Tobacco use among high school athletes and nonathletes: Results of the 1997 Youth Risk Behavior Survey.

Melnick, Merrill J. ; Miller, Kathleen E. ; Sabo, Donald F. 等

INTRODUCTION

Conventional wisdom has it that participation in high school athletic programs positively influences adolescent health-related behavior. Coaches, athletic administrators, the mass media, and the general public often assert that interscholastic athletic participation helps teenagers develop healthy habits while steering them away from tobacco, alcohol, drugs, dangerous dietary practices, physical inactivity, and other detrimental behaviors.

On the other hand, some sport critics focus on a variety of negative health-related behaviors they believe are associated with athletic participation such as binge drinking (Zill, Nord & Loomis, 1995); drug use; on- and off-the-field aggression (Nixon, 1997; Pilz, 1982); the "female triad," namely eating disorders, amenorrhea, and osteoporosis (Taub & Blinde, 1992); actions that result in unintentional injury and death, such as irresponsible automobile, motorcycle, and bicycle use (Baumert, Henderson, & Thompson, 1998); and unprotected sex.

The contribution of athletic participation to adolescent healthful living is a contentious issue ("Studies Raise Doubts," 1993). Despite an impressive corpus of research literature on the subject, the precise nature of the relationship between high school athletic participation and healthy behavior remains unclear, with research studies providing more questions than answers (Thorlindsson, Vilhjalmsson, & Valgeirsson, 1990). The present study was undertaken to investigate how nationally representative samples of moderately and highly involved female and male high school athletes compare with their nonathletic peers on a single health-related behavior, tobacco use (i.e., cigarette and cigar smoking and smokeless tobacco use), an addictive behavior that often begins in adolescence and is associated with increased rates of vascular and pulmonary disease in adulthood. We examined the relationship between high school athletic participation and tobacco use by replicating as well as improving upon previous research. Specifically, the strengths of the investigation include: (1) a recently sampled, nationally representative population of public and private high school students; (2) a greater focus on female users; (3) use of a measure of athletic participation which differentiates between highly involved and moderately involved athletes; (4) attention to both cigarette smoking and smokeless tobacco use; (5) a look at cigar smoking among high school athletes and nonathletes; and (6) a carefully controlled research design.

High School Students and Cigarette Smoking

Almost all current adult tobacco users initiated their habit as adolescents (Evans, 1998). Today, cigarette smoking is an entrenched part of teenage culture with an estimated 3,000 teens taking up the habit every day. Approximately 3.1 million adolescents are current smokers, and nearly one-third say they started smoking at age 12 or younger (Page & Koch, 1998). More than one in three high school students (35% of girls and 38% of boys) have smoked at least one cigarette in the past month ("High School Smokers," 1999). Despite the fact that the percentage of 12th graders who report smoking halfa pack of cigarettes or more a day dropped for the first time in years in 1998 (12.6%), it still remains well above where it stood in 1988 (10.6%) ("Up in Smoke," 1999).

Cigarette smoking during adolescence is associated with a number of significant health problems, such as respiratory illness (U.S. Department of Health and Human Services, 1994), and with several health-compromising behaviors, such as higher risk sexual behavior (Evans, 1998). Adolescent smoking is a complex, addictive behavior with a wide array of correlates including: ethnic group, attitude toward females who smoke, close friend's use of cigarettes, personal use of alcohol, school self-esteem, enrollment in college preparatory classes, grades, truancy, attendance at religious services, time spent in peer-oriented activities and peer group memberships (de Moor et al., 1994; Glendinning & Inglis, 1999; Husten, Chrisman & Reddy, 1996; McDermott et al., 1992; "Religion and Teen Drug Use," 1999; Wallace & Bachman, 1991).

Researchers in the United States have investigated the relationship between participation in organized sports and patterns of tobacco use at the county,

city, state, and national levels (Baumert, Henderson, & Thompson, 1998; Escobedo, Marcus, Holtman, & Giovino, 1993; Pate, Heath, Dowda & Trost, 1996; Rainey, McKeown, Sargent & Valois, 1996; Winnail, Valois, McKeown, Saunders & Pate, 1995; Zill, Nord & Loomis, 1995). These researchers have generally found that youth who are physically active and/or participate on high school athletic teams are less likely to smoke cigarettes or consider themselves regular smokers than their less active, less athletic peers. This finding appears to be the case for both genders through the high school years. On the other hand, other researchers have failed to replicate this finding (Buckhalt, Halpin, Noel, & Meadows, 1992) which can be explained in part by differences in sampling procedures (i.e., convenience samples versus regional or national samples), operational definition s of cigarette smoking (e.g., number of cigarettes smoked in the previous 30 days), and research designs.

High School Students and Smokeless ("Spit") Tobacco

The two major forms of smokeless tobacco--chewing tobacco (e.g., Redman, Levi Garrett, Beachnut) and dipping tobacco, also referred to as dipping snuff (e.g., Skoal, Skoal Bandits, Copenhagen)--pose major health risks, especially for their most frequent users, teenage males and young adult males. The documented health dangers associated with chronic smokeless tobacco use include cancer of the mouth, larynx, throat and esophagus, periodontal disease, nicotine addiction, coronary and peripheral vascular disease, delayed wound healing, reproductive disturbances, peptic ulcer disease, and decreased ability to taste (Hu et al., 1996; Riley et al., 1996; Walsh et al., 1994; Wichmann & Martin, 1994).

The Centers for Disease Control estimate that approximately 9.3% of male and 1.5% of female high school students use smokeless tobacco at least once a month (Koch, 1998). Among high school students who ever used smokeless tobacco, 73% did so by the 9th grade (Centers for Disease Control, 1994), suggesting that similar to cigarette smoking, the behavior begins in the pre-teen years. Tomar and Giovino (1998) estimate that in the United States, approximately 824,000 young people 11 to 19 years of age experiment with smokeless tobacco each year, and about 340,000 eventually become regular users.

Besides peer group influence (Ary, Lichtenstein, & Severson, 1987; Hill, Harrell, & McCormick, 1992), gender is a powerful predisposing factor to smokeless tobacco use. Research shows that the overwhelming number of smokeless tobacco users are males (Kegeles, Burleson, & Miozza, 1998; Lopez & Hamlin, 1995; New York State Education Department, 1997). This can be explained in part by the advertising industry's machismo image of the user, the well-publicized use of the product by celebrity male athletes, and the belief that chewing/dipping and winning are somehow related (Wichmann & Martin, 1994).

Researchers who have investigated the relationship between participation in organized sports and smokeless tobacco use have reported mixed results. While some have found that high school male athletes are more likely than nonathietes to be regular users (Hu et al., 1996; Tomar & Giovino, 1998), others have found no differences in smokeless tobacco use Baumert et al., 1998; Buckhalt et al., 1992; Rainey et al., 1996). The inconsistency can be explained by differences among subjects and operational definitions of smokeless tobacco use, the inclusion (or exclusion) of control variables, and variations in statistical analysis.

Far from an exclusively rural, blue-collar, Southern phenomenon (Horn et al., 2000), the use of smokeless tobacco among male adolescents is a growing national health concern. This substance is every bit as addictive as cigarette smoking, and in fact, the psychosocial predictors for smokeless tobacco use are not unlike those associated with cigarette smoking, namely, "... family structure, family conflicts, parental alcohol use, peer drug use, activities with peer (parties and sports), risk taking, stress, low achievement in school, and use of other substances" (Hu et al., 1996).

While recent statistics indicate that the mid-1990s cigar smoking frenzy in the United States has subsided (Shriver, 1999), estimated retail sales of premium cigars still reached $500 million in 1999. Far from a safe alternative to cigarettes (5 cigars a day is equivalent to smoking 1 pack of cigarettes), the Federal Trade Commission required health warnings on cigar packages and in advertisements beginning in early 2001 ("FTC, Cigar Makers Agree on Warnings," 2000). The fact that the number of cigar smokers in the U.S. has doubled since 1986 has the FTC especially concerned. Given the paucity of research on teenage cigar smoking, the present study fills an important void in the literature.

The primary goal of the investigation was to assess whether high school athletic participation is associated with greater or lesser odds of cigarette and cigar smoking and smokeless tobacco use. We also sought to determine whether there are gender differences with respect to tobacco use among athletes and nonathletes. The following null hypotheses were tested: (1) there will be no significant differences in cigarette and cigar smoking and smokeless tobacco use between athletes and nonathletes; and (2) there will be no significant differences in cigarette and cigar smoking and smokeless tobacco use between male and female athletes.

METHOD

Survey Design

The 1997 Youth Risk Behavior Survey (YRBS) was developed by the Centers for Disease Control and Prevention for the purpose of monitoring the prevalence of health-risk behaviors among adolescents. These behaviors, which are frequently interrelated, include those resulting in intentional or unintentional injuries; use of substances such as tobacco, alcohol, and other drugs; sexual behavior with the potential of leading to unintended pregnancy or HIV infection; unhealthy dietary behaviors; and physical inactivity. The national, school-based YRBS was conducted in 1990, again in 1991, and biennially during odd-numbered years since then. For the 1997 target year, students completed a self-administered, 84-item questionnaire that included questions about basic demographic characteristics and adolescent lifestyle behaviors that contribute either directly or indirectly to morbidity and mortality.

Data Collection and Definitions

A total of 16,262 questionnaires were completed in 151 public and private high schools for a response rate of 79%, and a student response rate of 87%. Although YRBS assesses adolescent health-risk behaviors in seven domains, the present investigation focuses only on tobacco use, specifically, cigarette and cigar smoking and smokeless tobacco use (chewing tobacco and dipping snuff). Since some respondents chose not to answer all of the tobacco use questions, sample sizes vary across the different variables.

The athletic participation variable was measured by combining responses to two items on YRBS, namely (1) "During the past 12 months, on how many sports teams run by your school did you play (Do not include physical education classes)?"; and (2) "During the past 12 months, on how many sports teams run by organizations outside of your school did you play?" This allowed the researchers to create dichotomous ("Did not participate on any teams"; "Participated on one or more teams") and tripartite ("Did not participate on any teams"; "Participated on one or two teams"; "Participated on three or more teams") response categories.

Four tobacco use variables were constructed to help assess the overall relationship between adolescent sports participation and tobacco use: (1) Have you ever smoked cigarettes regularly, that is, at least one cigarette every day for 30 days? (2) During the past 30 days, did you smoke any cigarettes? (3) During the past 30 days, did you smoke any cigars, cigarillos, or little cigars? (4) During the past 30 days, did you use chewing tobacco or snuff, such as Redman, Levi Garrett, Beachnut, Skoal, Skoal Bandits, or Copenhagen?

Data Analysis

For each of the major statistical analyses reported, two versions are given; one for the dichotomous athletic team variable (nonathletes and athletes), and the other for the tripartite athletic team variable (nonathletes; moderately involved athletes playing on 1 or 2 teams; highly involved athletes playing on 3 or more teams).

Gender-specific descriptive statistics were first calculated for the whole sample (not divided by athlete status) in order to gain a better understanding of general cigarette and cigar smoking and smokeless tobacco use patterns. Descriptive statistics were weighted to compensate for oversampling of Black and Hispanic respondents. After controlling for age, race/ethnicity (White, Black, Hispanic/Latino, Asian! Pacific Islander), parental educational achievement (a proxy for socioeconomic status), and metropolitan status (urban, suburban, rural), logistic regression was used to calculate odds ratios (ORs) in order to examine the association between sports team participation and tobacco use. For all analyses, the reference category designation consisted of all nonathietes.

RESULTS

Descriptive Statistics: Sports Participation Variable

Approximately 68.1% (n = 5,973) of male respondents were members of one or more teams; for females, the statistic was 51.1% (n = 3,727). Among male athletes, 56.8% (n = 3,390) were moderately involved in sports (participated on 1 or 2 teams), and 43.2% (n = 2,583) highly involved (participated on 3 or more teams). Among female athletes, 65.9% (n = 2,457) were moderately involved in sports, and 34.1% (n = 1,270) highly involved (see Table 1).

Descriptive Statistics: Tobacco Use Variable

To the first question ("Have you ever smoked cigarettes regularly, that is, at least one cigarette every day for 30 days?"), approximately one-fourth of the sample (n = 3,967) indicated that they had ever been regular smokers (25.4% of the females and 24.3% of the males).

To the second question ("During the past 30 days, did you smoke any cigarettes?"), 34.7% of the females (n = 2,441) and 37.7% of the males (n = 3,181) indicated they had smoked one or more cigarettes during the previous month.

Based on responses to the third question ("During the past 30 days, did you smoke any cigars, cigarillos, or little cigars?"), 78.0% answered the question in the negative. Approximately 89.2% of the females (n = 6,485) and 68.8% of the males (n = 6,062) responded "no."

For the fourth question ("During the past 30 days, did you use chewing tobacco or snuff?"), relatively few females answered in the affirmative. Of the 7,313 females who responded to the question, only 1.5% (n = 109) said they had used snuff or chewing tobacco during the previous month compared to 15.8% (n = 1,399) of the males.

Summarizing these findings, we see that approximately 75% of all respondents had never smoked regularly (defined as at least one cigarette every day for 30 days). Slightly more males than females had smoked recently (37.7% vs. 34.7%). Cigar smoking was very unpopular among females with only 10.8% indicating that they had smoked at least one cigar during the previous 30 days. By contrast, cigar smoking was three times more popular with males; 31.2% indicated that they had tried the product at least once during the previous month. Chewing tobacco and dipping snuff were even more unpopular with females as evidenced by the fact that less than 2% admitted trying the product in the previous month; males, by comparison, were ten times more likely to have tried smokeless tobacco (15.8%). Logistic Regression with Odds Ratios for the Dichotomous and Tripartite Sports Participation Variables

Logistic regression and odds ratios for each of the four tobacco use variables revealed significant findings for the dichotomous model (see Table 2). Female athletes were 29% less likely than female nonathletes (OR .71, p < .001) to be regular smokers, but 83% more likely to use smokeless tobacco (OR = 1.83, p < .05). Male athletes were 36% less likely than male nonathletes to smoke regularly (OR = .64, p < .001), 21% less likely to smoke cigarettes during the previous month (OR = .79, p < .001), but 41% more likely to use smokeless tobacco (OR = 1.41, p < .001). The likelihood of female (OR = 1.01) and male (OR = 1.10) athletes smoking cigars did not differ significantly from the likelihood of their nonathletic counterparts doing so.

Logistic regression analysis revealed several significant findings for the tripartite model. Moderately involved female athletes were 25% less likely to smoke regularly (OR = .75, p <.001). No significant odds ratios were discovered for the other three tobacco variables. Highly involved female athletes were 39% less likely to smoke regularly (OR = .61, p < .001), and 17% less likely to smoke during the previous month (OR = .83, p < .05). However, they were three times as likely (OR 3.20, p < .001) to use smokeless tobacco.

Moderately involved male athletes were 32% less likely to smoke regularly (OR = .68, p < .001), and 20% less likely to smoke during the previous month (OR = .80, p < .001). However, they were 26% more likely to use smokeless tobacco (OR = 1.26, p < .05). Highly involved male athletes were 42% less likely to smoke regularly (OR = .58, p < .001), and 23% less likely to smoke during the previous month (OR = .77, p < .001). However, they were 64% more likely to use smokeless tobacco (OR = 1.64, p <.001).

Analysis of the cigar smoking data was rated a "toss up" for both moderately involved (OR = 1.09) and highly involved (OR = 1.12) male athletes. That is, respondents who were moderately or highly involved in organized sports were no more (or less) likely to smoke cigars than were nonathletes.

To summarize, the gender-specific odds ratio findings based on logistic regression analysis support the following conclusions.

1. Both male and female athletes were less likely to have ever smoked regularly compared to their nonathletic counterparts. The effect was stronger for more highly involved athletes of both genders.

2. Male athletes were less likely than male nonathletes to have smoked within the past month. For female athletes, the finding was weaker, and applied only to highly involved athletes.

3. Cigar smoking was not significantly associated with athlete status.

4. Both male and female athletes were more likely to use smokeless tobacco than were their nonathletic counterparts. The effect was stronger for more highly involved athletes of both genders.

DISCUSSION

These overall findings confirm the fact that cigarette smoking is very much part of teen culture in the United States. Recall that approximately 25% of the sample indicated they had been regular smokers at some point, and over a third had smoked within the past month. Given the early age at which teens initiate tobacco use, if the U.S. Congress is serious about its goal of reducing teen smoking in the country by 60% over the next 10 years (Page & Koch, 1998), it will need to focus its efforts on the critical preteen years when young people appear to be especially vulnerable.

School-based anti-smoking programs that target children in the elementary grades make a lot of sense in light of these findings. Relatedly, the transition from ages 11-12 to 13 has been determined by Columbia University's National Center on Addiction and Substance Abuse to be very important in the nation's battle against teen drug use ("Age 13 Critical," 1998). It is equally important in the life history of the adolescent cigarette smoker.

Given the fact that there are approximately 3.1 million adolescents in the United States who are current smokers, it was encouraging to find that among both females and males, high school athletes were significantly less likely than nonathletes to have ever smoked regularly or smoked within the past month. Of further interest was the finding that the effect was strongest for highly involved athletes who participated on three or more teams. While the research design does not allow us to discuss the findings in terms of causality, the patterns were consistent and robust across all statistical treatments. Why were athletes less likely to smoke cigarettes, more likely to use smokeless tobacco, and similar to nonathletes in their use of cigars? A number of theoretical explanations can. be offered.

Health Awareness Factor

Compared to nonathletes, teen athletes are more apt to be exposed to health information and health counseling about the dangers of smoking from athletic directors, coaches, athletic trainers, school nurses, and other health providers. As a consequence they are more likely to perceive cigarette smoking as a real danger to their health, and more likely to avoid smoking as a result. We speculate that both female and male athletes have a greater awareness of the negative consequences associated with the use of tobacco in general, and cigarettes in particular.

Performance Factor

One of the major concerns of athletes is how to improve and/or maximize their performance. The highly competitive nature of organized sports in the United States is such that the product or outcome is often regarded as a lot more important than the process ("playing the game"). An "ugly win" is still a win, regardless of how well a team played. Doing well is looked upon in American sports as a lot more desirable than merely doing. This performance-based sports model encourages participants to perform at the highest level of which they are capable. Most athletes believe that smoking will hurt their performance, and compromise their chance for success. If they are not personally aware of the debilitating effects associated with smoking, then the mass media are sure to remind them. As a result, adolescent athletes equate smoking with lower cardiovascular endurance, a lack of stamina, slower reaction time/movement time, and loss of the "competitive edge." We speculate that athletes, not wanting to ruin their own or their team's chances for success, are more likely to steer clear of cigarettes.

Stress Reduction Factor

While initial introduction to cigarette smoking is generally related to social norms, adolescent social pressure and curiosity, once the behavior becomes established, the user is likely to report that it is a way of relaxing, a source of pleasure, an opportunity to alleviate stress, and for some females, a way of losing weight (Centers for Disease Control, 1994). High levels of physical activity including participation in organized sports provide athletes with another way of dealing with their anxieties and stresses. A vigorous workout or a hard-fought competition may provide the athlete with the stress reduction that his/her nonathletic peers perhaps find in cigarette smoking.

Mood Elevation Factor

Nicotine has been found to be a powerful inducement for cigarette addiction because it elevates mood (Wagner & Atkins, 2000). Physical activity is known to influence the physiological production of endorphins in the brain that also elevate mood. We speculate that athletes may be less attracted to cigarette smoking as a mood elevator because competitive sports provide them with adequate physiological stimulation.

Social Status Factor

A consistent predictor of adolescent cigarette smoking behavior is low self-esteem and low self-confidence (McDermott et al., 1992). Relatedly, research shows that athletes generally score higher than nanathletes on self-esteem and self-confidence scales (Kamal et al., 1995). Whether this difference is the result of positive socialization benefits resulting from participation in organized sports or sample bias resulting from adolescents already high in self-esteem self-selecting themselves into the athletic stream is difficult to say. In any case, because sport is such a highly valued activity in American society, those who perform at the highest levels typically receive considerable status and prestige, and this is no less true 'within the adolescent social system. Because athletes generally enjoy higher social status than do their nonathletic peers, we speculate that they are less inclined to use cigarettes for status enhancement and self-esteem boosting. Interestingly, these same positive consequences asso ciated with membership in the athletic subculture provide a rationale for "cultural resource theory," a theoretical explanation for understanding the differential sexual behavior patterns of adolescent female and male athletes (Miller et al., 1998, 1999).

Athletic Identity Factor

As a result of their participation in organized sports, teenagers are likely to develop some variation of an athletic identity. For obvious reasons, the athletic identity is antithetical to that of a smoker. On the other hand, the smoking behavior of teens not involved in organized athletic programs may function as a signifier of teenage, quasi-adult authority. Thus, athletic participation helps youngsters associate with a peer group and a subcultural value system which nurtures and reinforces the athletic identity, while enhancing their social status in the process. Cigarette smoking devalues personal identity and status, and compromises the neat and conventional appearance many athletes seek (Escobedo et al., 1993).

Athletic Subculture Factors

In describing Jessor and Jessor's (1977) problem behavior theory, Wallace and Bachman (1991) write: "The theory hypothesizes that young people who are less invested in traditional versus deviant behaviors, who are more strongly tied to peers than to parents, who are alienated from society, who have low self-esteem, and who hold unconventional beliefs, values, and attitudes are prone to become involved in problem behavior" (p. 334). Clearly, background and lifestyle factors play important roles in the etiology of adolescent problem behavior.

How might membership on a high school athletic team deter tobacco use? If problem behavior is related to nontraditional activities and unconventional beliefs, values, and attitudes, what impact might team membership have on cigarette smoking? In high school sports, we have a traditionally conservative, highly valued activity capable of influencing the participant's beliefs, values, and dispositions in prosocial ways. Moreover, the "family" atmosphere of an athletic team, and the support provided by caring adult coaches and staff members, can help offset negative family background and lifestyle factors.

We speculate that these subcultural forces involve the participant in a social system that devalues nonconformist beliefs and unconventional attitudes and behaviors frequently associated with adolescent smokers. Related to this point, Glendinning and Inglish (1999) found that disaffected youth who held largely negative attitudes toward authority and control, especially with respect to school and community, were much more likely to report frequent smoking behavior. In contrast, "conventional" peer groups whose members were well-integrated into school and family contexts viewed cigarette smoking in largely negative terms.

Adolescent smoking is also related to social insecurity and social isolation (Evans, 1998). Membership on an athletic team can serve as an antidote to both because it is a situation "ablaze with communal possibility" (Novak, 1976). We speculate that rather than seek security and identity in smoking-related group behavior, athletes are more likely to satisfy their social needs through the social relationships available through membership in the athletic subculture.

In addition to providing a prosocial, normative environment in which the concept of "family" is nurtured and reinforced, team members are expected to abide by a code of conduct and a set of training rules which place considerable restraints on several behaviors, including cigarette smoking (Pate et al., 1996). Athletes who are caught smoking can experience a variety of negative sanctions, including a strong reprimand, an extra workout, and even suspension from the team. For example, the following training rule pertaining to smoking applies to all varsity, junior varsity, and freshman athletes at a suburban high school located in western New York State: "Indulging in smoking is not permitted at any time. Any squad member violating this rule will be dismissed from his or her respective squad for the balance of that particular season. A second violation (during a second season) means automatic dismissal from all squads for the balance of the school year." Most athletes are not likely to risk getting caught smoki ng if it means the loss of their athletic identity and team membership.

Peer influence has been documented as an especially strong predictor of adolescent smoking behavior. Because athletes are less apt to smoke as a group, we speculate that they are less likely to come under peer pressure to do so. If smoking is less prevalent in youth athletic subcultures, it may well be because peer pressures to smoke are less pervasive.

It is a well-documented fact that smoking by adult role models exerts a powerful influence on the smoking behavior of children and youth. Because coaches, athletic trainers, and adult fitness enthusiasts are less likely to smoke, we speculate that athletes are much more likely to come into contact with positive role models, which helps lower their smoking risk.

Community Factor

In addressing the issue of how to reduce if not eliminate at-risk adolescent behavior, Irwin (1990) observed that "the protective role of supportive environments during adolescence must be acknowledged and may be critical in developing prevention and intervention programs" (p. 10). If cigarette smoking is an act of rebellion against school and community as some claim, then athletic participation and team membership can help draw young people into the life of the school and community. By fostering strong ties between family, school, and community, interscholastic sports programs provide athletes with a supportive, valued environment, one which encourages and reinforces prosocial values and dispositions. We suggest that athletes have less need to rebel, less reason to "act out," less inclination to engage in a variety of risky behaviors, including cigarette smoking.

The above theoretical explanations are helpful in understanding why the female and male athletes most highly involved in sports (i.e., membership on three or more teams) were least likely to report being regular smokers or having smoked within the previous month. Because of their multiple team memberships, they have greater access to health information and counseling; they are more "locked into" the high performance principle; they are more influenced by the athletic subculture; they more strongly identify with being "an athlete"; and, they are more socially integrated into school and community life.

Cigar Smoking Findings

Recall that while moderately and highly involved athletes were no more (or less) likely than their nonathletic counterparts to smoke one or more cigars during the past month, almost one-third of the male respondents indicated that they had smoked one or more cigars during the past month (less than 11% of the females responded the same).

Given the association between cigar smoking and mouth and throat cancer, this finding is disturbing. We speculate that we might even see greater cigar use among male adolescents if not for the following factors: Cigars are likely to stain and discolor the user's teeth, lips, and fingertips; the odor of cigar smoke is offensive to some; and advertisers have created few cigar-smoking celebrities with whom male adolescents can identify.

Smokeless Tobacco Findings

If the "good news" of the study is that athletes are less likely than nonathletes to smoke cigarettes, the "bad news" is that they are much more likely to chew and dip smokeless tobacco. Far from a safe alternative to cigarette smoking, smokeless tobacco has been causally linked to several oral health problems ranging from halitosis and periodontal disease to cancers of the mouth, throat, and larynx. Despite Major League Baseball's ban on the use of smokeless tobacco products in the minor leagues, and the NCAA's all-sport ban on the use of smokeless tobacco during NCAA practices and games (Wichmann & Martin, 1994), the product continues to appeal to a growing number of adolescent athletes. Recall that female and male athletes were 83% and 41% more likely, respectively, to use smokeless tobacco. Highly involved female athletes were three times as likely, and highly involved male athletes 64% more likely to use smokeless tobacco than were their nonathletic peers.

The attraction of young people to smokeless tobacco can be explained, in part, by the advertising industry's endorsement and glamorization of the product via the print media, outdoor advertising, entertainment events, stadium signs, on-site promotions, and sponsorship of major sporting events (Evans, 1998; Madden & Grube, 1994). In addition to the role advertisers and sponsors play in helping make smokeless tobacco the significant health problem it has become, we also should not overlook the powerful influence coaches and celebrity athletes have in shaping adolescent perceptions and attitudes. Consider the following comments made by teenagers in focused interviews about smokeless tobacco (Wichmann & Martin, 1994, p. 97): "Famous athletes use dip. One of the things that goes along with playing ball is taking a dip." "Walt Garrison [former star player for the Dallas Cowboys of the NFL] talked about Skoal. I wanted to be like him. He's a hero. I wanted to be one also." "My wrestling coach wrestles with Copenhag en in his mouth."

The fact that smokeless tobacco is openly used by the very role models young people look to for guidance and direction is disturbing to say the least. Horn et al. (2000) recently surveyed a sample of 566 West Virginia middle and high school coaches and found that 16% were current smokeless tobacco users, and 36% used the product when their athletes were present. The fact that smokeless tobacco has come to be associated in the minds of some young athletes with professionalism and success further exacerbates the problem. For example, it is estimated that 35--40% of Major League baseball players chew tobacco ("Bad News for Players," 1998). When youngsters can buy a product labeled Big Chew, a bag of shredded, pink confection (essentially gum) made to look like chewing tobacco, we see that changing young people's attitudes about smokeless tobacco will not be an easy task.

To focus only on a role model explanation for the smokeless tobacco behavior associated with high school students, however, is to minimize the complexity of the problem. The use of smokeless tobacco by teenage male athletes must also be analyzed within the context of historically situated, specific, men's sport subcultures (e.g., the use of chewing tobacco by baseball players is as old as the sport itself). By doing so, we appropriately recognize the powerful institutional and cultural processes which heavily impact on individual athletic behavior.

Perhaps the most striking finding of the study was that highly involved female athletes (membership on three or more teams) were three times more likely (OR = 3.20, p < .001) than female nonathletes to have used smokeless tobacco at least once during the preceding month. Although only 2.6% (n 33) of the highly involved athlete sample, these young women have opted for a behavior that can have serious health risk consequences. We wonder whether their use of smokeless tobacco is further evidence that girls' and women's sports are coming more and more to resemble the male athletic model. The use of smokeless tobacco by highly involved female athletes may be just another manifestation of a developing pattern of male imitative behavior. Other signs include: (1) an increase in player violence; (2) physically and emotionally abusive hazing practices; (3) the use of performance-enhancing drugs and nutritional supplements; (4) preference for an instrumental over an expressive sport orientation (Pilz, 1982); and (5) a tendency toward moral callousness when confronted with sport-specific, ethical decision-making questions (Beller & Stoller, 1995).

If, in fact, female athletes in particular sport subcultures (e.g., basketball, ice hockey, soccer) are being assimilated into a male athletic model, one that encourages and reinforces risky behavior and the adoption of destructive health practices (e.g., binge drinking, tobacco use, aggression), then the larger and more challenging solution to the problem lies in changing and reforming men's sports. Alternatively, the female athlete's use of smokless tobacco may be less a matter of imitating males than an attempt to break loose from feminine expectations and gender-based normative prescriptions.

While these findings strongly suggest a negative association between high school sports participation and cigarette smoking for both sexes, the cross-sectional nature of the data rules against making causal statements. The study is also limited in the following ways: (1) the use of self-report data invites distortion; (2) no attempt was made to verify the smoking data with appropriate physiological measures; (3) the findings probably underestimate the prevalence of tobacco use because the data do not include students who dropped out of school, a group known to have relatively high rates of tobacco use; (4) although efforts were taken to control for important mediating variables, a certain amount of self-selection into the "athletic stream" no doubt occurred; and (5) all responses were contingent upon accurate recall by the respondents.

Despite these limitations, the negative associations between athletic participation and cigarette smoking are well worth the attention of those interested in reducing if not altogether eliminating teen smoking in the United States. The popularity and salience of sport within teen culture is an open invitation to educational leaders to explore the inclusion of organized sports activities in comprehensive intervention programs aimed at changing adolescent cigarette smoking behavior.

On the other hand, the positive associations between athletic participation and smokeless tobacco use are disturbing, and encourage health and physical educators, coaches, and public health advocates to redouble their efforts at opposing the behavior while educating for its prevention (Walsh et al., 2000). The National Spit Tobacco Education Program (NSTEP) which seeks to educate players and coaches about the dangers of smokeless tobacco is a move in the right direction (Balog, 1997). Adolescent use of chewing and dipping tobacco may prove just as difficult a problem to solve as teen smoking unless adult leaders take a proactive role in dispelling prevailing myths about its use, provide factual information wherever and whenever appropriate, and most importantly, acknowledge the serious health problem it has become.

Table 1.

Weighted Descriptive Statistics: Sociodemographics and Tobacco Variables

 Females Males
 (n=7,352) (n=8,899)
 % n % n

Age (in years)
 Under 15 11.2 824 8.7 772
 15 23.3 1,715 21.1 1,885
 16 25.5 1,875 26.6 2,364
 17 26.2 1,923 26.4 2,341
 l8 and older 13.8 1,013 17.1 1,522
Mean age 16.08 16.22
Race/ethnicity
 White 68.9 4,414 72.4 5,639
 Black 16.0 1,027 12.7 991
 Hispanic 11.3 724 11.0 858
 Asian 3.8 241 3.9 305
Highest level of parental
education
 Less than high school 8.5 585 5.3 436
 High school graduate 18.6 1,283 14.7 1,205
 Some college 25.0 1,723 22.5 1,846
 College graduate 47.9 3,303 57.5 4,714
Area of residence (school
district)
 Urban 32.1 2,357 34.9 3,006
 Suburban 53.1 3,904 51.2 4,409
 Rural 14.8 1,091 13.9 1,194
Athletic participation
in past year
 No 48.9 3,566 31.9 2,799
 Yes 51.1 3,727 68.1 5,973
Level of athletic involvement
 Not involved (0 teams in 48.9 3,566 31.9 2,799
 past year)
 Moderately involved 33.7 2,457 38.6 3,390
 (1-2 teams)
 Highly involved (3+ teams) 17.4 1,270 29.4 2,583
Ever smoked cigarettes regularly
 No 74.6 5,394 75.7 6,645
 Yes 25.4 1,834 24.3 2,133
Smoked cigarettes in past month
 No 65.3 4,601 62.3 5,248
 Yes 34.7 2,441 37.7 3,181
Smoked cigars in past month
 No 89.2 6,485 68.8 6,062
 Yes 10.8 788 31.2 2,749
Chewed or dipped tobacco in
past month
 No 98.5 7,204 84.2 7,433
 Yes 1.5 109 15.8 1,399
Table 2.

Logistic Regressions of Tobacco Variables on Age, Race/Ethnicity,
Parental Education, Residential Type, and Athletic Participation/Level
of Athletic Involvement: Odds Ratios

 Females
 ever smoked smoked chewed or
 smoked cigarettes, cigars, dipped
 regularly past month past month tobacco,
 past month

Age 1.11 (***) 1.09 (***) 1.02 1.00
Black .16 (***) .30 (***) .97 .59
Hispanic/Latino .36 (***) .58 (***) .93 1.07
Asian/Pacific Islander .28 (***) .40 (***) .63+ 1.99
Parental Education .97 .96 1.00 1.04
Urban .89 .92 .97 .96
Rural 1.10 .98 1.16 1.51
Athletic Participation .71 (***) .91 (+) 1.01 1.83 (***)
Athletic Involvement
 Moderately Involved .75 (***) .94 .95 1.28
 Highly Involved .61 (***) .83 (*) 1.15 3.20 (***)

 Males
 ever smoked smoked chewed or
 smoked cigarettes, cigars, dipped
 regularly past month past month tobacco,
 past month

Age l.20 (***) 1.18 (***) 1.21 (***) l.21 (***)
Black .32 (***) .53 (***) .70 (***) .10 (***)
Hispanic/Latino .49 (***) .87 (+) .67 (***) .32 (***)
Asian/Pacific Islander .53 (***) .55 (***) .48 (***) .31 (***)
Parental Education .97 .97 1.10 (**) .95
Urban .94 .94 .92 .73 (***)
Rural 1.18 1.37 (***) 1.08 1.50 (***)
Athletic Participation .64 (***) .79 (***) 1.10 1.41 (***)
Athletic Involvement
 Moderately Involved .68 (***) .80 (***) 1.09 1.26 (*)
 Highly Involved .58 (***) .77 (***) 1.12 1.64 (***)

(+)p<.10, (*)p<.05, (**)p<.01, (***)p<.001

REFERENCES

Age 13 critical in anti-drug fight survey finds. (1998, September 2). Democrat and Chronicle (Rochester, NY), p. A5.

Ary, D. V., Lichtenstein, E., & Sverson, H. H. (1987). Smokeless tobacco use among male adolescents: Patterns, correlates, predictors, and the use of other drugs. Preventive Medicine, 16, 385-401.

Bad behavior in junior high. (1999, August 9). USA Today, p. A1.

Bad news for players who chew tobacco. (1998, April 1). USA Today, p. C6.

Balog, K. (1999, Oct. 29-31). Smart strategies to keep kids tobacco-free. USA Weekend, p. 11.

Baumert, P. W., Henderson, J. M., & Thompson, N. J. (1998). Health risk behaviors of adolescent participants in organized sports. Journal of Adolescent Health, 22, 460-465.

Beller, J. M., & Stoll, S. K (1995). Moral reasoning of high school students and general students: An empirical study versus personal testimony. Pediatric Exercise Science, 7, 352-363.

Buckhalt, J. A., Halpin, G., Noel, R., & Meadows, M. E. (1992). Relationship of drug use to involvement in school, home, and community activities: Results of a large survey of adolescents. Psychological Reports, 70, 139-146.

Centers for Disease Control and Prevention. (1994). Reasons for tobacco use and symptoms of nicotine withdrawal among adolescents and young adult tobacco users-United States, 1993. Journal of the American Medical Association, 272, 1648-1649.

de Moor, C., Johnston, D. A., Werden, D. L., Elder, J. P., Semi, K., & Whitehorse, L. (1994). Patterns and correlates of smoking and smokeless tobacco use among continuation high school students. Addictive Behaviors, 19, 175-184.

Escobedo, L. G., Marcus, S. E., Holtzman, D., & Giovino, G. A. (1993). Sports participation, age at smoking initiation, and the risk of smoking among U.S. high school students. Journal of the American Medical Association, 269, 1391-1395.

Evans, D. W. (1998). Tobacco use and adolescents. In A. Henderson & S. Champlin (Eds.), Promoting teen health (pp. 46-57). Thousand Oaks, CA: Sage.

FTC, cigar makers agree on warnings. (2000, June 21). USA Today, p. D9.

Glendinning, A., & Inglis, D. (1999). Smoking behavior in youth: The problem of low self-esteem? Journal of Adolescence, 22, 673-682.

High school smokers. (1999, March 16). USA Today, p. A1.

Hill, M. B., Harrell, J. S., & McCormick, L. K. (1992). Predictors of smokeless tobacco use by adolescents. Research in Nursing and Health, 15, 359-368.

Horn, K. A., Maniar, S. D., Dino, G. A., Gao, X., & Meckstroth, R. L. (2000). Coaches' attitudes toward smokeless tobacco and intentions to intervene with athletes. Journal of School Health, 70, 89-100.

Hu, F. B., Hedeker, D., Flay, B. R., Sussman, S., Day, L. E., & Siddiqui, O. (1996). The patterns and predictors of smokeless tobacco onset among urban public school teenagers. American Journal of Preventive Medicine, 12, 22-28.

Husten, C. G., Chrismon, J. H., & Reddy, M. N. (1996). Trends and effects of cigarette smoking among girls and women in the United States, 1965-1993. Journal of the American Medical Women's Association, 51, 11-18.

Irwin, C. B., Jr. (1990). The theoretical concept of at-risk adolescents. Adolescent Medicine: State of the Art Reviews, 1, 1-14.

Jessor, R., & Jessor, S. (1977). Problem behavior and psychosocial development. New York: Academic Press.

Kamal, A. F., Blais, C., Kelly, P., & Ekstrand, K. (1995). Self-esteem attributional components of athletes versus nonathletes. International Journal of Sport Psychology, 26, 189-195.

Kegeles, S. S., Burleson, J. A., & Miozza, J. (1989). Cigarette and smokeless tobacco use among Connecticut adolescents. American Journal of Public Health, 79, 1413-1414.

Koch, W. (1998, May 27). Garagiola takes aim at snuff tobacco. USA Today, p. C4.

Lopez, L. C., & Hamlin, P. A. (1995). Use of smokeless tobacco by Mexican-American high school students. Psychological Reports, 77, 808-810.

Madden, P. A., & Grube, J. W. (1994). The frequency and nature of alcohol and tobacco advertising in televised sports 1990 through 1992. American Journal of Public Health, 84, 297-299.

McDermott, R. J., Sarvela, P. D., Hoalt, P. N., Bajracharya, S. M., Marty, P. J., & Emery, E. M. (1992). Multiple correlates of cigarette use among high school students. Journal of School Health, 62, 146-150.

Miller, K. E., Sabo, D. F., Farrell, M. P., Barnes, G. M., & Melnick, M. J. (1998). Athletic participation and sexual behavior in adolescents: The different worlds of boys and girls. Journal of Health and Social Behavior, 39, 108-123.

Miller, K. E., Sabo, D. F., Farrell, M. P., Barnes, G. M., & Melnick, M. J. (1999). Sports, sexual behavior, contraceptive use, and pregnancy among female and male high school students: Testing cultural resource theory. Sociology of Sport Journal, 16, 366-387.

Nixon, H. L., II. (1997). Gender, sport, and aggressive behavior outside of sport. Journal of Sport and Social Issues, 21, 379-391.

Novak, M. (1976). The joy of sports. New York: Basic Books.

Page, S., & Koch, W. (1998, May 4). Students say government remedies will fall short. USA Today, pp. A1-A2.

Pate, R. R., Heath, G. W., Dowda, M., & Trost, S. G. (1996). Associations between physical activity and other health behaviors in a representative sample of U.S. adolescents. Journal of Public Health, 86, 1577-1581.

Pilz, G. A. (1982). Changes of violence in sport. International Review of Sport Sociology, 17, 47-71.

Rainey, C. J., McKeown, R. E., Sargent, R. G., & Valois, R. F. (1996). Patterns of tobacco and alcohol use among sedentary, exercising, nonathletic, and athletic youth. Journal of School Health, 66, 27-32.

Religion and teen drug use. (1999, June 10). USA Today, p. B1.

Riley, W. T., Barenie, J. T., Woodard, C. E., & Mabe, P. A. (1996). Perceived smokeless tobacco addiction among adolescents. Health Psychology, 15, 289-292.

Shriver, J. (1999, November 22). Not just blowing smoke. USA Today, p. D7. Studies raise doubts. (1993). Journal of the American Medical Association, 270, 798-801.

Taub, D. E., & Blinde, E. M. (1992). Eating disorders among adolescent female athletes: Influence of athletic participation and sport team membership. Adolescence, 27, 833-848.

Thorlindsson, T., Vilhjalmsson, R., & Valgeirsson, G. (1990). Sport participation and perceived health status: A study of adolescents. Social Science Medicine, 31, 551-556.

Tomar, S. L., & Giovino, G. A. (1998). Incidence and predictors of smokeless tobacco use among U.S. youth. American Journal of Public Health, 88, 20-26.

University of the State of New York. (1997). Youth risk behaviors in New York State: In Their Own Words 1997 survey results. New York: The State Education Department.

Up in smoke. (1999, September 23). USA Today, p. A18.

U.S. Department of Health and Human Services. (1994). Preventing tobacco use among young people: A report of the Surgeon General. Washington, DC: Author.

Wagner, E. F., & Atkins, J. H. (2000). Smoking among teenage girls. Journal of Child and Adolescent Substance Abuse, 9, 93-110.

Wallace, J. M., Jr., & Bachman, J. G. (1991). Explaining racial/ethnic differences in adolescent drug use: The impact of background and lifestyle. Social Problems, 38, 333-357.

Walsh, M. M., Ellison, J., Hilton, J. F., Chesney, M., & Ernster, V. L. (2000). Spit (smokeless) tobacco use by high school baseball athletes in California. Tobacco Control, 9(Suppl. 2), ii32-ii39.

Walsh, M. M., Hilton, J. F., Ernster, V. L., Masouredis, C. M., & Grady, D. G. (1994). Prevalence, patterns and correlates of spit tobacco use in a college athlete population. Addictive Behaviors, 19, 411-427.

Wichmann, S., & Martin, D. R. (1994). Snuffing out smokeless tobacco use. The Physician and Sportsmedicine, 22, 97-110.

Winnail, S. D., Valois, R. F., McKeown, R. E., Saunders, R. P., & Pate, R. R. (1995). Relationship between physical activity level and cigarette, smokeless tobacco, and marijuana use among public high school adolescents. Journal of School Health, 65, 438-442.

Zill, N., Nord, C. W., & Loomis, L. S. (1995). Adolescent time use, risky behavior, and outcomes: An analysis of national data. Rockville, Maryland: Westat, Inc.

Kathleen E. Miller, Research Institute on Addictions, University of Buffalo, Buffalo.

Donald F. Sabo, Department of Sociology, D'Youville College, Buffalo.

Michael P. Farrell, Department of Sociology, University of Buffalo, Buffalo.

Reprint requests to Merrill J. Melnick, Department of Physical Education and Sport, State University of New York, College at Brockport, 350 New Campus Drive, Brockport, New York 14420. Electronic mail may be sent via Internet to mmelnick@brockport.edu.

Grace M. Barnes, Research Institute on Addictions, University of Buffalo, Buffalo.