Use of Anabolic-Androgenic Steroids in Adolescence: Winning, Looking Good or Being Bad
by LARS WICHSTROM , WILLY PEDERSEN USE OF anabolic-androgenic steroids (AAS) was first known to the public as a potential means to enhance performance in "power" sports (e.g., field events and weight lifting). The main reason for use is the anabolic effect (Tanner et al., 1995). Although this has been somewhat hard to document experimentally in humans, the general trend is to accept its positive skeletal muscle-enhancing effect (Celotti and Negri-Cesi, 1992). AAS use has numerous adverse effects; among the most serious are increased risk of coronary heart disease, liver disease, testicular atrophy, prostate cancer, and breast enlargement in men and decrease in women (Bahrke et al., 1998). Possible psychological side effects include decreased libido, increased aggression including homicide and suicide, affective and psychotic disorders (Pope and Katz, 1988; Pope et al., 2000; Riem and Hursey, 1995). AAS is suggested to be addictive in some users (Brower et al., 1991). In addition, AAS users often use other illicit drugs and there is the risk for spread of hepatitis and HIV to originally low-risk populations through sharing needles in AAS injection (DuRant et al., 1993b). AAS use in adolescence may cause premature closure of the growth plates over the bones resulting in permanent short stature (Hallagan et al., 1989). The initiation of AAS use has mostly been found to take place sometime during adolescence (Beel et al., 1998; Faigenbaum et al., 1998; Yesalis et al., 1993). The prevention of AAS use should therefore be considered an important task not only within sports but also for the public at large. The search for risk and protective factors for AAS use in adolescence is vital. Rate of lifetime use among United States high-school students varies between 4% and 12% for males and 0.5% and 2% for females (Bahrke et al., 1998). Lower figures have been reported in other countries; rates vary between 1.2% and 3.2% for males and 0.2% and 2.0% for females in Australia (Beel et al., 1998), for example. One report showed a fairly similar rate (12-month) in Canada (4.1% in males and 1.5% in females) (Canadian Centre for Drug-Free Sports, 1993). Prevalence of AAS use among Afrikaans-speaking sport participants has been reported to be in the same range (e.g., 2.5%), whereas much lower rates were found in the general adolescent student population in South Africa (Schwellnus et al., 1992). Studies from different regions in Sweden have provided varying lifetime estimates for adolescents: 5.8% for males and 1.0% for females (Nilsson, 1995) versus 2.1% and 0.2%, respectively (Kindlundh et al., 1999). Substantial regional differences have also been found in the U.S. (DuRant et al., 1995) as well as in South Africa (Lambert et al., 1998). With the exceptions of Canada (e.g., Canadian Centre for Drug-Free Sports, 1993) and the U.S. (e.g., Yesalis et al., 1993), national data are lacking. The current knowledge about AAS use, therefore, needs to be supplemented by large-scale and nationally representative samples from sites outside North America. A policy of unannounced out-of-competition testing, with continuous escort coverage from the moment of notification until delivery of the sample (as opposed to the 48-hour unescorted warning periods given by some countries) is employed by the Norwegian Confederation of Sports (NCS) for participants in all types of sports (Bahr and Tjornhom, 1998). Almost all athletes in Norway, even at the high-school and college levels, are associated with the NCS. By comparison, only 16% of U.S. colleges and universities reported that they had tested for specific performance-enhancing drugs (Fields et al., 1994). It would be of interest to contrast prevalence rates in Norway (particularly among sports participators) with prevalence rates in countries with a less strict testing regimen. Thus, the first aim of this research was to estimate the prevalence of AAS use in Norway according to sports involvement, geographical region and other demographics. AAS research has been mostly descriptive in nature. There have been some attempts at more theoretically driven research, but differing theoretical perspectives have rarely been contrasted. Theories of AAS use have had at least three different viewpoints. First, many studies have taken a sports perspective, and these studies have noted such motives as winning and performing well in athletics (Scarpino et al., 1990). Thus, the prevalence of AAS use among top athletes has been reported to be high in such sports as football, track and field, weight lifting, body building and possibly also self-defense sports and martial arts (VanHelder et al., 1991). High rates of use have been found at subelite levels and among college and high-school athletes (Bahrke et al., 1998; Beel et al., 1998). Second, muscles not only make you win, they fit the present-day body ideal. For men, this body ideal implies upper torso strength and a mesomorph body type. For women, this is defined as slimness, particularly from the waist down, coupled with large breasts (Cohn and Adler, 1992; Wichstrom, 1999). To look good or to be big are the prime motives for AAS use among gym-based weight trainers (Gridley and Hanrahan, 1994) and among the top two motives in the general adolescent population (Buckley et al., 1988; Whitehead et al., 1992). Eating problems and concerns have been found in male body-builders and AAS users (Blouin and Goldfield, 1995; Komoroski and Rickert, 1992; Wroblewska, 1997). Possibly "reverse anorexia" (Pope et al., 1993) and hence AAS use have partly the same etiology as eating disorders, including poor self-concept and poor body satisfaction. However, among normal adolescents one of the prime motives for looking good is to enhance one's chances with the opposite gender, and one could therefore hypothesize that perceived "romantic appeal" is associated with AAS use. Third, although the motive for use of AAS may be to win in sports or to enhance appearance, only a minority of young people actually employs AAS as the means for this common goal. AAS are substances that share many of the same characteristics as other substances that are illegal or socially condemned in many societies. Thus, models of adolescent drug use could be extended to AAS use. According to one such prominent model, Problem Behavior Theory (PBT), substance use during adolescence is part of a larger syndrome of problem behavior (Jessor and Jessor, 1977). PBT argues that delinquent-type behavior, underage drinking, problem drinking, marijuana use, use of other illicit drugs and precocious sexual behavior are all part of the same pattern and constitute a broad class or syndrome of problem behavior. Young people who are heavily involved in one area of problem behavior tend also to be heavily involved in others. It has been suggested that for the majority of adolescents, the motive for being involved in such problem behavior is to gain the adult status that their body is prepared for but which society denies them (Moffitt, 1993). AAS use may be part of such a problem-behavior syndrome. These are substances strongly condemned by society at large that have the potency of actually making the young person look more adultlike. Several pieces of research lend support to viewing AAS use as part of the PBT spectrum. Substance use has repeatedly been found to be prevalent among AAS users (e.g., DuRant et al., 1993a; Kindlundh et al., 1999; Whitehead et al., 1992). Other aspects of problem behavior (e.g., sexual involvement and conduct problems) have not been analyzed systematically, but several case studies have noted violent acts and even homicide among AAS users (Choi and Pope, 1994; Pope and Katz, 1994). Such violence has mostly been interpreted as an effect of the AAS use itself. However, it is a definite possibility that increased antisocial behavior among AAS users is partly due to selection effects. In addition, high-school students who have used AAS played truant more often than nonusers (Kindlundh et al., 1999). Researchers addressing the nature of conduct problems and conduct disorder disagree as to whether this should be treated as a single syndrome or as different dimensions or subgroups (Loeber and Stouthamer-Loeber, 1998). Based on a meta-analysis of dimensional analyses of conduct problems and crime, Loeber and Schmaling (1985) suggested that such behavior could be divided into four different types based on two dimensions: overt-covert (aggressive vs nonaggressive) and destructive-nondestructive (property vs persons). Because aggression has traditionally been singled out as the important dimension in previous AAS research, we ask whether interpersonal aggression (covert aggression) is predictive of AAS use or if other types of conduct problems also play a role. Narratives from AAS users often reveal that they were introduced to or offered the drug by training partners or trainers in gyms (Beel et al., 1998). Although most of these offers may be turned down, those who are offered the drugs are at particular risk for experimenting with them. For preventive purposes it is important to know the characteristics of those who are offered AAS and what predicts use as opposed to rejection of the offered AAS. In summary, the present study had three aims: (1) to estimate the prevalence of AAS use among adolescents in Norway according to sports involvement and demographic factors; (2) to test three perspectives, sports involvement, appearance and eating concerns, and problem behavior theory regarding prediction of AAS use; and (3) to identify predictors of AAS use versus rejection of AAS. Method Participants Students from 67 schools in grades 7-12 (ages 12-20) participated in the first round of data collection (Time 1; T1) of the Young in Norway study during the spring and fall of 1992. Data for the present research are from the second round of data collection (Time 2; T2) that took place in 1994. Details about participants and the procedures at T1 have been given elsewhere (Wichstrom, 1995a; 1999) and only a brief outline will be presented here. Participants at T1 were representative of the Norwegian high-school population of grades 7 to 12. The response rate was 97.0%. As a result of dropouts and courses that take less than 3 years to complete, about 80% of the 18-year-olds were still attending high school. The overall response rate at T2 was 80.1%. In all, 8,508 of the 8,877 participants provided information about AAS use. Mean (SD) age was 17.33 (2.18) (range 14 - 25) and there were slightly more girls (53.8%) than boys. The main categories for the respondents were junior high school (31.9%), senior high school (49.6%), in education after senior high-school graduation (4.2%), full-time employment (4.9%), part-time employment (2.3%), unemployment (1.8%), housewife (0.4%), military service (2.0%), unknown activity (2.9%). Among those who we're still at their original school, 91.8% responded; the figure was 67.9% for those who received a postal questionnaire. Apart from this, the attrition was selective according to a large number of variables. Logistic regression (LR) analysis identified the following predictors of T2 attrition: gender (boy), vocational training, poor grades, conduct problems, low parental SES, few hours spent on homework, low parental monitoring and urbanity. By entering these variables, 65% of the attrition group was classified correctly. Procedure Every student gave his/her consent in writing based on both an oral and written description of the project, formulated according to the standards prescribed by the Norwegian Data Inspectorate. According to these standards, a written informed consent was also obtained from parents of students below the age of 15. The students were instructed to place the completed questionnaires in an envelope and to seal it themselves. A teacher trained by the liaison officer monitored the students in the class during completion. To avoid students influencing each others' responses, all eligible students at each school completed the questionnaire at the same time. Students who had consented to participate but were not present in class during that day completed the questionnaire together on a later occasion. At T2, participants who had left their original school (51.5%) received the questionnaire by mail. Those not responding within 4 weeks were mailed another questionnaire with a reminder letter. The students still at their original school filled out the questionnaire at school according to the same procedure as in 1992. Instruments AAS involvement. Participants were asked whether they had ever used anabolic steroids (doping) (yes/no), whether they had used them during the preceding 12 months (6-point scale) and whether they had ever been offered AAS (yes/no). Follow-up questions were used to find out whether they had used or been offered types of doping other than AAS (yes/no). Involvement in power sports. The adolescents were posed an open-ended question about whether they had competed in or were currently competing in any sports and asked to state the type of sport. They were also asked whether they had been involved in noncompetitive sports. Those indicating weight lifting, bodybuilding, boxing, gymnastics, wrestling or martial arts were grouped as power sports participators. Subjects who currently or previously had competed in sports indicated their highest level of competition (community, county, national or international level). In addition, hours per week spent on training in private gyms were recorded. Perceived athletic: competence was measured using a revised version of the Self Perception Profile for Adolescents (SPPA-R; Harter, 1988; Wichstrom, 1995b) ([Alpha] = 0.82). Appearance and eating problems. Eating problems were measured using a 12-item version of EAT-26 (Garner et al., 1982) developed by Lavik et al. (1991). Several studies suggest favorable reliability and validity for boys and girls (Engelsen, 1999; Engelsen and Hagtvet, 1999a, b); cross-sectional correlates (Wichstrom, 1995a) as well as prospective predictors (Wichstrom, 2000) of EAT-12 do not differ between genders. A 7/8 cut-off point was used. Body mass index (BMI; kg/[cm.sup.2]) was based on self-report. Desired weight change was calculated by subtracting self-reported weight from desired weight (provided the same height). Two measures of physical appearance were included. The Body Areas Satisfaction Scale (BASS; Brown et al., 1989) asks for ratings of satisfaction with specific body parts: face, lower torso, mid torso, upper torso, muscle tone, weight and height ([Alpha] = 0.81). Perceptions of global physical appearance were measured using the Physical Appearance subscale in SPPA-R ([Alpha] = 0.82), and romantic appeal was measured using another subscale in SPPA-R ([Alpha] = 0.75). Problem behavior. There were five measures of problem behavior, including three dimensions of conduct problems (overt destruction, overt nondestruction and covert nondestruction) that were based on previous analyses of the Young in Norway project (Pedersen and Wichstrom, 1995). Frequency of conduct problems during the preceding 12 months was measured on a 6-point scale ranging from "0 times" to "more than 50 times." In order to obtain an approximation of DSM-III-R symptoms of conduct disorder, only participation that was at hypothetically clinically meaningful levels of intensity was included (Wichstrom et al., 1996). A description of the dimensions follows. Overt destruction: stolen from someone in the family (6+ times, 4.9%); stolen goods of value between 100 and 500 Norwegian kroner (NOK; approx. $12) (2+ times, 3.9%); stolen goods of value between 500 NOK and 1,000 NOK (1+ time, 2.1%); minor vandalism (destroyed bus seats, postboxes, telephone booths, etc.; 6+ times, 1.9%); stolen a car or a motorcycle (1+ time, 1.6%); stolen goods of value more than 1,000 NOK (1+ time, 2.3%); and broken in to steal (1+ time, 2.7%). Overt nondestruction: fought with a weapon (1+ time, 2.1%); fought without a weapon (6+ times, 2.8%); threatened someone to obtain money or goods (1+ time, 2.6%); bullied or intimidated others (10+ times, 2.7%); forced someone to have sex (1+ time, 1.1%); or robbed someone (1+ time, 0.8%). Covert nondestruction: refrained from paying on buses, at the cinema, etc. (10+ times, 5.6%); was truant (50+ times, 1.6%); or stayed out a whole night without parental permission (10+ times, only for those aged 17 or younger, 1.1%). Sexual involvement was measured by asking the adolescents if they had ever had sexual intercourse. Drug use measured lifetime use of marijuana (yes/no) and having been offered marijuana (yes/ no). Period prevalence (12 months) of alcohol intoxication, marijuana use, solvent use and use of hard drugs was measured on a 6-point scale ranging from "0 times" to "more than 50 times." Background information included degree of urbanization according to five categories: the capital of Oslo (approximately 500,000 inhabitants), large town (100,000 to 200,000 inhabitants), small town (fewer than 75,000 inhabitants and official status as a town), suburban area (less than 25 km from Oslo or large town) and rural area (not fulfilling any of the previous criteria). Parental socioeconomic status (SES) was measured by classification of mother and father's occupations according to ISCO-88 (International Labour Office, 1990). The country was divided into five regions (East, South, West, Middle and North). Because reports of AAS may be confounded with response style, a brief version of the Marlowe-Crown Social Desirability Scale (Crowne and Marlowe, 1964) developed by Schuessler (1982) on a youth sample was included. Analysis For bivariate analyses, subjects were grouped for AAS involvement into those who had used AAS, those who had been offered AAS but not used them, and those who had never used or been offered AAS. Two logistic regression analyses were conducted. The dependent variables were having been offered AAS and use of AAS, with no offer and no use, respectively, as reference categories. Because no specific theoretical models were introduced, a stepwise procedure with backward elimination according to [Delta] LR was employed. Differences in regression coefficients between the two equations were tested with a t test (Paternoster et al., 1998). Results A total of 72 subjects (0.8%; 95% CI: 0.7%-1.0%) reported that they had ever used AAS and 430 subjects (5.1%; 95% CI: 4.6%-5.5%) reported that they had been offered AAS. Only eight AAS users (11.1%) stated that they had not been offered the drugs. After correcting for attrition by weighting with the results from the logistic regression analysis predicting attrition, only five additional cases of AAS use were estimated. AAS use did not vary according to whether the subjects completed the questionnaire at school or at home, degree of urbanization, geographical region, parental SES, or previous or current sport competition. However, those competing at high levels were slightly more prone to use AAS: community level, 0.5%; county level, 0.9%; national level, 1.3%; international level, 2.5% ([chi square] = 11.0, 3 df, p [is less than] .001). Those offered AAS more often lived in urban areas: capital area, 7.6%; town, 6.2%; small town, 5.6%; suburban area, 3.9%; and rural area, 3.7% ([chi square] = 31.13, 4 df, p [is less than] .001). Current (5.9%) and past (6.9%) sports competitors had been offered AAS more often than those without sports involvement (3.4%) ([chi square] = 28.52, 2 df, p [is less than] .001). Those competing at a high level had received offers of AAS the most: community level, 4.8%; county level, 5.8%; national level, 9.9%; international level, 14.3%, ([chi square] = 44.14, 3 df, p [is less than] .001). AAS use was not associated with a socially desirable response style, whereas those who had been offered AAS scored slightly higher on this measure (t = -3.04, 1/8,536 df, p [is less than] .01). Almost half (n = 31) of the AAS users had used the drug during the preceding 12 months (0.34%; 95% CI: 0.25%-0.50%). Those who had used the drug during the preceding 12 months differed from the past users in several ways. They were more often male (80.6% vs 48.8%; [chi square] = 7.65, 1 df, p [is less than] .01); participated more in power sports (56.7% vs 13.3%; [chi square] = 12.38, 1 df, p [is less than] .001); and were more often involved in overt destruction (53.6% vs 23.7%; 2 = 6.22, 1 df, p [is less than] .05) and overt nondestruction (50.0% vs 17.9%; [chi square] = 7.78, 1 df, p [is less than] .01). They also used illicit drugs more often in the preceding 12 months (marijuana [cannabis]: 58.1% vs 19.5%; [chi square] = 11.37, 1 df, p [is less than] .001; solvents: 32.3% vs 9.8%; [chi square] = 5.71, 1 df, p [is less than] .05); had better self-perceived romantic appeal (score: 3.17 vs 2.85;; t = 2.36, 1/70 df, p [is less than] .05); and trained currently more often in private gyms (2.17 vs 0.80 hours per week; t = 2.79, 1/ 68 df, p [is less than] .01). Entering these significant variables into a logistic regression predicting use within the previous 12 months compared with past use showed that only engagement in power sports (adjusted odds ratio [Adj. OR] = 14.09; 95% CI: 3.16-62.89) and current use of marijuana (Adj. OR = 9.11; 95% CI: 2.22-37.31) were multivariately significant. The AAS users and those who had been offered AAS but had never used them shared many of the same characteristics, and differed from those who were not at all involved in AAS (Table 1). They were predominantly male, about one third were involved in power sports, they more often trained in private gyms and they perceived their athletic competence to be better than did those not involved in AAS. They were less inclined to want to lose weight, they perceived their romantic appeal to be higher and they were more often involved in all types of problem behavior, in particular aggressive behavior and drug use. The AAS users, as opposed to those offered AAS, were less often male and had used illegal substances more often: hard drugs (12 months) 19% vs 5%, respectively; marijuana (lifetime) 58% vs 24% and marijuana (12 months) 36% vs 20%. [TABULAR DATA 1 NOT REPRODUCIBLE IN ASCII] Logistic regression analysis showed that to have been offered AAS was predicted by gender (boy), sexual intercourse, having been offered marijuana, disordered eating, conduct problems of both overt destructive type and overt nondestructive type, and involvement in power sports (Table 2). Use of AAS was predicted by gender, offered marijuana and used marijuana, disordered eating, conduct problems of overt nondestructive type, and involvement in power sports. Thus, AAS involvement was mostly associated with variables within the problem behavior spectrum, although variables from sports involvement (power sports) and appearance (disordered eating) were also included in the final model. Neither demographic variables nor social desirability was multivariately predictive. Significant interaction effects between the covariates and the variable resulting from the logistic regression analysis predicting attrition would indicate that the results were influenced by selective attrition; however, there was no such interaction. The variables included in the logistic regression model of those offered AAS differed little from the model of those who used AAS. The only exceptions were marijuana involvement, which was significantly more predictive of AAS use than of having been offered AAS (p [is less than] .001), and sexual intercourse, which was less predictive of AAS use compared to being offered AAS (p [is less than] .05). TABLE 2. Adjusted odds ratios (adj. OR) for AAS use and AAS offer (no AAS use/no AAS offer, as contrast) | |
Offered AAS Used AAS
Adj. Adj.
Predictors OR 95% CI OR 95% CI
Gender (boy) 3.65 2.68-4.79 1.98 1.02-3.83
Sexual intercourse(*) 2.47 1.80-3.37 1.12 0.57-2.22
Offered marijuana 6.06 4.40-8.32 6.10 2.54-14.61
Used
marijuana([double
dagger]) 0.73 0.52-1.02 3.24 1.68-6.22
Disordered eating 2.10 1.15-3.85 4.07 1.53-10.75
Conduct problems:
Overt destructive 1.67 1.22-2.29 1.59 0.80-3.16
Conduct problems:
Overt nondestructive 2.57 1.85-3.56 2.29 1.12-4.68
Involvement in
power sports 2.23 1.66-2.97 2.90 1.56-5.35
| | Note: Regression coefficient significantly different (*) p < .05; ([double dagger]) p < .001. Discussion The prevalence and correlates of AAS use among Norwegian adolescents were investigated in a large representative sample. The lifetime prevalence of AAS use was 0.8% and 12 months prevalence was 0.3%. Prevalence did not differ according to age, degree of urbanization, geographical region or SES. Three explanations for AAS use were contrasted: body image and eating problems, involvement in power sports and problem behavior. Multivariate analyses showed that AAS use was first and foremost associated with types of problem behavior (i.e., drug [marijuana] involvement and aggressive-type conduct problems). In addition, involvement in power sports and disordered eating were multivariately associated with lifetime use. Recent AAS users were more often current marijuana users than were previous AAS users. A substantial proportion of Norwegian adolescents (5.1%) had been offered AAS, and they differed little from those who had actually used the substances except that the AAS users more often used such illegal substances as marijuana. The prevalence of AAS use among Norwegian adolescents was substantially lower than in most studies of adolescents from Western societies; in the U.S. the range is typically 2.5% to 7% and in Australia, Canada, Sweden and South Africa it is 2% to 3.5%. However, the present figure of 0.8% is close to the following estimates: 1.3% from a national study of visitors to registered gyms (Okstad et al., 1995a), 1.9% among visitors to gyms in a Norwegian county (Bergsgard and Tangen, 1994) and 1.4% among military personnel, mostly consisting of drafted young males (military service is mandatory for Norwegian men) in the age range 17-29 (Okstad et al., 1995b). Therefore, there are reasons to believe that AAS are less of a problem in the general Norwegian adolescent population than in many other Western countries, particularly the U.S. Among adolescent athletes, AAS are more prevalent among participators in strength events (e.g., wrestling, track events and football) that are popular among men in the U.S. In Norway, the top three events are soccer, team handball and cross-country skiing, in that order. Thus, the lesser need for explosive strength among young Norwegian athletes compared to American athletes may explain some of the difference in AAS prevalence. This fact may also explain why there was no difference in AAS use among sports competitors and noncompetitors, as is often (for review, see Bahrke et al., 1998) but certainly not always (DuRant et al., 1995; Tanner at al., 1995; Whitehead et al., 1992) the case among North American samples. Sports competition is the norm for adolescents in Norway (69.9% were present or past competitors in the current sample). Although AAS use, but not AAS dealing, was legal when this study was conducted (1994), doping control has been a topic of considerable public interest. The overwhelming majority of Norwegian athletes are associated with the NCS and therefore risk unannounced doping tests. The number of tests performed in Norway increased during the 1990s, whereas the percentage of positive tests declined compared with earlier periods (Bahr and Tjornhom, 1998). The strong position against AAS taken by agencies associated with sports in Norway may have led to a strong opinion against the use of doping in sports. Indeed, only 5% of Norwegian adolescents report that they agree to the statement "I would in fact like to try anabolic steroids (doping)" (Wichstrom, 1995c). By contrast, in Australia where the prevalence is lower than in the U.S., 10% considered future use (McGufficke et al., 1990). In addition, sports competitors had been offered AAS more frequently than noncompetitors, but they were no more likely to use them. This finding, along with the low prevalence rate, might suggest that AAS have been fought quite successfully in Norwegian sports. To understand individual differences in AAS use, researchers have looked at the issue from three points of view: physical attractiveness and eating problems, participation in power sports and drug use. The present study indicates that although all three perspectives add to the understanding of those who use AAS, it is the illegal substance aspect that explains most of the variance in AAS use. This association has been identified in several studies. However, drug use is correlated with other potential characteristics of AAS users, including disordered eating (Jonas et al., 1987; Zweben, 1987) and living in urban areas (Wichstrom et al., 1996). The present research adds to the understanding of AAS use by demonstrating an association with drug use when such alternate explanations and potential confounders are controlled for. Drug use fits into a wider syndrome of problem behavior; however, the effect of drug use is still strong when other aspects of problem behavior (e.g., sexual involvement and conduct problems) are controlled for. Those offered AAS are at risk of experimenting with the substances. Previous research has identified weight trainers in gyms to be at particular risk, with lifetime prevalence rates of AAS use ranging from 7.7% (Korkia and Stimson, 1993) in the U.K., to 90% (Taylor and Black, 1987) in the U.S. Although most AAS users in the present study had been offered the substances, only 15% of those offered them had ever used them. It has been suggested that personality disorders within Clusters A (odd-eccentric) and B (dramatic-emotional) may be more prevalent among body builders who had used AAS than in body builders who had not used them (Porcerelli and Sandler, 1995). In addition, subclinical or clinically manifest personality disorders may increase the risk of accepting an offer to use AAS. Others, however, have not found the two groups of bodybuilders to differ in these respects (Moss et al., 1992; Perry et al., 1990; Yates et al., 1990). Researchers report that eating concerns and a desire for weight gain differentiate AAS body builders who are AAS users from nonusers (Blouin and Goldfield, 1995) and predict AAS use among high-school students (Komoroski and Rickert, 1992). They suggest that this might play an etiological role. However, these findings were not replicated in a sample of high-school graduates (Drewnowski et al., 1995) nor in a study comparing body builders who use or do not use AAS (Schwerin et al., 1997). The present study suggests that, although disordered eating was multivariately associated with AAS use, such problems might be considered a selection effect due to the fact that those offered AAS had equally distorted eating patterns. Previous studies have indicated that alcohol use is correlated with AAS use (Bahrke et al., 1998). More restrictions are placed on alcohol use in Norway than in most Western countries. For purchase, an age limit of 18 applies, wine and liquor are only sold in a limited number of government-owned outlets and alcohol is heavily taxed. Al though these policies have kept the average consumption at a comparably low level, the explosive drinking culture of Nordic countries prevails. Drinking to intoxication is more common among Norwegian adolescents than in many Mid or Southern European countries (Wichstrom, 1998). Hence, alcohol intoxication may be seen as socially acceptable in adolescence, whereas illicit drug use (e.g., marijuana use) is less prevalent than in many other countries and may be a more marginalized phenomenon. This may offer some explanation as to why alcohol was not multivariately predictive once illicit drug use was controlled. This finding is in keeping with results from Sweden (Kindlundh et al., 1999), as well the state of Georgia (DuRant et al., 1993b). It may thus be the illicit aspect of drug use that is important with respect to AAS. A substantial proportion of AAS users obtains the drugs on the black market (Beel et al., 1998). There is a definite possibility that their AAS supplier also deals other drugs, thereby increasing the risk for AAS users of experimenting with other illegal substances or using them regularly. The previously identified association between AAS and other drugs might merely represent a selection into an AAS-using milieu or of getting offers to use or buy AAS. The present study takes this research forward by demonstrating that AAS users more often use illegal substances than do those who have been offered AAS but refrained from using them. Marijuana possibly functions as a gateway drug for AAS in the Norwegian context, making the choice of experimenting with yet another illegal substance easier. Limitations The present study is cross-sectional and causal conclusions are therefore precluded. Numerous physiological, emotional and behavioral effects of AAS use have been suggested, including irritability, aggression, depression, euphoria, variable libido and aggressive behavior (Riem and Hursey, 1995). Most associations have not been substantiated in studies with adequate control for selection and expectancy biases, but recent studies indicate that AAS supplementation may increase mania and possibly also aggression in some men (Pope et al., 2000). It is thus possible or even likely that some of the observed differences between nonusers and users are the result of the intake of the drug or correlated events (e.g., weight training, AAS expectancies and being in an AAS milieu). That lifetime AAS use was recorded, whereas correlates were measured in a more narrow and recent time frame, adds to the possibility that the present findings represent drug effects. However, the findings also showed that such potential effects of AAS intake as aggressive behavior and eating problems were not multivariately predictive of recent use; neither did they differentiate between AAS users and those who had been offered AAS. This may indicate that the present results were not solely due to the drug. There has been considerable concern about underreporting in self-report studies of AAS, and a minority of users of doping agents may also be misinformed about which substances they have actually used (Thompson et al., 1993). One study found sensitivity and specificity to be 74% and 82%, respectively, in male weightlifters (Ferenchick, 1996) when self-reports were compared to urine samples. Based on the same method, DuRant et al. (1993a) found acceptable positive and negative predictive values of anabolic steroid metabolites in self-reports as well as acceptable test-retest reliability over a 4-month span. In general, adolescents tend to give valid and reliable information about their drug use in anonymous surveys (see, e.g., Rutter et al., 1998), and it is questionable whether self-reports about AAS use are less reliable and valid than, for example, self-reported cocaine or heroin use. Several aspects of the present study may have minimized potential underreporting. The general aim of the survey was concerned with the development and living conditions of adolescents; questions about AAS occupied only a small part, possibly making AAS users more willing to complete the survey than otherwise. In addition, the sample was drawn from an almost complete register of all adolescents in Norway and the participators were assured anonymity by placing the questionnaire in an unmarked envelope which they sealed themselves. The response rate was favorable and allowed for correction for attrition; such correction had minimal impact on the findings as had data collection method (school-based versus postal questionnaire). To ensure that the subjects did not report other doping agents, a follow-up question asked about doping agents other than AAS. It should be noted that other alleged anabolic agents sold over the counter in the U.S. (e.g., Creatine Monohydrate and AAS precursors such as Andronostenedione and Norandrostenediol) had very limited availability at the Norwegian market as shown by other Norwegian studies (Bergsgard and Tangen 1994; Okstad et al., 1995a). However, some subjects could have classified human growth hormone as AAS. A cautionary note on generalizing the present findings to other cultures is warranted. The use of AAS as well as other drugs, in particular marijuana, is less frequent in Norway than in other countries (e.g., the U.S.); therefore, such use may be a more marginalized phenomenon. Marijuana use may thus be more strongly associated with other forms of deviance among Norwegian adolescents than among adolescents in countries with a higher prevalence of its use. Indeed, hard drugs (e.g., cocaine and injected drugs) have been associated with AAS use in the U.S. (DuRant et al., 1993b). However, although the types of behavior associated with AAS may be different in other countries, the conclusion may still hold that AAS use in the general adolescent population is first and foremost another type of problem behavior and secondly related to power sports and appearance. (*) This research has been funded by grants from the Norwegian Research Foundation, the Norwegian Council for Mental Health and the Norwegian Confederation of Sports. References BAHR, R. AND TJORNHOM, M. Prevalence of doping in sports: Doping control in Norway, 1977-1995. Clin. J. Sport Med. 8: 32-37, 1998. BAHRKE, M.S., YESALIS, C.E. AND BROWER, K.J. Anabolic-androgenic steroid abuse and performance-enhancing drugs among adolescents. Child Adolesc. Psychiat. Clin. No. Amer. 7: 821-838, 1998. BELL, A., MAYCOCK, B. AND MCLEAN, N. Current perspectives on anabolic steroids. Drug Alcohol Rev. 17: 87-103, 1998. BERGSGARD, N.A. AND TANGEN, J.O. To Dope or Not to Dope (Rapport 1: En Kartlegging av Dopingbruk i et Idrettsmiljo, Report No. 80), Bo, Norway: Telemarksforskning-Bo, 1994. BLOUIN, A.G. AND GOLDFIELD, G.S. Body image and steroid use in male bodybuilders. Int. J. Eat. Disord. 18: 159-165, 1995. BROWER, K.J., BLOW, F.C., YOUNG, J.P. AND HILL, E.M. Symptoms and correlates of anabolic-androgenic steroid dependence. Brit. J. Addict. 86: 759-768, 1991. BROWN, T.A., CASH, T.F. AND LEWIS, R.J. Body-image disturbances in adolescent female binge-purgers: A brief report of the results of a national survey in the U.S.A.J. Child Psychol. Psychiat. 30: 605-613, 1989. BUCKLEY, W.E., YESALIS, C.E., 3RD, FRIEDL, K.E., ANDERSON, W.A., STREIT, A.L. AND WRIGHT, J.E. Estimated prevalence of anabolic steroid use among male high school seniors. JAMA 260: 3441-3445, 1988. CANADIAN CENTRE FOR DRUG-FREE SPORTS. National School Survey on Drugs and Sport: Final Report, Gloucester, Ontario, Canada: Canadian Centre for Drug-Free Sports, 1993. CELOTTI, F. AND NEGRI-CESI, P. Anabolic steroids: A review of their effects on the muscles, of their possible mechanisms of action and of their use in athletics. J. Steroid Biochem. Mol. Biol. 43: 469-477, 1992. CHOI, P.Y.L. AND POPE, H.G. Violence toward women and illicit androgenic-anabolic steroid use. Ann. Clin. Psychiat. 6: 21-25, 1994. COHN, L.D. AND ADLER, N.E. Female and male perceptions of ideal body shapes: Distorted views among Caucasian college students. Psychol. Women Q. 16: 69-79, 1992. CROWNE, D.P. AND MARLOWE, D. The Approval Motive: Studies in Evaluative Dependence, New York: John Wiley & Sons, 1964. DREWNOWSKI, A., KURTH, C.L. AND KRAHN, D.D. Effects of body image on dieting, exercise, and anabolic steroid use in adolescent males. Int. J. Eat. Disord. 17: 381-386, 1995. DURANT, R.H., ESCOBEDO, L.G. AND HEATH, G.W. Anabolic-steroid use, strength training, and multiple drag use among adolescents in the United States. Pediatrics 96: 23-28, 1995. DURANT, R., RICKERT, V. AND ASHWORTH, C. Validity of adolescents' self-report of anabolic steroid use. New Eng. J. Med. 329: 889, 1993a. DURANT, R.H., RICKERT, V.I., ASHWORTH, C.S., NEWMAN, C. AND SLAVENS, G. Use of multiple drugs among adolescents who use anabolic steroids. New Eng. J. Med. 328: 922-926, 1993b. ENGLESEN, B.K. Multidimensionality in adolescent eating problems. A two-phase measurement study. Eat. Weight Disord. 4: 63-75, 1999. ENGLESEN, B.K. AND HAGTVET, K.A. A generalizability study of the Eating Attitudes Test (EAT-12) in non-clinical adolescents. Eat. Weight Disord. 4: 179-186, 1999a. ENGLESEN, B.K. AND HAGTVET, K.A. The dimensionality of the 12-item version of the Eating Attitudes Test: Confirmatory factor analyses. Sand. J. Psychol. 40: 293-300, 1999b. FAIGENBAUM, A.D., ZAICHKOWSKY, L.D., GARDNER, D.E. AND MICHELI, L.J. Anabolic steroid use by male and female middle school students. Pediatrics 101: E6, 1998. FERENCHICK, G.S. Valdity of self-report in identifying anabolic steroid use among weightlifters. J. Gen. Intern. Med. 9: 554-556, 1996. FIELDS, L., LANGE, W.R., KREITER, N.A. AND FUDALA, P.J. A national survey of drug testing policies for college athletes. Med. Sci. Sports Exer. 26: 682-686, 1994. GARNER, D.M., OLMSTED, M.P., BOHR, Y. AND GARFINKEL, P.E. The Eating Attitudes Test: Psychometric features and clinical correlates. Psychol. Med. 12: 871-878, 1982. GRIDLEY, D.W AND HANRAHAN, S.J. Anabolic-androgenic steroid use among male gymnasium participants: Dependence, knowledge and motives. Sport Hlth 12:11-14, 1994. HALLAGAN, J.B., HALLAGAN, L.F. AND SNYDER, M.B. Anabolic-androgenic steroid use by athletes. New Eng. J. Med. 321: 1042-1045, 1989. HARTER, S. Manual for the Self-Perception Profile for Adolescents, Denver, CO: University of Denver, 1988. INTERNATIONAL LABOUR OFFICE. ISCO-88: International Standard Classification of Occupations, Geneva, Switzerland: International Labour Office, 1990. JESSOR, R. AND JESSOR, S.L. Problem Behavior and Psychosocial Development. A Longitudinal Study of Youth, San Diego, CA: Academic Press, 1977. JONAS, J.M., GOLD, M.S., SWEENEY, D. AND POTTASH, A.L. Eating disorders and cocaine abuse: A survey of 259 cocaine abusers. J. Clin. Psychiat. 48: 47-50, 1987. KINDLUNDH, A.M., ISACSON, D.G., BERGLUND, L. AND NYBERG, F. Factors associated with adolescent use of doping agents: Anabolic-androgenic steroids. Addiction 94: 543-553, 1999. KOMOROSKI, E.M. AND RICKERT. V.I. Adolescent body image and attitudes to anabolic steroid use. Amer. J. Dis. Child. 146: 823-828, 1992. KORKIA, P. AND STIMSON, G.V. Anabolic Steroid Use in Great Britain, London: Centre for Research on Drugs and Health Behaviour, 1993. LAMBERT, M.I., TITLESTAD, S.D. AND SCHWELLNUS, M.P. Prevalence of androgenic-anabolic steroid use in adolescents in two regions of South Africa. So. African Med. J. 88: 876-880, 1998. LAVIK, N.J., CLAUSEN, S.E. AND PEDERSEN, W. Eating behaviour, drug use, psychopathology and parental bonding in adolescents in Norway. Acta Psychiat. Scand. 84: 387-390, 1991. LOEBER, R. AND SCHMALING, K.B. Empirical evidence for overt and covert patterns of antisocial conduct problems: A recta-analysis. J. Abnorm. Child Psychol. 13: 337-353, 1985. LOEBER, R. AND STOUTHAMER-LOEBER, M. Development of juvenile aggression and violence: Some common misconceptions and controversies. Amer. Psychol. 53: 242-259, 1998. MCGUFFICKE, A., ROWLING, L. AND BAILEY, M. Drug use by adolescent athletes. Youth Stud. 9: 47-50, 1990. MOFFITT, T.E. Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychol. Rev. 100: 674-701, 1993. MOSS, H.B., PANZAK, G.L. AND TARTER, R.E. Personality, mood, and psychiatric symptoms among anabolic steroid users. Amer. J. Addict. 1: 315-324, 1992. NILSSON, S. Androgenic anabolic steroid use among male adolescents in Falkenberg. Europ. J. Clin. Pharmacol. 48:9-11, 1995. OKSTAD, A., Sow, E., KONGSVIK, T. AND STORVOLL, E. Dopingbruk i Registrerte Helsestudio: En Undersokelse av Omfang og Variasjoner (Report No. 13, 1995), Nord-Trondelag, Norway: Nord-Trondelagsforskning, 1995a. OKSTAD, A., SOLLI, E., KONGSVIK, T. AND STORVOLL, E. Dopingbruk i Forsvaret: En Undersokelse av Omfang og Variasjoner (Report No. 14, 1995), Nord-Trondelag, Norway: Nord-Trondelagsforskning, 1995b. PATERNOSTER, R., BRAME, R., MAZEROLLE, P. AND PIQUERO, A. Using the correct statistical test for the equality of regression coefficients. Criminology 36: 859-866, 1998. PEDERSEN, W. AND WICHSTROM, L. Patterns of delinquency in Norwegian adolescents. Brit. J. Criminol. 35: 543-562, 1995. PERRY, P.J., YATES, W.R. AND ANDERSEN, K.H. Psychiatric symptoms associated with anabolic steroids: A controlled, retrospective study. Ann. Clin. Psychiat. 2:11-17, 1990. POPE, H.G. AND KATZ, D.L. Affective and psychotic symptoms associated with anabolic steroid use. Amer. J. Psychiat. 145: 487-490, 1988. POPE, H.G. AND KATZ, D.L. Psychiatric and medical effects of anabolic-androgenic steroid use: A controlled study of 160 athletes. Arch. Gen. Psychiat. 51: 375-382, 1994. POPE, H.G., KATZ, D.L. AND HUDSON, J.I. Anorexia nervosa and "reverse anorexia" among 108 male bodybuilders. Comp. Psychiat. 34: 406409, 1993. POPE, H.G., JR., KOURI, E.M. AND HUDSON, J.I. Effects of supraphysiologic doses of testosterone on mood and aggression in normal men: A randomized controlled trial. Arch. Gen. Psychiat. 57: 133-140, 2000. PORCERELLI, J.H. AND SANDLER, B.A. Narcissism and empathy in steroid users. Amer. J. Psychiat. 152: 1672-1674, 1995. RIEM, K.E. AND HURSEY, K.G. Using anabolic-androgenic steroids to enhance physique and performance: Effects on moods and behavior. Clin. Psychol. Rev. 15: 235-256, 1995. RUTTER, M., GILLER, H. AND HAGELL, A. Antisocial Behavior by Young People, New York: Cambridge Univ. Press, 1998. SCARPINO, V., ARRIGO, A., BENZI, G., GARATTINI, S., LA VECCHIA, C., BERNARDI, L.R., SILVESTRINI, G. AND TUCCIMEI, G. Evaluation of prevalence of "doping" among Italian athletes. Lancet 336:1048-1050, 1990. SCHUESSLER, K.F. Measuring Social Life Feelings, San Francisco, CA: Jossey Bass, 1982. SCHWELLNUS, M.P., LAMBERT, M.I., TODD, M.P. AND JURITZ, J.M. Androgenic anabolic steroid use in matric pupils. A survey of prevalence of use in the western Cape. So. African Med. J. 82: 154-158, 1992. SCHWERIN, M.J., CORCORAN, K.J., LAFLEUR, B.J., FISHER, L., PATTERSON, D. AND OLRICH, T. Psychological predictors of anabolic steroid use: An exploratory study. J. Child Adolesc. Subst. Abuse 6: 57-68, 1997. TANNER, S.M., MILLER, D.W. AND ALONGI, C. Anabolic steroid use by adolescents: Prevalence, motives, and knowledge of risks. Clin. J. Sport Med. 5:108-115. 1995. TAYLOR, W.N. AND BLACK, A.B. Pervasive anabolic steroid use among health club athletes. Ann. Sports Med. 3: 155-159, 1987. THOMPSON, P.D., ZMUDA, J.M. AND CATLIN, D.H. Use of anabolic steroids among adolescents. New Eng. J. Med. 329: 888-889, 1993. VANHELDER, W.P., KOFMAN, E. AND TREMBLAY, M.S. Anabolic steroids in sport. Can. J. Sport Sci. 16: 248-257, 1991. WHITEHEAD, R., CHILLAG, S. AND ELLIOT, D. Anabolic steroid use among adolescents in a rural state. J. Faro. Pract. 35: 401-405, 1992. WICHSTROM, L. Social, psychological and physical correlates of eating problems: A study of the general adolescent population in Norway. Psychol. Med. 25: 567-579, 1995a. WICHSTROM, L. Harter's Self-Perception Profile for Adolescents: Reliability, validity, and evaluation of the question format. J. Pers. Assess. 65: 100-116, 1995b. WICHSTROM, L. Hvem? Hva? Hvor? -- Om ungdom og idrett: En landsdekkende surveyundersokelse (Statens trykksakekspedisjon: V0837B). Oslo, Norway: Kulturdepartementet, 1995c. WICHSTROM, L. Alcohol intoxication and school dropout. Drug Alcohol Rev. 17: 413-421, 1998. WICHSTROM, L. The emergence of gender difference in depressed mood during adolescence: The role of intensified gender socialization. Devel. Psychol. 35: 232-245, 1999. WICHSTROM, L. Psychological and behavioral factors unpredictive of disordered eating: A prospective study of the general adolescent population in Norway. Int. J. Eat. Disord. 28: 33-42, 2000. WICHSTROM, L., SKOGEN, K. AND OIA, T. Increased rate of conduct problems in urban areas: What is the mechanism? J. Amer. Acad. Child Adolesc. Psychiat. 35: 471-479, 1996. WROBLEWSKA, A. Androgenic-anabolic steroids and body dysmorphia in young men. J. Psychosom. Res. 42: 225-234, 1997. YATES, W.R., PERRY, P.J. AND ANDERSEN, K.H. Illicit anabolic steroid use: A controlled personality study. Acta Psychiat. Scan& 81: 548-550, 1990. YESALIS, C.E., KENNEDY, N.J., KOPSTEIN, A.N. AND BAHRKE, M.S. Anabolic-androgenic steroid use in the United States. JAMA 270: 12171221, 1993. ZWEBEN, J.E. Eating disorders and substance abuse. J. Psychoact. Drugs 19: 181-192, 1987. Received: February 14, 2000. Revision: June 21, 2000. LARS WICHSTROM, PH.D.,([dagger]) AND WILLY PEDERSON, PH.D.([dagger]) |
|
