lifetime AAS users (N = 44) M (SD) / %	p
Age	22.9 (3.9)	26.4 (3.4)	< 0.001
BMI	24.5 (5.1)	26.2 (4.4)	0.046
Gender			< 0.001
Cisgender woman	54.1	13.6
Cisgender man	37.7	81.8
TGNC	8.2	4.6
Race/Ethnicity			0.236
White	62.4	65.9
Black	3.2	0.0
Latino/a	2.3	4.6
East Asian	10.0	4.6
South Asian	7.0	6.8
Middle Eastern	2.3	6.8
Indigenous	1.3	2.3
Other	1.3	2.3
Multi-Racial	10.2	6.8
Sexual Identity			0.052
Heterosexual	58.6	70.5
Gay/Lesbian	7.9	18.2
Bisexual	18.2	4.6
Asexual	2.8	2.3
Pansexual	3.2	2.3
Queer	4.7	2.3
Questioning	3.9	0.0
Other	0.7	0.0
Highest Completed Education			0.007
High School Diploma or Less	44.5	22.7
College or Undergraduate Degree	42.6	54.6
Master’s Degree or Higher	11.6	18.2
Other	1.3	4.6
Current Weight Change Behavior			0.003
Lose weight	47.5	27.3
Stay the same weight	16.2	13.6
Gain weight	21.4	45.5
Not doing anything	14.9	13.6
Weight Perception			0.431
Very underweight	1.5	2.3
Slightly underweight	11.3	4.6
About the right weight	48.9	59.1
Slightly overweight	30.8	29.6
Very overweight	7.6	4.6
Weight Training, Past 12 months			0.122
No	38.9	27.3
Yes	61.1	72.7

Note. p value calculated using independent samples t-tests for continuous variables and Fisher’s exact test for categorial variables. AAS = Anabolic-androgenic steroids; BMI = Body mass index; TGNC = Transgender/gender non- conforming

AAS users were asked to provide details on what form of AAS they used. Lifetime AAS users were most likely to use pills (70.5%), followed by injections (54.6%), liquid (11.4%), and cream (6.8%). AAS users in the past 12 months more commonly reported using injections (60.9%), followed by pills (52.2%), liquid (8.7%), and cream (8.7%). Users were asked to report their motive(s) for using AAS. To increase muscularity (72.7%), increase muscular strength (70.5%), and improve appearance (68.2%) were the most common motives reported. Just under half (45.5%) of AAS users wanted to improve athletic performance, while 34% reported use to appear more masculine, 15.9% reported use due to pressures from peers, 6.8% reported use due to pressures from coaches and to do a job better, and 4.6% for self-defense.

To further understand motives for AAS use, body satisfaction of several body parts/areas were examined including satisfaction with height, weight, shape, waist, hips, thighs, stomach, face, body build, shoulders, muscles, chest, and overall body fat (Table 3). Generally, AAS users reported significantly higher body satisfaction compared to non-users.

Table 3.

AAS dependence symptoms among users.

	%
Recurrent steroid use resulted in failure to fulfill obligations	2.3
Steroid used in situations that were physically hazardous	4.7
Unsuccessful effort to cut down or control use	6.8
Experienced diminished effect despite use of same amount of steroids	9.1
You gave up or reduced important social or occupational activities because of steroids	9.5
Continued to use steroids to relieve or avoid withdrawal symptoms	11.4
Needed increased steroids to achieve desired effect	11.4
Steroid use continues despite interpersonal issues exacerbated by steroids	11.6
Steroid use continued despite recurrent physical or psychological issues caused by steroids	14.0
Larger amounts used or used longer than anticipated	20.5
Experienced at least 2 side effects after stopping use	22.7
Considerable time spent to obtain, use or recover from steroid effects	23.3
Experienced cravings or strong desire to use	27.9
AAS dependence
None (0–1 symptoms)	53.9
Mild (2–3 symptoms)	23.1
Moderate (4–5 symptoms)	18.0
Severe (≥ 6 symptoms)	5.1
AAS = Anabolic-androgenic steroids

Table 3 depicts AAS users’ level of dependence. Roughly 1 in 5 users (20.5%) said they used a larger amount of AAS than intended or used longer than intended. Almost 1 in 4 users (23.6%) noted spending a considerable amount of time to obtain, use, or recover from side effects, and over 1 in 4 (27.9%) experienced cravings or a strong desire to use. About 23% of users also experienced at least two side effects after stopping AAS use. The most common side effects reported among the overall sample included acne (34.1%), mood changes (34.1%), insomnia or trouble falling asleep (29.6%), increased blood pressure (27.3%), hair loss (20.5%), and testicular atrophy (20.5%). This pattern was largely the same among both men and women; however, there were some notable differences based on sex (i.e., irregular menstrual cycle, testicular atrophy). See Table 4 for reported side effects.

Table 4.

AAS use side effects among users.

	Overall	Women	Men
	%	%	%
Abnormal lipids and/or cholesterol levels	27.3	16.7	27.8
Acne	34.1	50.0	30.6
Clitoral enlargement	2.3	0.0	0.0
Edema	11.4	0.0	13.9
Gynecomastia	9.1	0.0	11.1
Hair loss	20.5	16.7	22.2
Heart palpitations or racing heart	13.6	16.7	13.9
Increased blood pressure	27.3	16.7	30.6
Injection site pain	18.2	0.0	22.2
Insomnia or trouble falling or staying asleep	29.6	33.3	30.6
Irregular menstrual cycle	4.6	2.0	0.0
Mood changes	34.1	33.3	33.3
Sexual dysfunction	13.6	16.7	13.9
Striae	18.2	16.7	19.4
Testicular atrophy	20.5	0.0	25.0
Voice changes	9.1	0.0	8.3
Other side effects	11.4	0.0	11.1
No side effects	25.0	33.3	22.2

Note. Given small cell size, stratified analyses were not conducted among transgender/gender non-conforming participants.

AAS = Anabolic-androgenic steroids

Mental health, psychoactive drug use, and violence history were also assessed and contrasted between AAS users and non-users (Table 5). Regarding a history of mental health diagnoses, lifetime AAS users were more likely to report a history of psychosis compared to non-users (4.6% versus 0.7%). For eating disorders, only pica was significantly higher among AAS users compared to non-users. Regarding psychoactive drug use within the past 12 months, use of marijuana (82.6% versus 55.5%), cocaine (26.1% versus 7.9%), other stimulants (i.e., Ritalin, Adderall; 30.4% versus 5.1%), MDMA (17.4% versus 5.3%), and psilocybin (39.1% versus 15.3%) were significantly more prevalent among AAS users compared to non-users. No significant differences were reported for history of violent victimization except for AAS users being more likely to report a history of physical abuse (21.7%) than non-users (7.5%).

Table 5.

Mental health, psychoactive drug use, and violence history comparisons between AAS users and non-users.

	Lifetime mental health diagnosis			Mental health diagnosis, past 12 months
	AAS usersa	Non-users	p	AAS usersb	Non-users	p
Mental health diagnosisc
Anxiety	27.3	35.5	0.340	17.4	19.1	1.000
Bipolar	4.6	2.2	0.252	4.4	1.1	0.229
Depression	25	30.3	0.511	13.0	15.7	1.000
Neurodevelopmental disorder or intellectual disability	15.9	13.5	0.655	0.0	7.4	0.407
OCD	4.6	6.6	1.000	4.4	3.3	0.545
Personality disorder	4.6	2.5	0.311	0.0	1.2	1.000
Psychosis	4.6	0.7	0.039	0.0	0.3	1.000
Substance use disorder	4.6	1.9	0.211	4.4	1.0	0.209
Trauma and stressor related disorders	6.8	8.3	1.000	0.0	4.7	0.623
Don’t know	11.4	5.4	0.088	0.0	4.9	0.626
	Lifetime eating disorder diagnosis			Current eating disorder diagnosis
Eating disorder diagnosisc	AAS usersa	Non-users	p	AAS usersb	Non-users	p
Anorexia	2.3	7.6	0.253	0.0	3.1	1.000
AFRID	0.0	1.4	1.000	0.0	0.8	1.000
BED	0.0	1.8	1.000	4.4	1.0	0.202
Bulimia	0.0	2.9	0.637	0.0	1.1	1.000
OSFED	0.0	1.5	1.000	0.0	0.7	1.000
Pica	2.3	0.0	0.016	4.4	0.0	0.008
Rumination	0.0	0.1	1.000	0.0	0.0	1.000
UFED	0.0	0.7	1.000	0.0	0.4	1.000
Not sure	2.3	0.6	0.239	0.0	0.4	1.000
				Psychoactive drug use, past 12 months
Psychoactive drug use	-	-	-	AAS usersb	Non-users	p
Marijuana	-	-	-	82.6	55.5	0.010
Cocaine	-	-	-	26.1	7.9	0.008
Opioids	-	-	-	4.4	2.8	0.482
Benzodiazepines	-	-	-	4.4	2.8	0.482
Methamphetamines	-	-	-	0.0	0.5	1.000
Other stimulants (i.e., Ritalin, Adderall)	-	-	-	30.4	5.1	< 0.001
MDMA	-	-	-	17.4	5.3	0.032
Ketamine	-	-	-	4.4	1.6	0.309
LSD	-	-	-	0.0	4.6	0.623
Psilocybin	-	-	-	39.1	15.3	0.005
Kratom	-	-	-	0.0	0.7	1.000
Other drugs without prescription	-	-	-	0.0	2.4	1.000
				Violence history, past 12 months
Violence history	-	-	-	AAS usersb	Non-users	p
Raped	-	-	-	4.4	3.1	0.521
Sexually assaulted	-	-	-	8.7	8.0	0.705
Physically abused	-	-	-	21.7	7.5	0.027
Emotionally abused	-	-	-	34.8	26.9	0.478
Perpetrated physical violence	-	-	-	4.6	5.0	1.000

Boldface indicates statistical significance at p < 0.05. Differences between groups determined using Fisher’s exact test.

^aLifetime AAS use.

^bAAS use, past 12 months.

^cSelf-reported diagnosis by a health professional.

AAS = Anabolic-androgenic steroids; OCD = Obsessive-compulsive disorder; ARFID = Avoidant/restrictive food intake disorder; OSFED = Other specified feeding and eating disorder; UFED = Unspecified feeding or eating disorder

4. Discussion

The aim of this study was to investigate AAS use and characterize AAS users compared to non-users in a national sample of adolescents and young adults in Canada. Notably, this is the first large-scale study to uniquely report on AAS use within a community sample of adolescents and young adults (aged 16 to 30 years) across all 13 provinces and territories of Canada. The prevalence of lifetime AAS use in our sample was 1.6%, which is similar to the prevalence reported in studies among Western countries. The lifetime prevalence of AAS use for young adults (19–28 years of age) has also remained relatively stable, ranging between 1.1% to 1.9% (Pope et al., 2014). In a review of AAS prevalence globally, North American, Oceanic, and European countries had prevalence of 3.0%, 2.0% and 3.8% respectively (Sagoe et al., 2014). However, these higher numbers may be reflected by varying data collection methods such as including adolescents younger than 16 and adults older than 30 years of age. Kanayama et al. (2011) noted that the prevalence of AAS use rises considerably after adolescence, as AAS use is often initiated during young adulthood (between 20–30 years old), and Pope et al. (2014) further indicated only 22% of users started before age 20, aligning with the mean age of initiation in this study (22 years). However, a review by Sagoe et al. (2014) noted people under 19 years of age had a higher AAS use prevalence than those older than 19 years of age, indicating there are mixed reports as to when AAS use often begins, leading to varying prevalence depending on the population studied.

Our findings continue to emphasize the gendered aspects of AAS use, specifically the unique masculine elements of use. As prior research has documented, AAS use among males is commonly precipitated by and perpetuated by desires to adhere to a muscular and strong body to display adherence to masculine norms (Murray et al., 2016; Piatkowski et al., 2020; Piatkowski, White, et al., 2021). In line with this notion, the majority of AAS users in our sample were men (82%), aligning with reports indicating the vast majority of AAS users are men (e.g., 98%; Pope et al., 2014). AAS use in females is often said to be false-positives as participants may misinterpret what constitutes “steroids” (Kanayama et al., 2007); however, this study clearly indicated AAS for clarity, whereby only 8 women in our sample indicate lifetime AAS use. Given women do not often desire muscular and bulky physiques and may fear masculinization consequences of AAS use (Börjesson et al., 2021; Gruber & Pope, 2000), low prevalence of use among females is to be expected. However, the greater prevalence of AAS use among women in this study compared to others may be further evidence of changes to the ideal female body to one that is more toned and fit (Donovan et al., 2020; Tiggemann & Zaccardo, 2018). We must also note most studies on AAS prevalence rely on the gender binary, whereas we included the option to self-identity as TGNC. However, there was an overall small number of TGNC participants who reported lifetime AAS use in our sample (n = 2), limiting our ability to analyze this group specifically. Therefore, further research should continue to investigate prevalence and characteristics of AAS use among TGNC individuals particularly given that body dissatisfaction and gender dysphoria may uniquely influence body-change behaviors among this population (Nagata et al., 2022; Nagata, Ganson, & Austin, 2020).

Within our sample, the AAS users overall reported having received a higher level of education compared to non-AAS users. Ip et al. (2011) found their sample to be slightly less educated than the general population. However, they noted their users spent over 260 hours researching the topic of AAS prior to using, with strategically planned regimens (Ip et al., 2011). While speculative, it is possible that the higher education level of the AAS users in this study may suggest that individuals have the skills to research AAS use and possible negative consequences, yet determined the benefits outweighed the risks. It is also possible that those with a higher education have the discretionary income to sufficiently cover the extra expense of AAS use.

Given AAS use is illegal in Canada, it is surprising to see individuals receiving permission from doctors and health professionals to use AAS. However, the term “health professional” was not defined nor were examples given. Participants could have interpreted the term to include a wide range of professions, such as personal trainers, coaches, or unreliable sources from the Internet and social media (e.g., influencers; Ganson & Rodgers, 2022). Additionally, “permission” was not defined for participants, requiring more research to provide greater detail on the involvement of medical professionals among AAS users. However, prior research has documented that many AAS users do not report their use to a medical professional, in large part due to fears of being judged, having their AAS use documented in their medical record, and fear of legal complications (Bonnecaze et al., 2020; Zahnow et al., 2017). However, there is evidence that users who disclose their AAS use to a medical professional receive diagnostic tests to monitor their physiological functioning via laboratory blood work (Bonnecaze et al., 2020; Ip et al., 2011; Rowe et al., 2017; Zahnow et al., 2017) and many AAS users receive their injecting equipment from health professionals (Jacka et al., 2019). Indeed, many AAS users in one study reported wanting greater medical oversight and information of their AAS use (Jacka et al., 2019). This literature provides evidence for a need for a harm reduction approach to AAS use (Bates et al., 2022; Jacka et al., 2019; Kimergård & McVeigh, 2014).

Our study is one of few that inquired about forms of AAS administration among a community sample of users in Canada. Users in this study most frequently reported using pills (52–70%) to ingest AAS, followed by injections (55–60%), which differs considerably from Ip et al. (2011) who found 91% of their users administered via injection. Other research has shown considerable differences in mode of administration, with injections being most common (Zahnow et al., 2017, 2018). Additionally, 95% of users were found to administer AAS via intramuscular injection in a review by Mȩdraś et al. (2018). Findings from the Global Drug Survey only assessed administration via injection or orally (pills), and found about one third of the sample used injection and another third ingested AAS orally (van de Ven et al., 2020). Despite mixed findings, both injection and oral administration pose health risks, either via needle sharing, though this occurs a relatively low rates among AAS users (Hope et al., 2013), or liver toxicity with pills (Pope et al., 2014) that need to be addressed given that a primary point of healthcare engagement is needle and syringe programs (van de Ven et al., 2020).

Within our sample, the main motives for AAS use included increasing musculature and strength, and improving appearance, as commonly reported in prior research (Bonnecaze et al., 2020; Börjesson et al., 2021; Christiansen et al., 2017; Ip et al., 2011; Rowe et al., 2017). For many, body dissatisfaction peaks during adolescence and young adulthood (Bucchianeri et al., 2013), and research has reported connections between body dissatisfaction and AAS use (Hibbert et al., 2021). For boys and men, the link between body dissatisfaction and AAS use is likely related with muscle dissatisfaction (Griffiths et al., 2017) and greater drive for muscularity (Eik-Nes et al., 2018; Murray et al., 2016). Among girls and women, use of AAS is often connected with bodybuilding involvement and motivated by body dissatisfaction and attempts to increase strength and muscle size (Börjesson et al., 2016, 2021; Goldfield, 2009). Despite this prior research documenting the connection between body dissatisfaction and AAS use, our findings provided contrary evidence. Surprisingly, AAS users reported being more satisfied with almost all body parts compared to non-users. The higher body satisfaction in AAS users could be due to the significant effectiveness of AAS toward improving muscularity and body build, which were primary motivators of use, such that people are now more satisfied with their body than non-users. This is problematic as the increased body satisfaction may perpetuate continued use (Bonnecaze et al., 2020; Griffiths et al., 2016).

Relatedly, roughly half of AAS users were seeking to gain weight (most likely in the form of muscle mass as their motives for use suggest), whereas most non-users were seeking to lose weight. Attempts to gain weight have become particularly common, specifically among boys and men (Ganson, Nagata, et al., 2021; Nagata et al., 2019; Nagata, Ganson, Griffiths, et al., 2020), likely underscoring the pervasiveness of the muscular ideal among this group. Further, mean BMI of AAS users was also significantly higher than non-AAS users, as with prior research (Ip et al., 2011), aligning with their attempts to gain weight, and is likely due to greater muscle mass among AAS users, as BMI does not account for body composition (i.e., lean muscle mass versus fat mass; Ganson et al., 2021).

Despite the effects of using AAS on body satisfaction and appearance-related outcomes, our sample noted considerable side effects, with 75% of the sample reporting at least one side effect. The side effects most commonly identified among AAS users, including acne, mood changes, hair loss, and testicular atrophy (Bonnecaze et al., 2020; Kanayama et al., 2020; Rowe et al., 2017), were also reported within our sample. It should also be noted that there were differences in side effects experienced based on gender (i.e., cisgender men and cisgender women). Specifically, these differences emerged as sex-specific, including women reporting irregular menstrual cycles and men reporting testicular atrophy. Notably, women often report concerns about the side effects of AAS use reducing their femininity and masculinizing their presentation (Börjesson et al., 2021), whereas men appear to also experience significant side effects related to their sex, including testicular atrophy. However, this side effect for men was nearly equivalent (Rowe et al., 2017) or less common than other studies (Bonnecaze et al., 2020).

AAS users in our study did not report experiencing more mental health issues, with the exception of lifetime AAS users being more likely to report a history of psychosis. Prior research has reported mixed findings regarding the psychiatric commodities among AAS users (Gestsdottir et al., 2021; Griffiths et al., 2017; Ip et al., 2011), despite mood changes being a primary side effect in this and other studies (Bonnecaze et al., 2020; Gestsdottir et al., 2021; Ip et al., 2011). Regarding experiences of violent victimization, there were no significant differences between AAS users and non-users, except for AAS users being more likely to report physical abuse in the past 12 months. It may be that individuals who have experienced physical abuse resort to AAS in an attempt to protect against future victimization (Ganson, Murray, Mitchison, et al., 2021). Prior research has also documented inconclusive findings in regards to the relationships between violent victimization and AAS use (Ganson, Murray, Mitchison, et al., 2021; Gruber & Pope, 1999; Hibbert et al., 2021; Ip et al., 2011), which is likely due to sample characteristics (i.e., nationally representative cohorts, bodybuilders) and sampling strategies. It should be noted that determining the prevalence of mental health diagnoses and violent victimization among AAS users is challenging, and differences can vary depending on doses and length of use (Ip et al., 2011; Pope & Brower, 2009).

Just over 23% of AAS users in our sample had moderate or severe dependence (i.e., experiencing 4 or more symptoms), similar to previously reported AAS dependence rates (e.g., (Bonnecaze et al., 2020; Ip et al., 2011; Pope et al., 2014). Generally, studies from the late 1990’s and early 2000’s have noted about 30% of users develop AAS dependence (Pope et al., 2014). Despite documented prevalence, AAS dependence is one of the most understudied dependence disorders (Kanayama et al., 2011). Unlike other substance abuse disorders, there is no immediate gratification or “high” after using AAS, so researchers can only speculate and hypothesize how individuals develop AAS dependence (Kanayama et al., 2020). Finally, regarding use of psychoactive drugs, AAS users, compared to non-users, were significantly more likely to have used marijuana, cocaine, other stimulants (i.e., Ritalin, Adderall), MDMA, and psilocybin in the past 12 months. Prior research has corroborated these findings, underscoring the common nature of polysubstance use among AAS users (Bonnecaze et al., 2020; Gestsdottir et al., 2021; Hibbert et al., 2021; Ip et al., 2011; Piatkowski, Dunn, et al., 2021; Van de Ven et al., 2018; Zahnow et al., 2018).

4.1. Strengths and limitations

Although our findings contribute to further understanding AAS use and characteristics among a large Canadian sample, our study is not without limitations. First, data is from a cross-sectional survey, limiting our ability to determine causality and further explore the long-term impact and characteristics of AAS use. Additionally, the nonprobability sampling method (i.e., social media advertisements) may limit the generalizability of the findings to the Canadian population. Second, despite our large initial sample size, only a small subset of our sample (N = 44) reported a history of AAS use, whereby a larger sample would perhaps allow for further insight and analyses. Third, additional questions about AAS use could have been included, such as types of AAS used or what involvement a health professional had with the participant’s AAS use (i.e., supervision, medical management, etc.). However, due to the length of the survey and to reduce participant burden, we had to limit the inclusion of certain items, which should be included in future research. Additionally, the term “health professional” was not specifically defined for participants, possibly resulting in mixed interpretations. However, similar studies (i.e., the National Longitudinal Study of Adolescent to Adult Health, Healthy Minds Study, etc.) regularly use the term “health professional” in survey items investigating health-related behaviors. Finally, our sample was limited to participants between the ages of 16 and 30, even though AAS use can begin younger and continue into older adulthood. However, we have captured a highly relevant time for AAS initiation and use. Strengths of the study include the use of a large and diverse sample of Canadian adolescents and young adults, representing all 13 provinces and territories in Canada.

4.2. Implications and future directions

This is the first study within the 21^st century to determine prevalence of AAS use and characteristics of users among a national sample of adolescents and young adults in Canada. We, however, encourage future studies in Canada utilizing a nationally representative longitudinal cohort study to capture trends in AAS use that present greater generalizability, and can further investigate longitudinal associations between AAS use and various health and social behaviors and outcomes. Additionally, targeting specific samples where AAS use is more prevalent, such as gyms and recreational sports teams, would provide opportunities to further inquire about characteristics of AAS use.

Our sample noted considerable side effects, levels of dependence, and polysubstance use, which is alarming. Healthcare professionals should be knowledgeable of the side effects of AAS use to effectively counsel and treat symptoms and sequelae. As such, these professionals can provide further education, prevention programming, and medical oversight for AAS users. This harm reduction approach is warranted and often desired by AAS users (Bates et al., 2022; Jacka et al., 2019; Kimergård & McVeigh, 2014). In Canada’s publicly funded healthcare system, primary care providers are often a first point of healthcare contact for many individuals, without any personal financial burden (Allin et al., 2020; Government of Canada, 2019). Within this context, primary care may be a nexus point for medical oversight if providers have the knowledge and training on AAS use. Additionally, providers need to be aware of how judgement, stigma, and discrimination may impact disclosure of AAS use to healthcare providers (Bonnecaze et al., 2020), which is particularly important given the low healthcare utilization among AAS users who are experiencing adverse health effects (Zahnow et al., 2017). Funding for education and training for healthcare professionals from the federal and provisional government in Canada can be crucial to increase knowledge and awareness of AAS use among healthcare professionals. This may result in greater capability and comfort to treat patients who use AAS, as well as reduce stigmatization of AAS use (Atkinson et al., 2021).

5. Conclusion

The aim of our study was to assess AAS use and user characteristics among adolescents and young adults across Canada. Among our sample of over 2,700 participants, we found a lifetime prevalence of 1.6% for AAS use. Notably, while overall body dissatisfaction was lower among AAS users compared to non-users, AAS users reported considerable side effects and levels of AAS dependence. Future research is needed to continue to understand AAS use among Canadian adolescents and young adults, particularly adolescent boys and young men. Healthcare efforts are needed within Canada to begin to appropriately advise and support AAS use among the population.

Table 2.

Body satisfaction comparisons between AAS users and non-users.

	AAS non-users (N = 2,730) M (SD)	AAS users (N = 44) M (SD)	p
Lifetime AAS use
Body satisfaction
Height	2.5 (1.2)	2.6 (1.3)	0.347
Weight	3.5 (1.1)	3.1 (1.3)	0.029
Body shape	3.5 (1.2)	2.8 (1.2)	0.001
Waist	3.3 (1.2)	2.9 (1.3)	0.027
Hips	3.3 (1.2)	2.7 (1.2)	0.003
Thighs	3.2 (1.3)	3.0 (1.3)	0.305
Stomach	3.7 (1.2)	3.3 (1.2)	0.017
Face	2.9 (1.3)	2.1 (1.1)	0.001
Body build	3.2 (1.2)	2.6 (1.3)	0.001
Shoulders	2.8 (1.2)	2.3 (1.1)	0.007
Muscles	3.0 (1.2)	2.6 (1.3)	0.025
Chest	3.1 (1.2)	2.7 (1.3)	0.048
Overall body fat	3.6 (1.2)	3.1 (1.2)	0.007
AAS use, past 12 months
Body satisfaction
Height	2.5 (1.2)	2.7 (1.4)	0.360
Weight	3.5 (1.1)	2.9 (1.3)	0.016
Body shape	3.5 (1.7)	2.5 (1.0)	0.001
Waist	3.3 (1.2)	2.9 (1.1)	0.115
Hips	3.3 (1.2)	2.9 (2.4)	0.159
Thighs	3.2 (1.3)	2.7 (1.1)	0.069
Stomach	3.7 (1.2)	3.3 (1.1)	0.102
Face	2.9 (1.3)	2.2 (1.3)	0.008
Body build	3.2 (1.2)	2.6 (1.1)	0.014
Shoulders	2.8 (1.2)	2.2 (1.1)	0.023
Muscles	3.0 (1.2)	2.3 (1.1)	0.002
Chest	3.1 (1.2)	2.5 (1.1)	0.021
Overall body fat	3.6 (1.2)	2.8 (1.1)	0.002

Boldface indicates statistical significance at p < 0.05. Differences between groups determined using independent samples t-tests. AAS = Anabolic-androgenic steroids

Scores range from 0 to 4, where 0 = very satisfied, 1 = somewhat satisfied, 2 = neither satisfied nor dissatisfied, 3 = somewhat dissatisfied, 4 = very dissatisfied. A higher score indicates greater body dissatisfaction.