The Prevalence of Anabolic Steroid Use by Southern California High School Athletes Final Report Submitted to the LA84 Foundation November 2007 Gary A. Green, MD Clinical Professor, UCLA David Geffen School of Medicine UCLA Olympic Analytical Laboratory 2122 Granville Ave. Los Angeles, CA 90025 [email protected] 2 Table of Contents Executive Summary..............................................................................................3 I. Background.......................................................................................................6 II. Purpose ...........................................................................................................7 III. Methods..........................................................................................................7 a. Sampling Plan..............................................................................................7 b. Questionnaire Development.........................................................................8 c. Human Subject Approval .............................................................................9 d. Survey Administration..................................................................................9 e. Statistical Analysis.....................................................................................10 IV. Results ..........................................................................................................10 V. Discussion.....................................................................................................13 VI. Conclusions...................................................................................................17 VII.Future Directions...........................................................................................18 About the principal investigator:..........................................................................18 Acknowledgements.............................................................................................19 References .........................................................................................................20 Tables.................................................................................................................22 Table I.................................................................................................................22 Table II................................................................................................................23 Table III...............................................................................................................24 Table IV ..............................................................................................................24 Figures................................................................................................................25 Figure 1...............................................................................................................25 Figure 2...............................................................................................................25 Figure 3...............................................................................................................26 Figure 4...............................................................................................................27 Figure 5...............................................................................................................28 Appendix A .........................................................................................................29 Appendix B .........................................................................................................30 Appendix C.........................................................................................................40 Appendix D.........................................................................................................42 Appendix E .........................................................................................................43 3 Executive Summary Introduction The use of performance-enhancing drugs in youth sports has been a central concern of the LA84 Foundation. In 1988, the Foundation (known then as the Amateur Athletic Foundation) hosted a national conference sponsored by the National Forum on Anabolic/Androgenic Steroids. The Foundation, in the same year, published a booklet titled, “Sports Devastated,” that gave young athletes and their parents information about the risks of anabolic steroid use. More than 40,000 copies were distributed to schools. In 1998 the Foundation, recognizing that drug use in youth sports was closely related to doping in adult sport, organized another national conference titled “Doping in Elite Sport.” More recently, information on drug use in youth sports has become an important theme in the Foundation’s coaching education program. Since the late 1980s drug use in high school sports has become an increasingly prominent policy issue. Three states – Florida, New Jersey and Texas – have mandated drug testing programs for high school athletes. The California Legislature as well as other states legislatures and individual school districts have debated the merits of testing and educational programs to combat drug use in sports. To understand more about performance-enhancing drugs in Southern California high school sports, the LA84 Foundation commissioned a study by Dr. Gary Green, Clinical Professor in the UCLA Division of Sports Medicine and researcher at the UCLA Olympic Analytical Laboratory. The study examined the use of anabolic steroids and muscle-building dietary supplements, as well as attitudes about anabolic steroid use and drug testing, among high school athletes in Southern California. Dr. Green developed a questionnaire that athletes completed online. The study, which was approved by the UCLA Human Subject Protection Committee, yielded 252 usable responses from student athletes (53% male, 47% female) in the California Interscholastic Federation’s Southern Section. Respondents represented 11 sports and 12 schools of various sizes, located in Santa Barbara, Ventura, Los Angeles, Orange and Riverside Counties. The responses to the questionnaire were consistent across all demographic groups. 4 Findings Self reports of anabolic steroid use were low. Estimates of use by other athletes were significantly higher. • 1% of athletes reported using anabolic steroids at least once. • 15% believe that some of their teammates used anabolic steroids. • Over 50% estimated that 10% of the athletes at their schools (excluding their own teammates) used anabolic steroids; another 12% believed that a quarter of the athletes, on other teams, at their schools were users. • 70% felt that 10-25% of athletes at opponents’ schools used anabolic steroids; another 15% estimated use by opposing athletes at 50% or higher. Athletes self-reported using muscle-building dietary supplements more than anabolic steroids. • 15% of all athletes reported using a muscle-building supplement at least once. • Nearly a quarter (23%) of the boys reported the use of muscle-building supplements. Athletes’ attitudes towards anabolic steroids were overwhelmingly negative. • 97% disagreed with the statement that it is “okay to use anabolic steroids to get a scholarship.” • 90% disagreed that it is “okay to use anabolic steroids once or twice.” Athletes supported drug testing in sports. • 70% agreed that drug testing would be good for their sports. • 82% felt that drug testing makes sports fairer. • 62% believed drug testing would effectively catch people using anabolic steroids. • 60% thought that drug testing is a deterrent to use. • While only 43% supported drug testing at their schools aimed exclusively at steroids, 59% approved of drug testing for all illegal drugs. • Only 9% would view anabolic steroid testing as a violation of their rights. Discussion One of the most striking results of the study is the discrepancy between athletes’ self-reports of anabolic steroid use and their perception of use by others. Only 1% reported using anabolic steroids, but the general perception among the respondents was that about 10% of athletes at their schools used these 5 substances. Most respondents estimated use by athletes at opposing schools to be in the 10% to 25% range. Like virtually all studies of teenagers’ anabolic steroid use, this study relied on self-reporting. It is quite possible that some athletes who use steroids chose not to participate in the study. It also is possible that some athletes who did participate were unwilling to reveal their use of anabolic steroids. Thus, the actual level of use may be above 1%. Alternatively, the self-reporting may be an accurate reflection of reality, in which case, the perception of use by other athletes is in error. Finally, it may be the case that actual use falls somewhere between the self-reported rate and the estimates of other athletes’ use. While the self-reporting of a 1% usage rate is lower than some previous studies, it is consistent with other studies across several different age groups. For example, over the past 17 years the Monitoring the Future Study of 8th to 12th graders, including both athletes and non-athletes, has consistently demonstrated a self-report rate of 1-2.5% for anabolic steroids. The NCAA surveys collegiate athletes every four years regarding use and abuse habits of various drugs and their self-report rate for anabolic steroids has been about 1% over the past 12 years. Recent drug testing results in college and high school sports also suggest low use of anabolic steroids. The positive rate for anabolic steroids under unannounced NCAA drug testing ranged from 0.5 to 1% between 1996 and 2005. Anabolic steroid testing of 500 high school athletes by the New Jersey State Interscholastic Athletic Association in 2007 returned only one positive result. Whether these low percentages primarily reflect negative attitudes towards anabolic steroids, or the deterrent effect of drug testing is unknown. Another noteworthy finding is the level of support for drug testing among the group. Most athletes had faith in the ability of drug testing to ensure fairness in sport by catching offenders and deterring use. These findings and earlier studies suggest that effectively reducing anabolic steroid use will require a multidisciplinary effort. The most promising approach is a comprehensive one that combines education and drug testing. The educational component should include information about the health effects of these substances and provide training in ethical decision making. 6 I. Background Anabolic steroids, more properly called “anabolic-androgenic steroids” or “AAS” are testosterone or testosterone-like synthetic drugs that have both anabolic (muscle-building) and androgenic (male hormone) properties. Athletes take these drugs for their anabolic effects, but it is usually the androgenic expression that is responsible for their adverse reactions. It is likely that they were first introduced to athletics in the 1950’s in the former Soviet Union and have been present on the sports scene ever since. In his 2004 State of the Union speech, President Bush sounded a clarion call regarding the epidemic of anabolic steroids in the United States. “The use of performance-enhancing drugs like steroids . . . sends the wrong message-that there are shortcuts to accomplishments, and that performance is more important than character. So . . . get tough and get rid of steroids now.” According to national surveys, such as the Monitoring the Future Study1 the use of anabolic steroids among high school students has been steadily increasing. Additional studies have also demonstrated increasing usage patterns in adolescents with rates of AAS use ranging from 1-6%.2 Due to their inherent insecurities and emotional immaturity; teenagers are especially vulnerable to the allure of anabolic steroids. A recent study revealed that 30% of adolescent girls and boys reported thinking frequently about wanting more defined muscles3. Unfortunately adolescents’ physical immaturity may also make them more susceptible to the adverse effects of anabolic steroids. Although the main focus of drug testing has been on elite athletes in the Olympic Games, National Collegiate Athletic Association (NCAA) and professional sports, studies demonstrate that patterns of drug use are often established well before an athlete enrolls in college. The NCAA conducts a quadrennial survey of drug use and abuse habits of college student-athletes that involves more than 20,000 student-athletes nationwide. One of the items surveyed in the 2005 edition was the age at which anabolic steroid users began using these drugs. According to the survey, 55% began using anabolic steroids in either high school or junior high school4. It is apparent from this information that effective programs are needed at the high school level. These findings are similar to Dr. Green’s analysis of the1997 NCAA survey that found 52% of college anabolic steroid users had begun in junior high or high school.5 While it has been reported that anabolic steroid use is prevalent in high schools, there has been debate as to the appropriate population to target for interventions. Although Buckley found that 35% of 12th grade anabolic steroid users were not involved in a school-sponsored sport 6, other studies have determined that athletes have a higher prevalence of anabolic steroid use as compared to the general student population. A survey of Canadian youth ages 11 to 18 found that 7 students involved in sports and sport-related activity were significantly more likely to report having used anabolic steroids than those not involved in such activity.7 A study of collegiate athletes by the Dr. Green and others found that collegiate athletes were more likely to use anabolic steroids than their non-athlete peers.8 Finally, a recent study of high school students revealed that team sports participation was related to using products to enhance muscles and appearance, with participation in football being an even stronger correlation.3 Although national surveys of teens suggest that anabolic steroid use is a significant problem, there is very little firm data on the prevalence of anabolic steroid use in Southern California high school athletes. In order to develop a comprehensive prevention program, it is necessary to understand not only the prevalence of use, but also attitudes that may influence behavior. The LA84 Foundation recognized this as a significant issue facing high school sports and commissioned a study by Dr. Green to investigate this problem. II. Purpose The purpose of the study was to develop and administer a survey to Southern California high school student-athletes to determine the prevalence of anabolic steroid use and attitudes towards performance-enhancing drugs and drug testing. III. Methods a. Sampling Plan This study was begun in the fall of 2006 and designed to be completed by the end of the academic year in June of 2007. California High Schools compete under the aegis of the California Interscholastic Federation (CIF) which is made up of several Sections. It was decided to utilize the Southern Section of the CIF due to its large geographic area that encompasses a diverse population. The cooperation of the leadership of the Southern Section was enlisted and the president of the Southern Section, Dr. Jim Staunton, wrote a letter of support (Appendix A). The Southern Section provided a listing of all 564 schools that make up this Section. A biomathematical statistician was employed to conduct a randomization program for these schools. The first step was that all schools with a general enrollment of 50 or less were excluded. The remaining schools were stratified into small (enrollment 50-999) and large (over 1000) schools. From there, schools were sorted with respect to the second digit of their zip codes yielding four groups: 90, 91, 92, and 93, respectively. This eventually yielded eight groups: 90- small, 90- large, 91- small 91- large, etc. All eligible school were then sorted into the 8 groups and randomly selected. In order to ensure that every student-athlete in the Southern Section had an equal chance of being selected for the study, at least two schools were needed in each group. Furthermore, in order to prevent the smaller schools from being overrepresented, 8 three of the larger schools were selected for each zip code. The only exception to this was the 93- zip code in which only two large schools were selected due to the relative paucity of schools in this geographic area. This led to 19 schools ultimately being randomized for initial inclusion in the study. According to the sampling plan, the initial 19 selected schools would be invited to participate in the study. If a selected school declined to participate, the next school in that category in the randomization process would be invited. Schools would continue to be invited until either the target number of schools in that category was achieved or the study was terminated. b. Questionnaire Development As background to the study, a literature search was conducted in several areas. The first was a review of the literature with respect to drug surveys in high school students, with an emphasis on anabolic steroid research in athletes. Existing surveys that targeted this audience were reviewed and in many cases, communication with study authors was conducted to assess their methods. Questionnaires from validated research studies were obtained and questions reviewed for relevance with respect to meeting the objectives of the study. Permission was obtained from the authors to use their respective questions. Eventually 53 questions regarding the use of anabolic steroids, attitudes towards drug testing and anabolic steroids were compiled into a survey instrument. (Appendix B) A prototype of the survey was developed and a pilot survey was given to approximately 50 male and female intercollegiate athletes at two Division I universities and the respondents were surveyed with respect to any suggestions for improving the survey. Based on this pilot data, changes were made to the final questionnaire. A literature review was also conducted regarding the use of electronic versus paper surveys. The literature confirmed that the administration of electronic surveys had similar results when compared to the traditional paper and pencil surveys9. Furthermore, discussions with high school principals provided assurances that home internet access among today’s high school students was fairly universal. In addition, alternative methods, such as electronic scanning of paper questionnaire were explored with the UCLA Biomathematics Department and it became clear that an electronic survey was the most efficient and accurate method of sampling. A search was then conducted to determine the optimal format for the survey. After extensive research, Survey Monkey.com was selected as a web-based platform to conduct the survey. Their web-based design allows subjects to easily access the survey and answer questions quickly and efficiently with the results easily downloaded into an excel file for statistical analysis. The questionnaire was then formatted into their web site and it was piloted again and found to be very easy to use and required only about five to 10 minutes to complete. The LA84 Foundation (formerly the Amateur Athletic Foundation of Los Angeles) 9 agreed to host the survey on their web site. (http://www.la84foundation.org/3ce/ped_frmst.htm) c. Human Subject Approval In order to conduct this study, approval was necessary from the UCLA Human Subject Protection Committee (HSPC) that oversees the ethical conduct of all research studies at UCLA. An application was submitted. Owing to the fact that information regarding illegal drug use was being collected in the survey, extensive revisions were required and provisions for confidentiality were required which delayed approval. In addition, the UCLA HSPC had never before approved a study with electronic informed consent (as opposed to traditional paper signatures). After a great deal of negotiation, the study was ultimately approved by the Committee on November 1, 2006. As a condition for approval, all participating schools were required to submit written letters agreeing to comply with the UCLA code of ethical research. Each subject who participated in the study was required to provide informed consent with parental assent also required for subjects less than 18 years of age. (Appendix C) d. Survey Administration Once the survey instrument was completed and HSPC approval secured, the process of school recruitment began. Invitation letters to the principals and athletic directors were emailed to the initial 19 schools with follow up emails and telephone calls. If a school declined, the next school in that category was invited. Dr. Green conducted all of the communication with the schools until April of 2007. In March of 2007, a senior UCLA medical student, Michael Chen, was added to the project as a volunteer and approved by the UCLA Human Subject Protection Committee. When he joined the project, all 19 initial schools had been contacted. Mr. Chen was assigned to the task of inviting new schools to participate, as well as assisting in attending recruitment meetings at schools that had already agreed to participate. In addition to Mr. Chen, a research assistant was hired for approximately six weeks in April and May. At that point in the study, approximately nine schools already had been surveyed. However, the number of participants at each school was less than anticipated. Ms. Danya Sarembock was hired, with approval of UCLA Human Subject Protection Committee, to re-contact those schools that had already participated in order to return to those schools to remind students to participate. Once a school agreed to participate in the study, a time was arranged for Dr. Green to visit the school and address the student-athletes at that school. At that meeting, the purpose of the study was explained to the students and a flyer was distributed (see Appendix D) with the details of the survey, the web-site address and the password to enter the survey. Time was allotted for questions. Alternatively, schools could request paper copies of the parental assent form and students could use the school computer lab to take the survey if they had a signed copy of their parent’s assent. Following the initial meeting with the athletes, the school received a $100 donation to their athletic fund. 10 e. Statistical Analysis A statistician with the UCLA Department of Biomathematical Statistics developed statistical methods to analyze the data. Descriptive statistics were tabled for several subgroups, including school size, geographic area, gender, age, ethnicity, participation in team sports, and use of muscle-building dietary supplements. Statistical comparisons between subgroups were made with chi- square tests for percentages and Kruskal-Wallis rank sum tests for ordinally scaled variables. Sets of questions within subject were compared using the Friedman rank test, testing the null hypothesis that each response came from the same distribution. In addition, pairs of questions were compared using the Wilcoxon signed rank test. A p value of 0.05 was considered statistically significant. IV. Results Nineteen schools initially were selected as representative of the Southern Section High Schools of the CIF. All of those schools were contacted and nine eventually agreed to participate in and completed the study. The other 10 declined and additional schools were then recruited until either the category was completed or the 2006-07 school year ended. An additional 50 schools were contacted, and three agreed to participate and eventually completed the survey for a total of 12 schools. The reasons for declining to participate varied. Over 60% did not respond to repeated contacts and were deemed “passive declines” and 20% actively responded and declined to participate, although no firm reason given. The remaining reasons given were: the inability to obtain parental consent (e.g. boarding schools), the campus was moving, or the study required school board approval. Data was analyzed comparing schools that were initially selected for the survey (n=9) versus those that were invited later in the study (n=3). The major statistical difference was that athletes at the latter schools were more likely to have used creatine in the past 12 months and were more likely to have used other muscle- building supplements in the past 30 days, year and lifetime. The latter group also was more likely to be planning on using other muscle-building supplements in the next year. The only other finding was that the latter surveyed schools were more likely to favor drug testing for Olympic and professional athletes as compared to those in the original group. For the eight geographic/school-size categories, one achieved complete participation (93- large schools) and the remaining seven categories each were missing one school. Although the original plan called for the study to include 11 large and eight small schools, the final result was eight large schools and four small ones. The participating schools were distributed over a wide geographical area and were located in Santa Barbara, Ventura, Los Angeles, Orange, and Riverside Counties. The distance between the two furthest schools was 250 miles and the researchers drove a total of 1650 miles during the course of the