Influences of dietary supplement use in South Florida adolescent athletes.

Abstract: Dietary supplement use was assessed in adolescent athletes (n=184) in ten high schools in Broward County, Florida. The Survey to Predict Adolescent Athletes Dietary Supplement Use was used to assess how the supplement group users/non-users demonstrated a relationship with the Theory of Reasoned Action components. Significant supplement group differences existed in gender and grade. Male athletes had a stronger intention to use supplements than did females. The health risks that accompany supplement products show evidence of the need for health educators to understand more about the issue of supplement use in order to establish intervention strategies for this population.

**********

Adolescent athletes, striving to fit in and stand out athletically, are choosing dietary supplements as their means to attain optimum performance (Rubinstein & Federman, 2000). Dietary supplements, often used as ergogenic aids, are frequently thought to be the competitive edge. Gaining the desired edge from these dietary supplement products used as ergogenic aids creates a concern for health educators and those in related fields due to the lack of scientific evidence proving their worth and safety. However, with the lack of scientific support of the efficacy of these products, adolescents are choosing to utilize these products (Rubinstein & Federman).

Research exists assessing the uses of supplements in the general population and some specific to adolescents as well (Perko, 1999). With the increase in the use of dietary supplements, a concern is raised in the medical community and the media about the potentially serious health risks of the mostly untested and unregulated products (Blendon et al., 2001). The possible health risks that accompany many of these products show evidence of the need for health educators to understand the influences of the use of dietary supplements. As health educators, finding a way to determine what will be the most effective method to attain information and approach intervention strategies is vital.

Data have shown that adolescent athletes use supplements at a higher rate than the average for all adolescents (Sobal, J. & Marquart, L., 1994). However, medical evidence suggests only certain subgroups of people actually need dietary supplements. Adolescent use of supplements leads health professionals to wonder if the products are being prescribed or attained by the adolescents without any guidance. Gale et al. (2001) examined patterns of prescribing dietary supplements and found 51% of the patients, who received supplements for special diets, were less than 18 years old. Limited data were available on the nutritional status of the subjects before supplementation occurred. As a result, evaluating whether general practitioners were prescribing these supplements appropriately is difficult (Gale et al.).

The purpose of this study was to examine the subjective norms, attitudes, and intentions of the adolescent athletes' use/nonuse of dietary supplements by utilizing a survey based on the Theory of Reasoned Action (TRA). The researchers wanted to investigate if this questionnaire would be able to predict the strongest influence of supplement use and what the influences of use were while using the TRA as its basis.

The SPAADSU survey was developed based on the Theory of Reasoned Action (TRA), specifically for studying the influences of dietary supplement use in adolescent athletes. The SPAADSU contained two sections. The first section pertained to demographic information, sport participation information, and the use of dietary supplements. The second portion contained questions applicable to the three components of the TRA: attitude, subjective norms, and behavioral intentions. The SPAADSU has been proven to be a reliable and valid instrument, r=.9409 (Perko, 1999). Each individual question was based on a Likert scale (1- strongly disagree to 5- strongly agree) and was in one of the three component categories (attitude, subjective norm and behavioral intention).

METHOD

SUBJECTS

This study consisted of 184 athletes ranging from grades 9-12, ages 14-18 years (M = 129, F = 55) from ten high schools in Broward County, Florida. Broward County's population is over 1.6 million and has a median household income of $41,691. The Broward County School District athletic programs have approximately 12,500 athletes of the total 65,140 high school students in all 25 public high schools. The fall and winter sport athletes of each of the participating schools were asked to participate due to the timing of data collection.

Prior to administering the survey, Florida Atlantic University Institutional Review Board and the Broward County School District Administration granted permission for the study. One week prior to the administration of the survey, the parent/guardian consent forms were sent home to be signed and returned. Athletes were eligible if they had returned the consent form with their parents' signature and had signed the minor assent form. The survey (SPAADSU) was then administered and completed prior to the team practice in a school classroom.

DATA ANALYSIS

SPSS 12.0 software was used for data analysis. An item analysis was completed for each question to analyze the 36-item SPADDSU. A fixed factor analysis was completed to establish the three factors of the survey. Three factors (behavioral intention, attitude, and subjective norm) were set to be extracted from the total instrument when the analysis was done. Reliability analysis was completed to determine internal reliability of each of the survey components individually. In addition, a reliability analysis was done for all survey items or components together. Regression analysis was used to determine which of the theory components, attitude, or subjective norm, was a better predictor of behavioral intention. Both attitude and subjective norm are theorized to be predictors of ones intent to perform a behavior. Behavioral intention acted as the dependant variable, and subjective norm and attitude were predictors. The regression analysis yielded a value to indicate the strongest predictor of ones intent to perform the behavior of future supplement use. In addition to the theory predictors (subjective norm and attitude), age was also used to examine whether the various age categories played a role in the intent one had to perform the supplement use behavior.

Six two-way ANOVAs were completed in order to discover differences between the TRA components which acted as the dependent variable (DV) while supplement group, gender, and grade acted as independent variables. This analysis provided overall differences between the three components of the TRA with regard to supplement use and provided information regarding female and male athletes as well as any differences concerning grade level.

Regarding supplement groupings, the subjects were given an option for supplement use status that best represented them, 1) current user, 2) past frequent user, 3) experimental user, and 4) those who have never tried them. Based on the above, two supplement groups were used (Group A: 1, 2, & 3 (users); Group B: 4 (non-users) for data analysis between the TRA components and the differences in supplement groups with regard to gender and grade. Significance was established at p<.05.

RESULTS CHARACTERISTICS OF PARTICIPANTS

Fifty-five subjects were female (30%) and 129 (70%) were male. Majority of athletes (32%) were 17 years of age. The marginal majority of athletes were in grade twelve (29.3%; Table 1). Table 2 displays school sport participation and the gender breakdown. Majority of participating student athletes were football players. The ethnic majorities of student athletes were 37.5 % white, 19.6% Hispanic, and 18.5% black.

[TABLE 2 OMITTED]

Survey results demonstrated a strong relationship between the athletes' intention to use or not use dietary supplements and the attitude they have regarding supplement use and their subjective norms. In this study, coaches, parents, athletic trainers and physicians were all recognized as those who had influence on the young athletes.

Correlations between all three components were nearly equivalent with one another, indicating a strong correlation (Figure 1). Factor analysis completed for the 36-item survey did not yield the expected three-factor outcome. All survey items combined yielded a .8771 alpha value. Next, a reliability analysis for each individual component yielded alpha values of .931 for behavioral intention, .911 for attitude, and .932 for subjective norm. The regression analysis with behavioral intention acting as the constant indicated marginally that attitude was a stronger predictor of intent to use supplements than was subjective norm (Figure 2). The attitude component yielded a beta coefficient of .48, and t(3, 181) = 8.04, p < .01, while the subjective norm component yielded .40 beta coefficient and t(3, 181) = 6.75, p < .01. Age yielded a beta coefficient, -.09, and t(3, 181) = -1.97, p < .05. Age was not a positive predictor of behavior in the study analysis.

[FIGURES 1-2 OMITTED]

When the effect of gender on behavioral intention was examined using analysis of variance, F(1, 180) = 9.30, p < .01, and for supplement group, F(1, 180) = 11.59, p < .01, females scored (M = 38.44, SD = 10.01) lower on behavioral intention than males (M = 44.76, SD = 10.91). Adolescents who have used supplements or currently use supplements (M = 48.16, SD = 9.02) scored higher than adolescents who have never taken supplements (M = 40.24, SD = 11.00).

The effect of gender on the subjective norm was examined, F(1,180) = 7.01, p < .01. Females (M = 36.98, SD = 9.49) scored lower with regard to others and the need to comply with their desires than the males (M = 40.76, SD = 10.28).

Gender effects on attitudes was also examined, F(1, 180) = 10.02, p < .01 and for supplement group, F(1,180) = 4.94, p < .05. Specifically for gender, females (M = 23.33, SD = 5.93) scored lower on attitude questions than males (M = 27.20, SD = 5.93). Adolescents who currently use or have used supplements (M = 28.97, SD = 5.92) scored higher than those adolescents who have never taken supplements (M = 24.59, SD = 5.93).

Another analysis of variance examined the effect of grade in regard to behavioral intention for the supplement group, F(1,176) = 26.20, p < .01. Adolescent athletes who are current users or past users of supplements (M = 48.16, SD = 9.02) scored higher than those adolescents who have never taken any supplement products (M = 40.24, SD = 11.00).

The effect of grade in regard to the subjective norm was investigated, F(3, 176) = 3.27, p < .05 and for the supplement group, F(1,176) = 10.47, p < .01. Ninth graders (M = 42.10, SD = 9.90) scored higher than twelfth grade adolescents (M = 36.78, SD = 11.32). Adolescents who have used supplements or currently use supplements (M = 42.57, SD = 8.62) scored higher than adolescents who have never taken supplements (M = 38.17, SD = 10.59).

The effect of grade on attitude was studied for the supplement group, F(1,176) = 14.64, p < .01. Adolescents who never used supplements scored lower (M = 24.60, SD = 7.70) than those athletes who are current users or have used supplements in the past (M = 28.97, SD = 5.91).

No significant findings were discovered after examining the TRA components and supplement groups in conjunction with ethnicity or with the relationship between ethnicity and grade. No main effects emerged when testing supplement group differences while the subjective norm component acted as the DV. No significant findings were found when subject age and its relation to the theory components and supplement group differences were examined. Additionally, when examining group differences with grade as the independent variable, with both behavior and attitude as dependant variables, no significant main effects emerged.

A total of 11% of the athletes sampled currently used supplements on a regular basis while, 10% had tried them once or twice, 11% used to previously take them, and 67% had never tried them. A majority of those, who were currently or who were previous supplement users, took them to gain muscle size.

The survey questions on the SPADDSU were specific to the components of the TRA. For the behavioral intention portion, which actually holds the most immediate determinate of a persons intent either to perform or not perform a specific behavior, 37.5% of the athletes strongly agreed/agreed that they would use dietary supplements to improve their sports performance. To the statement, "I would use dietary supplements if my coach gave them to me," 28.8% strongly agreed/agreed, and athletic trainers were also noteworthy with 46.7% of athletes stating they would use dietary supplements if an athletic trainer would give them to him/her. Coaches also ranked highly with 82.6% of athletes strongly agreed/agreed they would ask their coach if dietary supplements are safe, and 75% strongly agreed/agreed they would ask them if they work. Parents were recognized as supports due to 79.3% of athletes answering strongly agree/agree to the question, "I would ask my parent or guardian if dietary supplements are safe," and 74.5% strongly agreed/ agreed they would ask their parent or guardian if dietary supplements work. A significant number of athletes (54.9%) answered they strongly agreed/ agreed they would use dietary supplements that they know work, and 65.7% strongly agreed/agreed they would use dietary supplements that were tested and safe.

The attitude component was the portion of the survey, which corresponds with a person's positive or negative evaluation of the performance of a specific behavior. In this scenario, the behavior was dietary supplement use. In response to the question, "Taking dietary supplements is a safe way for athletes to improve sports performance," 21.7% strongly agreed/ agreed. To the statement, "Taking dietary supplements is a good way to build muscles," 35.3% strongly agreed/agreed. In response to the question, "Taking dietary supplements is safe because they are tested by scientists," 20.1% of athletes strongly agreed/agreed.

The third component, subjective norm, was described as an interaction between a person's perception of how significant others will feel about the individual's participation of a behavior and how they feel they need to comply. To the statement, "My parent or guardian would support my using dietary supplements for better sports performance," 25.5% strongly agreed/ agreed. The same question referring to teammates support for better sports performance yielded 26.6% strongly agreed/agreed. In response to the question, "My teams' athletic trainer would support my using dietary supplements to improve sport performance," 21.2% strongly agreed/agreed. With reference to athletic trainers and their support of dietary supplement use for general health reasons, 40.2% strongly agreed/ agreed. In response to, "Generally speaking, I want to do what my parent or guardian wants me to do," 53.8% strongly agreed/agreed, 48.4% strongly agreed/agreed with reference to their coach, and 57.4% strongly agreed/agreed with reference to their doctor.

DISCUSSION

The Theory of Reasoned Action (TRA) framework for this study yielded behavioral intention was the strongest predictor of those who will be supplement users. This conclusion was not surprising considering Perko et al. (2000) in their investigation of influences of supplement use stated behavior intent continues to be a strong indicator of potential use. The most important determinant of health behaviors is the intention to perform the behavior (Unger et al., 2001).

At the time of survey design, the survey constructs were established through the TRA, and Perko (1999) yielded appropriate questions to fall into each of the three components of the survey. When the factor analysis was used by Perko to develop the instrument, the purpose for its use was different than how it was utilized in this study. Only one factor was extracted in this current study. The analysis continued based on the three established theory components that were constructed for the TRA. This enabled investigators to yield results with differences between theory components. The factor analysis did not conclude the finding of the three factors the investigator had predicted, instead one factor was pulled from the analysis. This conclusion may be due to the strong correlation between the theory components and the difficulty to distinguish between all of them through the survey items.

Dunn et al. (2001) found attitude (.492) as a stronger predictor of supplement use among adolescents than subjective norm (.186) with a sample size of just over 1,600 subjects. An investigation on the health risk behavior of smoking utilizing the TRA as a predictor found the attitude variable was a larger indictor than the subjective norm variable. Indicating the perceived consequences of smoking were more important when compared to the motivation to comply with peer norms (Unger et al., 2001). However, a study that addressed predicting intention to have sex in Latina adolescents yielded opposing results to the above examples and subjective norms positively predicted behavioral intention, not general attitude with both of test groups (Fores et al., 2002). In this study, both attitude and subjective norm scored almost equally on the regression analysis to indicate attitude was marginally a stronger predictor of behavioral intention than was the subjective norm. These findings indicated the investigators were not able to make a strong report in support for the theory with either attitude or subjective norm as a stronger predictor of behavioral intent.

Researchers also used the regression analysis to investigate the relationship age had with the intention to use supplements and no significant findings were found (Figure 2). The general knowledge regarding the theory concept is all components are highly correlated, which in turn one can understand why this theory was based on how the components inter-relate with various behaviors. The small sample size of this current study is thought to be the significant reason why results of the regression analysis yielded the similar beta value for attitude and subjective norm and also for the negative value for age. Consequently, the data did not exhibit a profound example of the TRA theoretical concept.

Regarding the components specifically, the attitude component analysis demonstrated that in both males and females a need exists to further educate the athletes on the perceived beneficial effects of dietary supplements, and more importantly, the safety of these products. Among all participating athletes, 21.7% believed taking dietary supplements was a safe way to improve sports performance. Considering this alone, a need for concern is shown. According to current literature, nutritional supplements are not subject to FDA testing. The safety information of these products is often unknown (Metzel et al., 2001). The attitudes demonstrated by athletes in this study regarding supplement safety and testing questions leads educators to have concern for the lack of adequate testing conducted on dietary supplement products. FDA regulations only prevent manufacturers from making claims about the benefits of supplements to cure disease on its labeling. No requirements exist to state the dangers of the product.

After analyzing the study results, current or past use of dietary supplements was shown to play a role in the emerging main effect of behavior intention differences. Those athletes who were current users or had a history of use have a higher intent to perform the behavior in the future. O'Callaghan et al. (1997) reported past behavior may effectively predict future behavioral control because successful performance of the behavior in the past implies the person has been able to obtain the proper resources or has demonstrated the ability to perform the behavior. The O'Callaghan study results dealing with alcohol consumption and young adults concluded that attitudinal factors played a little role when examining those who had a history of alcohol use. This suggested the strong influence exhibited by past behavior and subjective norm left minimal explanation to be made by the attitude component.

Conclusions in this current study determined that athletes indicated coaches as the strongest determinant of behavioral intention when asked questions regarding their intention to use dietary supplements. Perko, et al. (2000) discussed the expectation of coaches to be a strong determinant of behavior due to the contact a coach has on a daily basis with the athletes. Coaches, if properly educated about the supplement products and the dangers they possibly possess from credible sources, could potentially be a great resource for our young athletes. The study results did exhibit others to be established influences as well.

In general, physicians and athletic trainers would possibly be better prepared to answer questions regarding nutritional information (Perko, et al., 2000). Athletic trainers were strong resources with regard to intent and both male and female athletes as well. Dunn et al. (2001) discussed how trainers were generally less influential than parents and coaches (Marquart & Sobal, 1993; Graves et al., 1991), but while the schools which have full-time athletic trainers, would have a greater opportunity to influence students than those athletes in schools with no athletic trainer. Thus, trainers may be named as significant influences to athletes on supplement survey questionnaires. All of the Broward County schools who participated had fulltime trainers employed, who could support the strong influence of athletes' behavior. Student athletes who spend significant time with their school's athletic trainer may opt to utilize them for questions or concerns regarding personal supplement use or use by teammates or peers.

The physician received the most positive response for being the most influential with regard to subjective norms followed by their parent or guardian. The study revealed those who currently are users or have had a history of use of dietary supplement products were more likely to be influenced by their coach and parents to use supplement products than those who had never tried them. The above resulting influences were for both male and female athletes, but it is important to remember the possible differences in them with regard to supplement use.

Male athletes exhibited a stronger behavioral intention to utilize dietary supplements than their female counterparts in this research. According to Radimer et al. (2000), amino acid supplement users were primarily males. One study in particular found adolescent males tended to receive greater social support regarding exercise and social norms for exercise on exercise behavior (Garcia et al., 1995). Alone, the above could indicate why males had a stronger intention to utilize supplements in this study, which in turn can reflect sport performance.

Findings of the current study relating to gender differences should be considered when examining the subject matter. However, with the significantly smaller total number of female athletes surveyed could create an influenced result, considering the greater number of male athletes who participated (Table 1).

After utilizing the SPAADSU as instrumentation for the research, a conclusion regarding the set up of the questions was found. The project investigator found the items in each of the theory components were all grouped together. A response bias may have occurred due to this survey set up. The monotony of answering the survey questions established for one component in its entirety could have been affected when similar questions were consequently being asked until questions for the next component was reached due to the similarity between the items.

CONCLUSION

The need for further research examining the reasons for supplement use behavior is necessary in order to continue to understand this health risk our young athletes are choosing. More thorough investigations regarding how specific sports influence differences in male and female athletes would be interesting and beneficial for those dealing with particular teams, giving more specific insight into the needs and appropriate intervention for athletes.

The Theory of Reasoned Action (TRA) is a valid predictor of intention one may have to use dietary supplements. However, further studies may benefit from examining past behavior and the influence on one's intention to perform a behavior. One's past behavior would act as an additional component to the TRA. Thus, using the SPAADSU instrument, will provide additional knowledge to assist health educators to determine the best intervention approach for this young athletic population.

REFERENCES

Blendon, R.J., DesRoches, C.M., Benson, J.M., Brodie, M., & Altman, D.E. (2001). American Views on the Use and Regulation of Dietary Supplements. Archives of Internal Medicine, 161, 805-810.

Dunn, M.S., Eddy, J.M., Wang, M.Q., Nagy, S., Perko, M.A., & Bartee, R.T. (2001). The influence of significant norms and intentions, regarding dietary supplement use among adolescent athletes. Adolescence, 36, 583-591.

Fores, E., Tschann, J.M., & Marin, B.V. (2002). Latina adolescents: Predicting intentions to have sex. Adolescence, 37, 559-580.

Gale, CR., Edington, J., Coles, S.J., & Martyn, C.N. (2001). Patterns of prescribing of nutritional supplements in the United Kingdom. Clinical Nutrition, 20, 333-337.

Garcia, A.W., Broda, M.A., Frenn, M., Coviak, C., Pender, N.J., & Ronis, D.L. (1995). Gender and developmental differences in exercise beliefs among youth and prediction of their exercise behavior. Journal of School Health, 65, 213-219.

Graves, K.L., Farthing, M.C., Smith, S.A., & Turchi, J.M. (1991). Nutrition training attitudes, knowledge, recommendations, responsibility and resource utilization of high school coaches and trainers. Journal of the American Dietetic Association, 91, 321-324.

Marquart, L., & Sobal, J. (1993). Beliefs and information sources of high school athletes regarding muscle development. Pediatric Exercise Science, 5, 377-382.

Metzel, J.D., Small, E., Levine, S.R., & Gershel, J.C. (2001). Creatine use among young athletes. Pediatrics, 108, 421-424.

O'Callaghan, F.V., Chant, D.C., Callan,V.J., & Baglioni, A. (1997). Models of alcohol use by young adults: an examination of various attitude-behavior theories. Journal of Studies on Alcohol, 58, 502-507.

Perko, M. (1999). Development of a theory based instrument regarding adolescent athletes and dietary supplements. American Journal of Health Studies, 15, 71-86.

Perko, M., Bartee, T., Dunn, M., Wang, M., & Eddy, J. (2000). Giving new meaning to the term "taking one for the team:" Influences on the use/nonuse of dietary supplements among adolescent athletes. American Journal of Health Studies, 16, 99-106.

Radimer, K.L., Subar, A.F., & Thompson, F.E. (2000). Nonvitamin nonmineral dietary supplements: Issues and findings from NHANES III. Journal of the American Dietetic Association, 100, 447-454.

Rubinstein, M.L., & Federman, D.G. (2000). Sports supplements: Can dietary additives boost athletic performance and potential? Postgraduate Medicine, 108, 103-112.

Sobal, J., & Marquart, L. F. (1994). Vitamin/mineral supplement use among high school athletes. Adolescence, 29, 835-843.

Unger, J.B., Rohrbach, L.A., Howard-Pitney, B., Ritt-Olsen, A., & Mouttapa, M. (2001). Peer influences and susceptibility to smoking among California adolescents. Substance Use Misuse, 36, 551-571.

CHES AREAS

Responsibility I--Assessing Individual and Community Needs for Health Education

Competency A: Obtain health related data about social and cultural environments, growth and development factors, needs, and interests. Sub-competency 1-Select valid sources of information about health needs and interests. Sub-competency 3-Employ or develop appropriate data-gathering instruments. Sub-competency 4-Apply survey techniques to acquire health data. Sub-competency 5-Conduct health-related needs assessment in communities.

Jennah R. Godo, M.S., ATC, B. Sue Graves, Ed.D., and Joseph A. O'Kroy, Ph.D. are affiliated with the Department of Exercise Science and Health Promotion at Florida Atlantic University. Steven Hecht, Ph.D. is affiliated with the Department of Psychology at Florida Atlantic University. Address all correspondence to B. Sue Graves, Ed.D., Department of Exercise Science and Health Promotion, Florida Atlantic University, 2912 College Avenue, #279, Davie, Florida 33314; PHONE: [phone]; FAX: [phone]; E-MAIL: [e-mail].

Table 1. Subject demographics of gender, age and grade with percentage
(%) of total subjects (n), mean and standard deviation (std.dev).

Gender                            %        N      mean     std. dev

  Male                          70.1      129
  Female                        29.9       55
Age (years)                                       15.98        1.19
  14                            12.5       23
  15                            26.1       48
  16                            20.7       38
  17                            32.1       59
  18                             8.7       16
Grade                                             10.68        1.07
   9                            16.8       31
  10                            27.2       50
  11                            26.6       49
  12                            29.3       54