|
 |
|
|||||||
|
|||||||||
© 2003 American Public Health Association
RESEARCH AND PRACTICE |
The authors are with the Harvard School of Public Health and Social Behavior, Boston, Mass.
Correspondence: Requests for reprints should be sent to Henry Wechsler, 677 Huntington Ave, Room 704, Boston, MA 02115-6096 (e-mail: hwechsle{at}hsph.harvard.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Objectives. This study examined whether colleges with larger enrollments of students from demographic groups with lower rates of binge drinking exert a moderating effect on students from groups with higher binge drinking rates.
Methods. The study analyzed data from 114 colleges included in the 1993, 1997, 1999, and 2001 College Alcohol Study surveys.
Results. The binge drinking rates of White, male, and underage students were significantly lower in schools that had more minority, female, and older students. Students who do not binge drink in high school are more likely to start binge drinking at colleges with fewer minority and older students.
Conclusions. Student-body composition and demographic diversity should be examined by colleges wishing to reduce their binge drinking problems.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Heavy episodic or "binge" drinking has been recognized as a major public health problem on many American college campuses. A study sponsored by the National Institute on Alcohol Abuse and Alcoholism estimated that 1400 college students die each year from alcohol-related injuries.1 Several national studies have found that approximately 2 of every 5 college students are binge drinkers.2–7 Binge drinking has been associated with harms to the drinker as well as to others on campus through secondhand effects, including physical assaults, property damage, unwanted sexual advances, and disruptions of sleep and study.5,7–10
In recent years, much emphasis has been placed on normative influences on college student drinking behavior.11 Social learning theories12–14 stress the importance of the interaction and identification with as well as the imitation of the behavior of others in acquiring new and reinforcing old behaviors.
Although various interventions have been attempted to lower the level of binge drinking,11,15,16 to our knowledge, colleges have not yet examined housing and admissions policies and student demographics to that end. Yet binge drinking rates vary among student subgroups. African American and Asian, female, and older students have lower rates of binge drinking than do White, male, and younger students.5,8,17
Johnson and Hoffmann18 examined the relationship of cigarette smoking rates to the proportion of minority students. They found that as the percentage of racial/ethnic minority students increases, rates of smoking decrease among African American students, but rates do not decrease among White students. However, the schools studied were primarily composed of minority students.
The purpose of the present study was to examine whether an increased presence of students from groups in which alcohol is less heavily consumed tends to moderate the level of binge drinking in high–binge drinking demographic subgroups. Having more minority students, older students, and women should provide more models of abstention and responsible drinking and should lower the overall binge drinking rate.
In the present study, we hypothesized that (1) binge drinking rates of White, male, and underage students would be lower at schools with higher enrollments of minority, female, and older students; (2) White, male, and underage students who did not binge drink in high school would be less likely to take up binge drinking at schools with more minority, female, and older students; (3) White, male, and underage students who were binge drinkers in high school would be less likely to continue binge drinking at schools with more minority, female, and older students.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Study Population
The study used data from the 1993, 1997, 1999, and 2001 College Alcohol Study surveys.5,7,19,20 Of the 140 colleges surveyed in 1993, 120 participated in all 4 surveys. For each survey, the administrators at each college were asked to provide a random sample of undergraduates drawn from the total enrollment of full-time students. The attrition of 20 schools was primarily due to the inability of these colleges to provide a sample of students and mailing addresses to meet the time constraints of the survey. We excluded 1 school with a response rate that was substantially lower than that of the other schools, leaving 119 schools. Details of the sampling design have been published elsewhere.5,7,19,20
College response rates differed by school year: 52% in 2001 (range: 22% to 86%); 59% in 1999 (range: 27% to 83%); 59% in 1997 (range: 29% to 88%); and 70% in 1993 (range: 48% to 100%). However, response rates at individual colleges were not associated with binge rates at those schools. The Pearson correlation coefficient of the associations between a college’s binge rate and its response rate was –0.057 (P = .536) in 1993, 0.044 (P = .635) in 1997, 0.002 (P = .984) in 1999, and 0.170 (P = .064) in 2001.
Measures
Heavy episodic or binge drinking was defined as the consumption of at least 5 drinks in a row for men or 4 drinks in a row for women during the 2 weeks preceding their completion of the questions.7,21 A college’s binge drinking rate is the percentage of students classified as binge drinkers on the basis of the aggregated self-report responses of students at that school to the binge drinking questionnaire. We calculated the college binge drinking rates for total students and for high-risk subgroups (White, male, and underage [< 21 years] students).
In addition, to rule out the change in binge drinking rates that is attributable to the changes in demographic characteristics in each survey year, estimation of the binge rate in each survey year was standardized based on a direct standardization procedure over 8 strata (gender by 2 age groups [< 22 years vs others] by 2 ethnic groups [White vs others]), using each school’s true demographic characteristic in 1993 as the reference. We then not only could compare the binge rate over time, given the assumption that demographic characteristics remained constant, but also could reduce the potential selection bias in the prevalence rate of each survey year. The adjusted rates can therefore be reliably interpreted over time.
Demographic distributions at each school, including the percentages of minority, female, and older (aged 
22 years) students, were obtained from the Integrated Postsecondary Education Data System.21 School enrollment size was obtained from the Integrated Postsecondary Education Data System as well. We further categorized the enrollment size into small/medium schools and large schools. The small/medium schools are those schools whose student population is less than or equal to 10 000. The large schools are those schools whose student population is more than 10 000.
Data Analysis
We excluded from these analyses 2 historically Black colleges and 3 religious colleges that prohibit alcohol use. Thus, the college-level analyses included 114 colleges at 4 time points (n = 456). The total numbers of participating students in the 114 colleges were 10 446 in 2001, 13 396 in 1999, 13 846 in 1997, and 14 624 in 1993.
Pearson correlation coefficients were used to examine the univariate associations between the proportions of demographic characteristics and college binge drinking rates among total students and high-risk subgroups by each survey year. The correlation coefficients also were weighted by the school size. Both unweighted and weighted correlation coefficients were presented.
We also performed longitudinal multiple regression analyses to simultaneously examine the effects of the percentages of minority, female, and older (aged 22 years) students in the school, as well as 3 dummy variables indicating 1993, 1997, and 1999 survey years (i.e., 2001 being the comparison year). The General Estimating Equations approach23,24 with a working independence correlation structure was used to fit the longitudinal regression models to appropriately handle the longitudinal repeated measures of the college-level outcomes over 4 survey years. These regression models were stratified by school size (
10 000 vs > 10 000) to demonstrate how the associations vary by school size. All of the variables in the model were standardized with the mean equal to 0 and the standard deviation equal to 1 so that the resulting regression coefficients could be on the same scale as the correlation coefficients.
Multiple logistic regression was used to examine how these demographic characteristics predict the individual students’ binge drinking in college stratified by their drinking status in high school. Adjusted odds ratios and their 95% confidence intervals are reported. The General Estimating Equations approach was used to fit the logistic regression models.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
The mean college binge drinking rate was 44.4 ± 14.2% (range: 13.2% to 83.1%) for all students, 48.6 ± 14.3% (range: 18.6% to 85.9%) for White students, 49.8 ± 14.4% (range: 5.9% to 91.6%) for male students, and 43.1 ± 15.5% (range: 0% to 82.5%) for underage students. The average percentage of minority students at each school was 27.0 ± 18.3% (range: 2.8% to 82.2%), the average percentage of female students was 54.3 ± 11.8% (range: 20.6% to 100%), and the average percentage of students aged 22 years or older was 32.8 ± 15.1% (range: 4.3% to 76.9%).
The correlation between binge drinking rates and demographic distribution by each survey year is shown in Table 1
. The results indicate that the binge drinking rates for total students and for high-risk subgroups (White, male, and underage) at each school were significantly correlated with the demographic distribution of the school. The higher the percentage of minority, female, and older (aged 22 years) students in a school, the lower the binge drinking rates for total students and high-risk subgroups. These correlations did not vary significantly by survey year. The weighted correlations associated with the percentages of minority students and older students appeared slightly stronger than the unweighted correlations, whereas the weighted correlation associated with the percentage of female students tended to become slightly weaker.
|
In the stratified multiple regression analyses (Table 2
), the percentages of minority, female, and older students in school and the survey years were included in the models. The percentage of minority students was still significantly associated with lower binge drinking rates for total students and the high-risk subgroups after adjustment for other covariates except for the White binge rates in small/medium schools. The percentage of female students was significantly associated with binge drinking rates for total students and high-risk subgroups among small/medium schools except for male students. The percentage of female students had no significant effect on binge rates in large schools. This may have been because the gender distribution did not vary much among large schools. The percentage of students aged 22 years and older was also significantly associated with the binge drinking rates for total students and high-risk subgroups regardless of school size. In general, the binge drinking rate did not vary with the exception of lower binge rates for Whites in 1997 compared with 2001 in large schools and higher binge rates for underage students in all other survey years compared with 2001 in large schools.
|
The results in Table 3
indicate that high school nonbinge drinkers at colleges with higher percentages of minority and older students were significantly less likely to engage in binge drinking in college, even after adjustment for survey year, gender, underage, and race. Similarly, high school binge drinkers at colleges with higher percentages of minority and older students were found to be significantly less likely to continue binge drinking in college after adjustment for survey year, gender, underage, and race. The percentage of female students at each school was not significantly associated with binge drinking rates for both high school binge and nonbinge drinkers.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
The results strongly suggest that significant moderating effects accrue from the large-scale presence of lower-risk subgroups on the binge drinking of the high-risk subgroups. The college binge drinking rates among White, male, and underage students were significantly lower in schools with larger enrollments of minority, female, and older students.
In general, the moderating effects were significant for small/medium and large schools with the exception of the effect of the percentage of female students. The effect of percentage of female students was significant for the small/medium schools but not for the large schools. This finding may have been because there was not enough variation of gender distribution among large schools. (The mean female distribution was 58 ± 21% for small/medium schools and 51 ± 5% for large schools).
In addition, students who did not binge drink in high school were less likely to start binge drinking at colleges with larger enrollments of minority and older students, and students who were high school binge drinkers were less likely to continue drinking this way.
The findings may help to explain why fraternities and sororities and segregated freshmen dormitories that provide the highest concentrations of binge drinkers account for the bulk of alcohol problems on campus. Student-body composition, as well as the value of diversity at the college, organizational, and dormitory levels, should be considered by colleges wishing to reduce their binge drinking problems. Encouraging more older students to live on campus and in fraternity houses may be one practical application of these findings; another may be decreasing the heavy concentration of young, male, and White students in residential arrangements.
The results of this study must be viewed within the context of its limitations. First, the College Alcohol Study is subject to the limitations of self-report surveys. However, such surveys have been considered generally valid in examining alcohol responses.25,26 Second, potential bias may have been introduced through nonresponse. However, several procedures were used to test for potential bias from nonresponse in both surveys, and we found no effect on the findings. Furthermore, the binge drinking rates reported in this study are almost identical to those found in other national surveys.2–4,6
Finally, because this is a correlational study, cause-and-effect relationships cannot be inferred. Although 1 interpretation of the findings suggests that the presence of lower-drinking demographic groups may moderate drinking in higher-drinking groups, selfselection also may be operating. Colleges that have larger numbers of minority and older students and women may attract White, underage, and male students with different attitudes about drinking. However, we found that the presence of more women students, older students, and minority students had an effect on the rates of binge drinking among higher-risk-group students who did not binge drink in high school as well as among those who did. This finding makes self-selection on the basis of drinking attitudes a less plausible alternative explanation of the findings.
|
Acknowledgments |
|---|
This study was supported by a grant from the Robert Wood Johnson Foundation.
We gratefully acknowledge the advice of Steve Gortmaker, Harvard School of Public Health.
Human Participant Protection
No protocol approval was needed for this study
|
Footnotes |
|---|
Contributors
H. Wechsler conceptualized the study and interpreted the findings. M. Kuo helped to conceptualize ideas, conducted the analyses, and interpreted the findings. Both authors cowrote the article.
Accepted for publication November 30, 2002.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
1. Hingson R, Heeren T, Zakocs RC, Kopstein A, Wechsler H. Magnitude of alcohol-related mortality and morbidity among US college students ages 18–24. J Stud Alcohol.2002;63:136–144.[Web of Science][Medline]
2. Centers for Disease Control and Prevention. Youth risk behavior surveillance: National College Health Risk Behavior Survey United States. MMWR Morbid Mortal Wkly Rep.1997;46(SS-6):1–54.[Medline]
3. Johnston LD, O’Malley PM, Bachman JG. College Students and Young Adults. Rockville, Md: National Institute on Drug Abuse; 1997. National Survey Results on Drug Use From the Monitoring the Future Study, 1975–1995; vol 2. NIH publication 98-4140.
4. Substance Abuse and Mental Health Services Administration. Summary of Findings From 1999 National Household Surveys on Drug Abuse. Rockville, Md: Office of Applied Studies; 2000. National Household Survey on Drug Abuse series.
5. Wechsler H, Lee JE, Kuo M, Seibring M, Nelson T, Lee J. Trends in college binge drinking during a period of increased prevention efforts: finding from four Harvard School of Public Health College Alcohol Study surveys 1993–2001. J Am Coll Health.2002;50:203–222.[Web of Science][Medline]
6. Presley CA, Meilman PW, Cashin JR. 1992–1994 Alcohol and Drugs on American College Campuses: Use, Consequence, and Perceptions of the Campus Environment; Vol 4. Carbondale, Ill: The Core Institute; 1996.
7. Wechsler H, Davenport A, Dowdall G, Moeykens B, Castillo S. Health and behavioral consequences of binge drinking in college: a national survey of students at 140 campuses. JAMA.1994;272:1672–1677.
8. Wechsler H, Dowdall W, Davenport A, Castillo S. Correlates of college student binge drinking. Am J Public Health.1995;85:921–926.
9. Wechsler H, Moeykens B, Davenport A, Castillo S, Hansen J. The adverse impact of heavy episodic drinkers on other college students. J Stud Alcohol.1995;56:628–634.[Web of Science][Medline]
10. Perkins HW. Surveying the damage: a review of research on consequences of alcohol misuse in college populations. J Stud Alcohol.2002;14(suppl):91–100.
11. Perkins HW. Social norms and the prevention of alcohol misuse in collegiate contexts. J Stud Alcohol.2002;14(suppl):164–172.
12. Akers R. Deviant Behavior: A Social Learning Approach. Belmont, Calif: Wadsworth Publishing Co; 1977.
13. Bandura A. Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice Hall; 1986.
14. Hawkins J, Catalano R, Miller J. Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: implications for substance abuse prevention. Psychol Bull.1992;112:64–105.[Web of Science][Medline]
15. Larimer ME, Cronce JMP. Identification, prevention, and treatment: a review of individual-focused strategies to reduce problematic alcohol consumption by college students. J Stud Alcohol.2002;14(suppl):148–163.
16. Toomey TL, Wagenaar AC. Environmental policies to reduce college drinking: options and research findings. J Stud Alcohol.2002;14(suppl):193–205.
17. O’Malley PM, Johnston LD. Epidemiology of alcohol and other drug use among American college students. J Stud Alcohol.2002;14(suppl):23–29.
18. Johnson RA, Hoffmann JP. Adolescent cigarette smoking in US racial/ethnic subgroups: findings from the national education longitudinal study. J Health Soc Behav.2000;41:392–407.[Web of Science][Medline]
19. Wechsler H, Dowdall GW, Maenner G, Gledhill-Hoyt J, Lee H. Changes in binge drinking and related problems among American college students between 1993 and 1997. J Am Coll Health.1998;47:57–68.[Web of Science][Medline]
20. Wechsler H, Lee J, Kuo M, Lee H. College binge drinking in the 1990s: a continuing problem. Results of the Harvard School of Public Health 1999 College Alcohol Study. J Am Coll Health.2000;48:199–210.[Web of Science][Medline]
21. US Department of Education. Fall Enrollment Surveys. National Center for Education Statistics, Integrated Postsecondary Education Data System. Available at: http://nces.ed.gov/ipeds. Accessed October 10, 2003.
22. Wechsler H, Dowdall G, Davenport A, Rimm E. A gender-specific measure of binge drinking among college students. Am J Public Health.1995;85:982–985.
23. Zeger SL, Liang KY, Albert PS. Models for longitudinal data: a generalized estimating equation approach. Biometrics.1988;44:1049–1060.[Web of Science][Medline]
24. Liang KY, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrics.1992;73:12–22.
25. Cooper AM, Sobell MB, Sobell LC, Maisto SA. Validity of alcoholics’ self-reports: duration data. Int J Addict.1981;16:401–406.[Web of Science][Medline]
26. Midanik L. Validity of self report alcohol use: a literature review and assessment. Br J Addict.1988;83:1019–1030.[Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  D. M. Gorman, J. Mezic, I. Mezic, and P. J. Gruenewald Agent-Based Modeling of Drinking Behavior: A Preliminary Model and Potential Applications to Theory and Practice Am J Public Health, November 1, 2006; 96(11): 2055 - 2060. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. R Weitzman and Y.-Y. Chen Risk modifying effect of social capital on measures of heavy alcohol consumption, alcohol abuse, harms, and secondhand effects: national survey findings J Epidemiol Community Health, April 1, 2005; 59(4): 303 - 309. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |