The relationship between religiosity, employment, and political beliefs on substance use among high school seniors.

Abstract

The purpose of this study was to provide a descriptive profile of the alcohol, cigarette, marijuana and cocaine use practices among adolescents and to examine the relationship between employment, political beliefs, religious beliefs and substance use behaviors among high school seniors participating in the Monitoring the Future Study during the 2002 academic school year. Data from the Monitoring the Future (MTF) study (2002 12th grade core data) was used for this study. Logistic regression models were used to examine the relationship between substance use behaviors and employment, political beliefs, and religiosity. Analysis showed that males and females who believed religion was very important were less likely to have initiated alcohol use, to be a current user, and to have binge drank. A significant association was found for all alcohol use variables for those individuals who worked moderate amounts at an after school job. Political beliefs were found to be associated for initiation and current alcohol use but not for binge drinking. Political beliefs, religion, and employment were all significantly associated with cigarette use and cocaine use. Conceptualization of substance use behavior and its prevention and treatment should include consideration of such key cultural and social factors as religiosity, employment and political beliefs of adolescents.

Key Words: Adolescents, Substance Use, Employment, Religiosity, Political Beliefs

INTRODUCTION

Alcohol, tobacco, and other drug use has been well documented among U.S. adolescents, but little is known about the role of religious affiliation, after school employment, and political beliefs on substance use among high school seniors. According to the Monitoring the Future Study, 79.7% of high school seniors had used alcohol in their lifetime and 49.8% had used in the past 30 days. Additionally, 29.7 of seniors had binge drank in the past 2 weeks. Cigarette use on a regular basis was reported by 19% of students and 49% of students had used marijuana and 8.2% had used cocaine in their lifetimes (Johnston, O'Malley, Bachman, 2002). Research has suggested that a host of factors are related to substance use among adolescents.

After school and weekend employment is an important factor in adolescent risk behaviors that has received little consideration. Adolescent employment has several benefits such as increased self-esteem, personal responsibility, and autonomy (Institute of Medicine, 1998). On the other hand, it has been reported that adolescent work has been shown to be related to greater substance abuse (Valois, Dunham, Jackson, Waller, 1999; Bachman & Schulenberg, 1993). Bachman, Safron, Sy, and Schulenberg (2003) found that those students who work long hours were more likely than the average student to use cigarettes, alcohol, and marijuana. This study also found a positive correlation with substance use and weekly income, specifically those who received more money, were more likely to be engaging in substance use behaviors.

Religion is another important variable in adolescent substance use. Studies consistently indicate that religiosity tends to be a protective factor for high-risk behaviors among adolescents. Specifically, higher levels of religiosity have been associated with lower levels of cigarette, alcohol, marijuana, and cocaine use among adolescents (Amey, Albrecht, Miller, 1996; Wallace & Forman, 1998; Miller, Davies, Greenwald, 2000; Brown, Schulenberg, Bachman, O'Malley, Johnston, 2001). Paulson, Eppler, Satterwhite, Wuensch, & Bass, (1998) found that college women with strong religious beliefs consumed less alcohol than females with weaker religious beliefs. In a study of substance use behaviors of 7th and 10th graders, it was found that strong religious beliefs was associated with a decrease in the likelihood of engaging in alcohol, tobacco, and marijuana use (Wills, Yaeger, Sandy, 2003). In addition, this study found that political beliefs were consistently linked to substance use behaviors.

The purpose of this study was to provide a descriptive profile of the alcohol, cigarette, marijuana and cocaine use among adolescents and to examine the relationship between employment, political beliefs, religious beliefs and substance use behaviors among high school seniors participating in the Monitoring the Future Study during the 2002 academic school year.

METHODS

Subjects and Data Collection

Data from the Monitoring the Future (MTF) study (2002 12th grade core data) was used for this study. The MTF study examines a wide variety of substance use behaviors and related attitudes and perceptions. Additionally data is collected on attitudes towards religion, changing roles for women, educational aspirations, self-esteem, exposure to sex and drug education, and violence and crime (both in and out of school). The sampling procedure for the MTF is a multistage probability sample design based on geographical area or primary sampling units (PSUs), schools within PSUs, and students within the sampled school. For the 2002 12th grade sample, approximately 133 public and private high schools nationwide participated. The overall response rate for 2002 was 83%; complete details of the sampling method have been published elsewhere (Johnston, Bachman, O'Malley, Schulenberg, 2002).

For data collection, the MTF collects data in-class with a questionnaire. About 10 days before the in-school administration, the students are given flyers explaining the study. Also, advance letters are sent to parents informing them about the study. The questionnaire administration was conducted by the local Institute for Social Research representatives and their assistants, following standardized procedures. The questionnaires are group administered in classrooms during normal class period whenever possible.

DATA ANALYSIS

Measures and Procedures

For this study six substance use measures (lifetime alcohol use, current alcohol use, current binge drinking, current cigarette use, lifetime marijuana use, and lifetime cocaine use) were included as dependent variables. Lifetime alcohol use was judged by response to the question, "On how many occasions have you had alcoholic beverages to drink (more than just a few sips) in your lifetime." The variable current alcohol use was based on the question, "On how many occasions have you had alcoholic beverages to drink (more than just a few sips) during the last 30 days." Binge drinking was indicated on the item, "How many times have you had five or more drinks in a row during the last two weeks." Current cigarette use was based on the question, "How frequently have you smoked cigarettes during the past 30 days." Lifetime marijuana use was measured with the question, "On how many occasions (if any)

have you used marijuana." Cocaine was assessed with the item, "On how many occasions (if any) have you taken cocaine in your lifetime." All substance use questions were dichotomized into yes/no categories.

The independent variables included political beliefs, religiosity, and number of hours worked after school. Political beliefs were assessed by the question, "How would you describe your political beliefs." This measure was tri-chotomized into conservative, moderate, and liberal political beliefs. Religiosity was based on the question, "How important is religion in your life." This measure was tri-chotomized into little important, pretty important, and very important. After school employment was based on the question, "On the average, over the school year, how many hours per week do you work in a paid or unpaid job." This variable was categorized into none, few (1 to 10 hours), moderate (11 to 20 hours), and high (21 or more hours) per week.

For data analysis purposes only white and black high school seniors were used. Frequencies were used to describe demographic characteristics, substance use behaviors and political beliefs, religiosity, and employment status. Logistic regression models were used to examine the relationship between substance use behaviors and political beliefs, religiosity, and employment. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for each gender.

RESULTS

Of the students who participated in the 2002 MTF study, 6029 valid observations were used in the analysis. The racial and gender distribution was 86.9% white with 52.1% being female. The data in Table 1 displays the frequencies and percentages for lifetime alcohol use, 30-day alcohol use, binge drinking, current cigarette use, and lifetime marijuana and cocaine use.

The data for lifetime alcohol use are displayed in Table 2. For lifetime use, significant associations were found between variables. Males with moderate and liberal political beliefs were more likely to have drank than those males with conservative views. Liberals (OR = 2.04) were twice as likely to have drank compared to those with conservatives beliefs. Similar finding were identified with females, in that liberals were more likely to have drank than moderate and conservative beliefs with liberal females (OR = 2.61) being almost 3 times more likely to drink than conservatives. Females working moderate and high work schedule were significantly more likely to have drank compared to those who did not work, with high employment females being almost twice as likely to have drank (OR = 1.72). Males and females who felt religion was very important were less likely to have initiated alcohol use compared to those who did not feel religion was important (OR = 1.32; 1.27).

The data for current alcohol use are displayed in Table 3. Both males and females with moderate and liberal views were significantly more likely to have drank in the past 30 days than those with conservative views. Males with liberal views had 1.60 times the odds of drinking in the past 30 days, whereas females with liberal views had 1.59 the odds of drinking in the past 30 days. Employment was associated with current alcohol use for both males and females. Males who worked moderate hours had 1.45 the odds of drinking and males with high working schedules had 1.61 the odds of drinking compared to those who did not work. Females who worked moderate hours had 1.40 the odds of being a current drinker compared to those who did not work. Males and females who felt religion was very important were less likely to be a current drinker compared to those who did not feel religion was important (OR = 1.40; 1.31).

The data for past 2 weeks binge drinking are displayed in Table 4. Moderate work schedule was found to be associated with binge drinking. Males and females who worked moderate hours had 1.34 and 1.45 the odds of being a binge drinking compared to those who did not work. These results were not found for those males and females who had high work schedules. It may be that those who worked long hours had less opportunity to engage in binge drinking. Males and females who felt religion was very important were less likely to be a binge drinker compared to those who did not feel religion was important (OR = 1.45; 1.34).

The data for current cigarette use are displayed in Table 5. Liberal political beliefs were associated with smoking for both males and females. Males had 1.67 times the odds of being a smoker and females had 2.00 times the odds of being a smoker compared to those individuals with conservative views. Employment was also associated with smoking for both males and females. Those individuals who work moderate or high work schedules were significantly more likely to smoke. Males who worked high levels were over three times more likely to be a smoker (OR = 3.34) and females with high work schedules were twice as likely to be a smoker compared to those who did not work (OR = 2.37). Males and females who believed religion was important were significantly less likely to be a current smoker (OR = 1.47; 1.40).

The data for lifetime cocaine use are displayed in Table 7. Males who held liberal political beliefs (OR = 1.86) were significantly more likely to have used cocaine than those with conservative beliefs. Employment was also a significant predictor of cocaine use. Moderate and high working schedule was associated with cocaine use for males (OR = 2.32; 2.82). High working schedule was a significant predictor for females. Females who worked high hours had 2.36 the odds of having used cocaine in their lifetime compared to those who did not work. Males and females who believed religion was important were significantly less likely to have tried cocaine (OR = 1.33; 1.30).

DISCUSSION

The purpose of this study was to provide a descriptive profile of the alcohol, cigarette, marijuana and cocaine use among adolescents and to examine the relationship between employment, political beliefs, religious beliefs and substance use behaviors among high school seniors participating in the Monitoring the Future Study during the 2002 academic school year. Analysis showed that among alcohol use variables (initiation, current, binge) similarity consisted among independent variables and across gender. Religiosity and employment were found to be significantly associated with all alcohol use variables. Males and females who believed religion was very important were less likely to have initiated alcohol use, to be a current user, and to have binge drank. A significant association was found for all alcohol use variables for those individuals who worked moderate amounts at an after school job. Political beliefs were found to be associated at the initiation and current alcohol use level but not for binge drinking. In regards to cigarette use and cocaine use, similarities were found among independent variables and across gender. Political beliefs, religion, and employment were all significantly associated with cigarette and cocaine use.

The results of this study seem to indicate that development of alcohol use patterns, cigarette use, and cocaine use are related to holding liberal views, not thinking religion is important, and working moderate to high amounts at an after school job. These results are similar to other studies, in that employment was found to be associated with increased rates of substance use (Miller & Miller, 1997). Valois et al., (1999) found that working more than 10 hours per week increased the odds of engaging in a host of alcohol, cigarette, marijuana and cocaine use practices among a sample of adolescents. Steinberg and Dornbush (1991), found that adolescents who worked more than 10 hours a week reported higher rates of alcohol use. Bachman and Schulenberg (1993) found that nearly two-thirds of adolescents who had used cocaine were employed at after school jobs. This link between employment and increased substance use may be best explained by two factors. First, adolescent workers are more likely to be exposed to older workers who may influence the adolescent to use various substances, through modeling behavior or providing opportunities to try substances (i.e., being offered). Second, adolescents who work will tend to have more disposable income to spend on substance use activities.

In terms of religiosity, this study's results support previous evidence that religiosity is a protective factor for a number of adolescent substance use outcomes. One study found that among college males and females, those with strong religious beliefs consumed less alcohol and were less likely to engage in risky sexual behavior than were students with weaker religious convictions (Poulson, et al., 1998). Nonnemaker, McNeely, Blum (2003) found that religiosity was a protective factor against cigarettes, alcohol, and marijuana use among adolescents. Wills, et al., (2003) found that religiosity was a protective factor for tobacco, alcohol, and marijuana use. Other studies have found similar results, in that religious affiliation was a protective factor for a host of substance use activities (Brown, Parks, Zimmerman, Phillips, 2001; Amey, et al., 1996; Miller, et al., 2000; Wallace & Forman, 1998). These findings seem to suggest that religion may act as a social control mechanism, in which adolescents are likely to avoid certain behaviors. It's also important to consider the fact that religion can promote the kinds of personal and social resources that, in turn, can function to inhibit substance use. Few studies have assessed political beliefs in relation to substance use. Wills et al., (2003) found that political beliefs were linked to substance use behaviors, with those who held liberal beliefs being more likely to use alcohol. Another study found that political beliefs were predictive of health related behaviors (Furnham, 1994).

Some inconsistent findings were observed in this study. For example, both males and females with high work schedules showed lower binge drinking. It may be that those who worked long hours had less opportunity to engage in binge drinking. In addition, political beliefs, religiosity, and work schedule did not affect lifetime marijuana use.

Since there are varying degrees of substance use and misuse among adolescents and the factors that contribute to such behavior, the generic approach in which everyone receives the same approach to alcohol treatment and prevention is likely unsuitable for effective behavior change. As such, conceptualization of substance use behavior and its prevention and treatment should include consideration of such key cultural and social factors such as religiosity, employment and political beliefs of adolescents. Implications of this study suggest that health education programs for adolescents regarding substance use should take into account the role of religion, employment, and political beliefs. First, research needs to address such questions as to whether strong religious adherence is a substitute for the high of substance use and/or social interactions and whether there is a personality difference between individuals who find religion important and those who do not. Secondly, the workplace may be an appropriate venue for establishing substance use prevention and early intervention programs focused on adolescent part-time workers. Additionally, it is important that parents, health educators, and policy makers continue to monitor after school employment hours. Lastly, future research needs to determine the mechanism in which political beliefs influence substance use. Such research will result in the kind of fine-tuning of existing prevention and intervention programs that will help reduce adolescent substance use.

Table 1. Gender Break Down for Alcohol, Tobacco,
and Other Drug Use Behaviors

                     Male          Female
Variable             [n (%)]       [n (%)]

Alcohol Lifetime
  No                 1271 (22.0)   1329 (21.0)
  Yes                4459 (78.0)   5009 (79.0)

Alcohol Past Month
  No                 2705 (47.1)   3418 (54.1)
  Yes                3038 (52.9)   2902 (45.0)

Binge Drink
  No                 3689 (64.7)   4811 (76.6)
  Yes                2013 (35.3)   1473 (23.4)

Cigarette Smoking
  No                 5145 (86.3)   5624 (86.7)
  Yes                 817 (13.7)    862 (13.3)

Marijuana Lifetime
  No                 2452 (48.9)   2591 (56.2)
  Yes                2472 (51.1)   2461 (43.8)

Cocaine Lifetime
  No                 5418 (91.7)   6046 (93.4)
  Yes                 492 (8.3)     426 (6.6)

Table 2. Odds Ratios and 95% Confidence Intervals for
Lifetime Alcohol Use by Gender

                         Males                 Females

Variable                 Odds    95%           Odds    95%
                         Ratio   Confidence    Ratio   Confidence
                                 Intervals             Intervals

Political Beliefs (1)
  Moderate               1.67    1.20-2.32 *   1.66    1.19-2.31
  Liberal                2.04    1.41-2.95 *   2.61    1.78-3.82 *
Religion (2)
  Pretty Important (T)   0.85    0.58-1.25     0.72    0.47-1.10
  Very Important (T)     1.32    1.23-1.45 *   1.27    1.18-1.40 *
Working (3)
  Few                    1.06    0.73-1.53     1.20    0.84-1.71
  Moderate               1.59    1.09-2.33 *   1.45    1.02-2.06 *
  High                   1.36    0.90-2.05     1.72    1.06-2.78 *

(1) The referent group for calculating odds ratios was conservative
views.

(2) The referent group for calculating odds ratios was little
importance.

(3) The referent group for calculating odds ratios was no after
school work.

(T) Inverse Relationship

* Significant

Table 3. Odds Ratios and 95% Confidence Intervals
for Current Alcohol Use by Gender

                         Males                 Females

Variable                 Odds    95%           Odds    95%
                         Ratio   Confidence    Ratio   Confidence
                                 Intervals             Intervals

Political Beliefs (1)
  Moderate               1.33    1.02-1.72 *   1.35    1.02-0.79 *
  Liberal                1.60    1.21-2.11 *   1.59    1.19-0.13 *
Religion (2)
  Pretty Important (T)   0.80    0.62-1.03     1.70    1.54-1.91 *
  Very Important (T)     1.40    1.31-1.51 *   1.31    1.24-1.40 *
Working (3)
  Few                    1.10    0.83-1.46     0.99    0.76-1.30
  Moderate               1.45    1.10-1.92 *   1.40    1.08-1.81 *
  High                   1.61    1.18-2.20 *   1.15    0.83-1.59

(1) The referent group for calculating odds ratios was conservative
views.

(2) The referet group for calculating odds ratios was little
importance.

(3) The referent group for calculating odds ratios was no after
school work.

(T) Inverse Relationship

* Significant

Table 4. Odds Ratios and 95% Confidence Intervals for
Binge Drinking by Gender

                         Males                 Females

Variable                 Odds    95%           Odds    95%
                         Ratio   Confidence    Ratio   Confidence
                                 Intervals             Intervals

Political Beliefs (1)
  Moderate               1.10    0.85-1.43     1.04    0.74-1.45
  Liberal                1.13    0.85-1.49     1.46    0.99-2.05
Religion (2)
  Pretty Important (T)   0.87    0.68-1.11     0.82    0.63-1.07
  Very Important (T)     1.45    1.35-1.59 *   1.34    1.25-1.46 *
Working (3)
  Few                    1.15    0.86-1.52     0.98    0.71-1.34
  Moderate               1.34    1.02-1.77 *   1.45    1.09-1.92 *
  High                   1.31    0.97-1.78     1.34    0.93-1.92

(1) The referent group for calculating odds ratios was conservative
views.

(2) The referent group for calculating odds ratios was little
importance.

(3) The referent group for calculating odds ratios was no after
school work.

(T) Inverse Relationship

* Significant

Table 5. Odds Ratios and 95% Confidence Intervals for
Current Smoking by Gender

                         Males                 Females

Variable                 Odds    95%           Odds    95%
                         Ratio   Confidence    Ratio   Confidence
                                 Intervals             Intervals

Political Beliefs (1)
  Moderate               1.31    0.90-1.91     1.37    0.85-2.23
  Liberal                1.67    1.14-2.45 *   2.00    1.23-3.25 *
Religion (2)
  Pretty Important (T)   0.99    0.72-1.36     0.76    0.54-1.06
  Very Important (T)     1.47    1.32-1.69 *   1.40    1.27-1.60 *
Working (3)
  Few                    1.34    0.86-2.09     0.92    0.59-1.44
  Moderate               2.50    1.69-3.70 *   1.63    1.13-2.40 *
  High                   3.34    2.22-5.02 *   2.37    1.54-3.63 *

(1) The referent group for calculating odds ratios was conservative
views.

(2) The referent group for calculating odds ratios was little
importance.

(3) The referent group for calculating odds ratios was no after
school work.

(T) Inverse Relationship

* Significant

Table 6. Odds Ratios and 95% Confidence Intervals for
Lifetime Marijuana Use by Gender

                         Males                Females

Variable                 Odds    95%          Odds    95%
                         Ratio   Confidence   Ratio   Confidence
                                 Intervals            Intervals

Political Beliefs (1)
  Moderate               0.46    0.23-1.91    0.86    0.42-1.76
  Liberal                0.75    0.40-1.58    1.37    0.65-2.87
Religion (2)
  Pretty Important (T)   0.86    0.45-1.65    0.58    0.29-1.15
  Very Important (T)     0.50    0.25-1.02    0.40    0.21-1.78
Working (3)
  Few                    0.60    0.29-1.25    0.82    0.40-1.69
  Moderate               0.74    0.35-1.56    1.81    0.95-3.46
  High                   1.46    0.61-3.50    1.33    1.08-3.02

(1) The referent group for calculating odds ratios was conservative
views.

(2) The referent group for calculating odds ratios was little
importance.

(3) The referent group for calculating odds ratios was no after
school work.

(T) Inverse Relationship

* Significant

Table 7. Odds Ratios and 95% Confidence Intervals for
Lifetime Cocaine Use by Gender

                         Males                 Females

Variable                 Odds    95%           Odds    95%
                         Ratio   Confidence    Ratio   Confidence
                                 Intervals             Intervals

Political Beliefs (1)
  Moderate               0.97    0.58-1.61     1.12    0.58-2.18
  Liberal                1.86    1.14-3.02 *   1.42    0.73-2.76
Religion (2)
  Pretty Important (T)   0.68    0.45-1.03     0.45    0.27-1.74
  Very Important (T)     1.33    1.19-1.56 *   1.30    1.17-1.53 *
Working (3)
  Few                    1.01    0.55-1.82     1.14    0.61-2.13
  Moderate               2.32    1.42-3.79 *   1.61    0.94-2.76
  High                   2.82    1.69-4.70 *   2.36    1.30-4.29 *

(1) The referent group for calculating odds ratios was conservative
views.

(2) The referent group for calculating odds ratios was little
importance.

(3) The referent group for calculating odds ratios was no after
school work.

(T) Inverse Relationship

* Significant

References

Amey, C. H., Albrecht, S. L., & Miller, M. K. (1996). Racial differences in adolescent drug use: The impact of religion. Substance Use and Misuse, 31, 1311-1332.

Bachman, J. G. & Schulenberg, J. (1993). How part-time work intensity relates to drug use, problem behavior, time use, and satisfaction among high school seniors. Developmental Psychology, 29, 220-235.

Bachman, J. G., Safron, D. J., Sy, S. R., & Schulenberg, J. E. (2003). Wishing to work: New perspectives on how adolescents' part-time work intensity is linked to educational disengagement, substance use, and other problem behaviors. International Journal of Behavioral Development, 27, 301-315.

Brown, T. L., Parks, G. S., Zimmerman, R. S., & Phillips, C. M. (2001). The role of religion in predicting adolescent alcohol use and problem drinking. Journal of Studies on Alcohol, 62, 696-705.

Brown, T. N., Schulenberg, J., Bachman, J. G., O'Malley, P. M., & Johnston, L. D. (2001). Are risk and protective factors for substance use consistent across historical time?: National data from high school classes of 1976-1997. Prevention Science, 2, 29-43.

Furnham, A. (1994). Explaining health and illness: Lay perceptions on current and future health, the causes of illness, and the nature of recovery. Social Science and Medicine, 39, 715-725.

Institute of Medicine. (1998). Protecting Youth at Work: Health, Safety, and Development of Working Children and Adolescents in the United States. Washington, DC.

Johnston, L. D., Bachman, J. G., O'Malley, P. M., & Schulenberg, J. E. Monitoring the future: A continuing study of American youth (12th grade survey), 2002 [computer file]. Conducted by University of Michigan, Institute for Social Research, Survey Research Center. ICPSR ed. Ann Arbor, MI: Inter-University Consortium for Political and Social Research, 2003.

Johnston, L. D., O'Malley, P. M., & Bachman, J. G. (2002). Monitoring the Future national results on adolescent drug use: Overview of key finding, 2001. (NIH Publication No. 02-5105). Bethesda, MD: National Institute on Drug Abuse.

Miller, L., Davies, M., & Greenwald, S. (2000). Religiosity and substance use and abuse among adolescents in the National Comorbidity Survey. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 1190-1197.

Miller, D. S., & Miller, T. Q. (1997). A test of socioeconomic status as a predictor of initial marijuana use. Addictive Behavior, 22, 479-489.

Nonnemaker, J. M., McNeely, C. A., & Blum, R. W. (2003). Public and private domains of religiosity and adolescent health risk behavior: Evidence from the National Longitudinal Study of Adolescent Health. Social Sconce and Medicine, 57, 2049-2054.

Paulson, R. L., Eppler, M. A., Satterwhite, T. N., Wuensch, K. L., Bass, L. A. (1998). Alcohol consumption, strength of religious beliefs, and risky sexual behavior in college students. Journal of American College Health, 46, 227-232.

Steinberg, L. D., & Dornbusch, S. M. (1991). Negative correlates of part-time employment during adolescence. Developmental Psychology, 27, 304-313.

Valois, F. V., Dunham, A. C., Jackson, A. B., Waller, J. (1999). Association between employment and substance abuse among public high school adolescents. Journal of Adolescent Health, 25, 256-263.

Wallace, J. M., & Forman, T. A. (1998). Religion's role in promoting health and reducing risk among American youth. Health Education and Behavior, 25, 721-741.

Wills, Yaeger, & Sandy (2003). Buffering effects of religiosity for adolescent substance use. Psychology of Addictive Behavior, 17, 24-31.

Author Note

Michael S. Dunn, Department of Public Health, East Tennessee State University.

Correspondence concerning this article should be addressed to Michael S. Dunn, Department of Public Health, East Tennessee State University, PO Box 70674, Johnson City, Tennessee 37614-0674. Phone: [phone]. E-mail: [e-mail]

Michael S. Dunn, East Tennessee State University