Drug And Alcohol Prevention
Outcomes of a federally funded program for alcohol and other drug prevention in higher educationJohn C. Licciardone INTRODUCTION
Alcohol is the drug of choice at institutions of higher education in the United States. Throughout the 1990s, data from the Monitoring the Future Study (MTF) (1,2), the Core Alcohol and Drug Study (3-6), and the College Alcohol Study (7-9) indicate that alcohol use and binge drinking was widespread on college campuses. Bingeing has adverse consequences not only for the drinker but also for nonbingers (10). Binge drinking is also associated with greater use of cigarettes, marijuana, cocaine, and other illegal drugs (11).
Problems with drugs such as marijuana and cocaine prompted the U.S. Congress to pass the Anti-Drug Abuse Act of 1986, which lead to a cascade of federal legislation and regulations, including Drug-Free Schools and Campuses (12). The U.S. Department of Education subsequently administered the largest block of Federal funds devoted to alcohol and other drug (AOD) prevention programming in higher education, including the Fund for the Improvement of Postsecondary Education (FIPSE), which addressed campus prevention initiatives such as those mandated by congressional legislation (13). In fact, the institution-wide programs funded by FIPSE from the late-1980s to the mid-1990s represent a unique national experiment in collegiate AOD prevention programming.
To date, there has been little research on the effectiveness of programs in response to such legislation, particularly FIPSE-funded programs. A study of state education agencies found that about one-third of respondents did not conduct an evaluation during the school year (14). Another study of randomly selected students at 75 institutions nationwide concluded that information provided in compliance with the Drug-Free Schools and Communities Act Amendments was ineffective and costly (15).
With regard to FIPSE-funded institutions, a survey of 336 colleges and universities found that most campus coordinators perceived no change in AOD use or related problems as a result of prevention programming (16). Nevertheless, the latter study was limited in drawing conclusions about the effectiveness of FIPSE-funded AOD prevention programs, because its data were based strictly on the opinions of campus AOD coordinators, and there was no comparison group of non-FIPSE funded institutions. An earlier study involving only one FIPSE-funded institution and an in-state control institution reported an increased perceived risk associated with AODs and reduced levels of alcohol and marijuana use among FIPSE students (17). However, this study included methodologic limitations, and its findings cannot be extrapolated to other colleges and universities. Recently, an analysis of 104 institutions supported by FIPSE during the early-to mid-1990s demonstrated that binge drinking decreased at 60 institutions, while it increased at 34 institutions and was unchanged at 10 others (18). Other outcomes of AOD prevention programming were not studied in this analysis.
Although the number of FIPSE-funded AOD prevention programs has diminished substantially since the mid-1990s, the singular role that FIPSE played by being a catalyst for collegiate AOD prevention initiatives cannot be disputed. Consequently, assessment of the impact of FIPSE funding remains of interest not only for an historical perspective but also because it may guide the development or refinement of present-day AOD prevention efforts. The purpose of the present study was to provide a comprehensive outcomes analysis of the AOD prevention programs funded by FIPSE in the early- to mid-1990s.
METHODS
Nationwide Campuses Study
The Nationwide Campuses Study (NCS) was established with support from two FIPSE Analysis Project grants to provide data on the structures, processes, and outcomes of funded AOD prevention programs at colleges and universities across the nation during the early- to mid-1990s. The NCS database is unique because it links campus-specific data on AOD prevention program activities with pre- and postprogram outcome measures. The methodology used in acquiring the NCS data elements are described in greater detail below.
Participating Institutions of Higher Education
Eligible institutions included those colleges and universities that were awarded a two-year Institution-Wide Drug Prevention Program grant by FIPSE for 1990-1992 (1990 FIPSE cohort) or 1991-1993 (1991 FIPSE cohort). Each institution conducted its AOD prevention program according to evolving campus-specific needs rather than by following a standard protocol. Moreover, because these programs were aimed at all students, no control group was available at any of the institutions. Required institutional evaluations used baseline data at the beginning of the funding period and outcomes data at the end.
Study eligibility was limited to those institutions that used representative sampling methods to administer the Core Alcohol and Drug Survey during both pre- and postprogram evaluations (19). The Core Institute (Southern Illinois University at Carbondale) verified these inclusion criteria. A total of 207 institutions were funded during the study period, including 106 in the 1990 FIPSE cohort and 101 in the 1991 FIPSE cohort. Of these, 188 (90.8%) used the Core survey, 164 (79.2%) used representative sampling, and 97 (46.9%) used representative sampling both pre- and post AOD prevention programming (4,5). The latter included 49 institutions in the 1990 FIPSE cohort (4) and 48 in the 1991 FIPSE cohort (5).
Each of these 97 institutions was invited to participate in the study by providing access to its aggregate Core survey data. A total of 82 (84.5%) institutions agreed to participate, including 38 (77.6%) and 44 (91.7%) in the 1990 and 1991 FIPSE cohorts, respectively. Aggregate data from institutions using the Core survey, such as those in this study, have been consistent with those of the National Center for Education Statistics with regard to gender and race and ethnicity of the American college student population (6,20). This has been attributed to the adequacy of sampling methods and a mean response rate of 75% in administering the Core survey at FIPSE institutions (20).
The Core Alcohol and Drug Survey
The Core survey was developed in the late 1980s to meet the survey needs of FIPSE grantees (19,21). It is capable of producing data that may be validly compared with data from other major surveys, including the MTF study (19-21). The Core survey forms the basis for the largest existing database on AODs in postsecondary education (21). About 500,000 questionnaires were completed at more than 920 FIPSE-funded and non-FIPSE-funded institutions during the study period (19). The Core survey has been shown to be a valid and reliable instrument, with its psychometric properties in the areas of content-related validity, construct validity, test-retest reliability, and item reliability having been well established (3). The Core survey used in this study (4) was slightly revised from the original version (3).
Data Collection
During 1993 and 1994, data on FIPSE funding levels, student enrollment, and AOD prevention program elements were abstracted from grantee final reports submitted to FIPSE. Tabular data on inclusion of 72 possible program elements were reported according to six dimensions: student activities (29 possible elements), development and distribution of program materials (14 elements), curriculum infusion (7 elements), administrative response (10 elements), faculty and community activities (3 elements), and early intervention (9 elements). Agreement rates for each of the three coder pairings used for data abstraction of program elements were 81%, 80%, and 85%. Discordant coding was resolved by majority opinion. On a blinded control item, each of the three coders reported 100% use of the Core survey at participating institutions.
Core survey data were acquired directly from the Core Institute throughout 1995. Aggregate pre- and postprogram data for each institution provided information on student demographic variables, including year in college (classification), gender, race and ethnicity, marital status, work status, full- vs. part-time enrollment (student status), and on- vs. off-campus residence, and on 25 outcome variables in three major areas: perceptions of the campus environment with regard to AODs; AOD use; and consequences of AOD use.
Crude data from dichotomous survey items were used without transformation. Four items addressing perceptions of the campus environment (e.g., "'Does your campus have drug and alcohol policies?") had "yes," "no," and "don't know" responses. For these items, "don't know" generally reflected a lack of awareness or knowledge of the campus environment. Thus, "don't know" responses were transformed to "no" responses for these items. The remaining polychotomous items were transformed by using weighting techniques to yield essentially continuous data (22). Duplicate sets of Core survey data were entered and transformed by using independent coders for each institution (23). The overall agreement rate between the duplicate sets was 98%, with agreement rates of 99%, 97%, and 98% for each of the three coder pairings. Discordant data were resolved by the author.
Statistical Analysis
Descriptive statistics were used to summarize baseline student characteristics and institutional AOD prevention program elements. Because Analysis Project grants funded by FIPSE did not support collection of data from non-FIPSE-funded institutions, data from a contemporaneous national referent group were subsequently used to adjust for temporal trends in AOD use when evaluating the performance of FIPSE-funded programs (24). The MTF study provided the most appropriate referent group because of its random sampling of college students nationwide (1) and because the Core survey was specifically designed to enable comparisons between the two surveys (19-21).
The MTF study uses a three-stage random sampling procedure to secure a nationwide sample of high school seniors; follow-up samples subsequently generate a national representation of college students (1). College students are defined as "those follow-up respondents one to four years past high school who say they were registered as full-time students in a two- or ]bur-year college at the beginning of March in the year in question" (1). Thus, in any given calendar year, the MTF study will survey a representative panel of college students, including comparable numbers in the first, second, third, and fourth year of college. As a cohort graduates from college, it is replaced by a cohort of entering students, thus ensuring that a representative sample of students is surveyed and available for comparison with FIPSE students. Further, response rates in the MTF study (79% in the first year post-high school) (1) are comparable with those of the Core survey (20).
A contemporaneous comparison of the baseline (preprogram) characteristics of 41,567 FIPSE students (18,796 and 22,771 students in the 1990 and 1991 FIPSE cohorts, respectively) with 2810 referents in the MTF study (1400 and 1410 students in the 1990 and 1991 panels, respectively) was conducted. Because an inception cohort of individual students was not assembled and followed over time as part of the FIPSE evaluation strategy, incidence rates and relative risks of AOD outcomes could not be measured. Therefore, cross-sectional comparisons between FIPSE and MTF students were summarized by using odds ratios (OR) and 95% confidence intervals (CI). Comparisons included the prevalences of alcohol, tobacco, marijuana, and cocaine use, and binge drinking. Annual prevalence was defined as having used alcohol, or the drug in question, at least once in the last year. Similarly, 30-day prevalence consisted of any use within the last 30 days and daily prevalence involved use in each of the last 30 days. Prevalent binge drinking was defined as five or more drinks at a sitting within the last two weeks. Data for 30-day and daily prevalences were not available for the 1990 FIPSE cohort. Moreover, summary measures could not be computed for daily prevalence of cocaine use because the MTF study did not report the necessary data (1).
Institutional program effects were measured by comparing student responses on the pre- and post-AOD prevention program Core surveys. Effect sizes were computed by adapting Glass's A to obtain a common metric for comparison across all variables (25). This statistic was computed as the difference of the mean post- and preprogram measures divided by the standard deviation of the preprogram measure. Thus, the effect size represents a standardized change score. Zero indicates no change over the two-year AOD prevention program period, while a positive score indicates an increase in the measure over time and a negative score indicates a decrease over time. Because there was a very high correlation between the mean unweighted and sample-size-weighted effect sizes (p = 0.97, P < 0.001), only unweighted effect sizes were analyzed and reported. These generally have been shown to be valid and, in some cases, preferable to weighted effect sizes (26). Effect sizes were not adjusted for demographic factors because representative sampling was used on all pre- and post-program surveys.
The expected postprogram prevalences of AOD use among 39,197 FIPSE students (16,936 and 22,261 students in the 1990 and 1991 FIPSE cohorts, respectively) were computed by using preprogram data and adjusting for two-year trends in prevalences of AOD use reported in the MTF study (1). To be contemporaneous with the pre- (1990 and 1991) and postintervention (1992 and 1993) measures used in the FIPSE cohorts, MTF two-year trends from 1990 to 1992 and from 1991 to 1993 were used for adjustment. There were 2980 students in the 1992 and 1993 MTF panels (1490 students in each panel). Expected prevalences of AOD use and nonuse among FIPSE students were then compared with the observed postprogram prevalences by using 2 x 2 contingency tables. Figure 1 summarizes the selection of FIPSE institutions and students and the analytic methods used to measure program effects.
[FIGURE 1 OMITTED]
Multiple linear regression was used to study associations between intensity of AOD prevention programming in each of the six dimensions and AOD use and its consequences. Independent variables included the number of institutional elements within each program dimension and the dependent variables included institutional effect sizes. All six dimensions were analyzed simultaneously to determine the effect of each dimension, independent of the other five. [3 coefficients and standard errors were used to assess statistical significance. All hypotheses were tested at the 0.05 level of statistical significance by using the SYSTAT software package (SPSS Inc., Chicago, IL).
RESULTS
FIPSE Institutions and Students
The study included institutions in 36 states with the following geographical distribution: South, 27 (32.9%); Northeast, 25 (30.5%); Northcentral, 17 (20.7%); and West, 13 (15.9%). The greatest number of institutions were located in Massachusetts (9.8%), Missouri (6.1%), and Texas (6.1%). Mean FIPSE funding for the two-year grant period was $112,977-+$33,683 (median, $107,910). Mean student enrollment, including both full- and part-time students, was 7629 [+ or -] 8370 (median, 4460).
The baseline characteristics of FIPSE students are presented in Table 1. The gender (42.9% male) and race (17.7% nonwhite) distributions were similar to those of the general population of higher education students reported by the Census Bureau for 1990 and 1991 (45.4% male and 22.9% nonwhite) (27). However, FIPSE students were much more likely to be attending classes full-time (87.7%) than students reported by the Census Bureau (56.6%) (27). Perceptions of the campus environment included a general awareness of the existence of AOD policies and the belief that campuses were concerned about the prevention of AOD use. A minority of students were aware of the existence of a campus AOD prevention program or believed that AOD policies were enforced. Few students were actively involved in campus AOD prevention efforts. Tobacco and alcohol use, including binge drinking, were commonly reported. The most frequently reported consequences of AOD use during the last year were having a hangover, nausea or vomiting, and driving a car while under the influence.
A comparison of the baseline prevalences of AOD use among FIPSE students with MTF referents is presented in Table 2. The FIPSE students reported significantly lower daily prevalences of alcohol (OR, 0.11; 95% CI, 0.08-0.16), tobacco (OR, 0.84; 95% CI, 0.71-0.99), and marijuana (OR, 0.29; 95% CI, 0.18-0.46) use, as well as a lower two-week prevalence of binge drinking (OR, 0.92; 95% CI, 0.85-0.99). The FIPSE students also reported significantly lower annual prevalence of alcohol use (OR, 0.70; 95% CI, 0.62-0.79); however, they reported significantly higher annual (OR, 1.18; 95% CI, 1.09-1.28) and 30-day (OR, 1.33; 95% CI, 1.17-1.52) prevalences of tobacco use.
AOD Prevention Programs and Outcomes
Institutional AOD prevention programming is summarized in Table 3. There was considerable variation in programming across institutions. The most commonly reported program elements included development and distribution of flyers or brochures (95.1%), other literature dissemination (90.2%), educational presentations (86.6%), alcohol awareness week activities (84.1%), faculty and staff training (84.1%), alcohol-free parties (82.9%), peer education programs (81.7%), new student orientation presentations (81.7%), and administrative AOD policy reviews (81.7%).
Effect sizes for demographic and AOD outcome variables are presented in Table 1. Effect sizes for demographic variables confirm the adequacy of sampling methods. The only statistically significant demographic effect was an increase in nonwhite students (P < 0.001). This finding is consistent with national trends in nonwhite college enrollment reported by the Census Bureau for 1990 through 1992 and for 1991 through 1993 (27) (mean [DELTA] = 0.062 vs. 0.073 observed in the study). Favorable changes in perceptions of the campus environment included an increased awareness of campus AOD prevention programs (P < 0.001) and increased student involvement in prevention efforts (P = 0.03). However, HPSE students reported an increased desire for availability and use of drugs at parties in and around campus (P < 0.001). The HPSE students also reported increased frequencies of tobacco (P = 0.002), marijuana (P=0.001), and cocaine (P = 0.002) use. Other highly significant findings were increased frequencies of arrests for driving while intoxicated or under the influence (P < 0.001) and of poor performance on a test or important project (P = 0.003). The frequencies of several other negative consequences of AOD use also increased sufficiently to achieve statistical significance (Table 1). After controlling for the number of students enrolled at each institution and their demographics, the only significant association with the amount of FIPSE funding was a decrease in trouble with police, residence hall, or other college authorities (P = 0.05).
Adjusted AOD use outcomes of prevention programs at FIPSE institutions are presented in Table 2. These demonstrate significantly increased prevalences of marijuana and cocaine use and significantly decreased prevalences of tobacco use for all annual, 30-day, and daily measures. The prevalences of alcohol use and binge drinking likewise were increased; however, only the annual and 30-day prevalences of alcohol use achieved statistical significance. The most notable effect of adjusting for time trends in AOD use was to identify significant decreases in tobacco use among FIPSE students that otherwise would have been inapparent.
Outcomes associated with AOD prevention program dimensions are presented in Table 4. Only 5 of the 120 contrasts were statistically significant. To better discern whether superior outcomes were associated with specific AOD prevention program dimensions, contrasts that approached significance (0.06 [less than or equal to] P [less than or equal to] 0.10) and the direction of the contrasts also were inspected. Several findings subsequently emerged from this process. Favorable outcomes were associated with three of the six dimensions: curriculum infusion, administrative response, and faculty and community activities. For these three dimensions combined, 53 of 60 contrasts, all 4 nearly significant contrasts and all 5 significant contrasts were favorable. Administrative response was associated with significant decreases in frequency of alcohol use (P = 0.02), missed classes (P = 0.02), and hangovers (P = 0.05), while curriculum infusion was associated with significant decreases in the desire for drugs at parties in and around campus (P = 0.01) and in driving a car while under the influence (P = 0.05). For development and distribution of program materials, student activities, and early intervention, 51 of 60 contrasts, 6 of 7 nearly significant contrasts, and both significant contrasts were unfavorable. Greater intensity in development and distribution of program materials was associated with significant increases in frequency of alcohol use (P = 0.05) and nausea or vomiting (P = 0.05).
DISCUSSION
This study illustrates the gravity of AOD use and its consequences in higher education during the early- to mid-1990s. On average, there were 1.3 binge drinking episodes per student in the last two weeks at the FIPSE-funded institutions, as well as 3.1 hangovers per student in the last year and 1.4 incidents of driving while under the influence per student in the last year (Table l). Despite these findings, the data suggest that students at FIPSE-funded institutions actually represented a relatively low-risk population with regard to AOD use, particularly daily alcohol and marijuana use (Table 2).
Although FIPSE-funded programs varied considerably across institutions (Table 3), they were very effective in promoting awareness of their campus AOD prevention activities. However, there were relatively few favorable outcomes attributable to these programs (Table 1). The increased desire for availability and use of drugs at parties in and around campus and increased arrests for driving while intoxicated or under the influence are disturbing. The latter may have been a consequence of stronger enforcement efforts, however, because there was a trend toward decreased driving while under the influence. Nevertheless, the increased frequencies of marijuana and cocaine use suggest that students acted on their stated desire for drugs. Moreover, increased marijuana and cocaine use persisted even after adjusting for temporal trends in AOD use (Table 2). The only favorable adjusted outcomes involved tobacco use. Curriculum infusion, administrative response, and faculty and community activities were most clearly associated with favorable outcomes, while development and distribution of program materials, student activities, and early intervention were most consistently associated with unfavorable outcomes (Table 4). Given the fiscal crisis in higher education and challenges to institutionalization of AOD prevention programming after FIPSE funding (28), it is questionable whether longer-term AOD outcomes would have been more favorable than observed in these FIPSE cohorts during the study period.
It is difficult to explain why FIPSE funding would negatively impact AOD prevention efforts, particularly as suggested by the findings relevant to illegal drugs (Tables 1 and 2). Aside from potential methodologic limitations, at least two other possibilities exist. First, it is possible that FIPSE programs were successful in helping to retain AOD-using students, including illegal drug users, who would have otherwise dropped out. The retention of such students and their participation in the post-program surveys would tend to result in less favorable outcomes than would have been observed if they had dropped out and were not included in the surveys. Second, it is possible that FIPSE-funded AOD prevention programming was aimed at the more prevalent alcohol and tobacco users rather than other illegal drug users. Nevertheless, given the large differential in AOD prevention program funding between FIPSE and non-FIPSE funded institutions (15), even a small proportion of prevention efforts at FIPSE institutions targeted at illegal drug users would likely have equaled or surpassed, in magnitude, such programming at non-FIPSE-funded institutions.
Several comments about the study methodology warrant mention. First, students at FIPSE-funded institutions were generally at lower risk of AOD use than students at non-FIPSE-funded institutions (Table 2). With over 3500 institutions of higher education nationwide during the study period (27), only a relative few were able to compete successfully for FIPSE funding. Students at high risk of AOD use and consequences may not reflect the student profiles of these select colleges and universities that received FIPSE funding. Therefore, discrepant AOD use among students at FIPSE-funded and non-FIPSE-funded institutions is a manifestation of the inherent selection bias introduced by FIPSE-funding decisions that cannot be fully resolved by the study methodology or statistical adjustments. However, it is important to note that FIPSE students were comparable with college students nationwide with regard to gender and race and ethnicity (6,20,27). Also, because close to 90% of FIPSE students attended classes full-time, there was little bias introduced by using full-time MTF students as referents. The 79% and 75% response rates among MTF and FIPSE students, (1,20) respectively, on surveys containing sensitive items, further support the validity of study findings.
Second, this was an ecologic study having available for analysis only institution-level data for 82 colleges and universities rather than individual-level data on the hundreds of thousands of students attending these institutions. A limitation of ecologic studies is that an observed association between two variables at the aggregate level may not hold at the individual level because other extraneous factors may affect the association. The latter phenomenon, when it occurs, is known as the ecological fallacy and limits the use of ecologic studies in testing etiologic hypotheses (29). Nevertheless, ecologic studies can be useful in evaluating the impact of intervention programs on outcomes in a target population and ecologic studies are enhanced by comparing temporal trends in outcomes among different study groups (29), as done in this study.
Third, because AOD prevention programming was mandated by congressional legislation during the study period, there exists the possibility of contamination bias when selecting referents from any American college student population. This is unlikely to be a significant problem because the median amount spent in complying with such legislation was less than $5000 (15), whereas median FIPSE funding was $107,910. Fourth, adjusted probabilities were not used to maintain a study-wide statistical significance level of 0.05 when testing multiple hypotheses (30) (Table 4). Recent debate appears to favor Bayesian approaches rather than adjustment for multiple comparisons (31-34). Nevertheless, use of such adjustments would not have materially altered the major conclusions of this study (35).
This study evaluated the outcomes of AOD prevention programs during peak years of FIPSE funding. The possibility exists that such programs experienced a learning curve in providing effective AOD prevention services and that later FIPSE-funded programs may have experienced better outcomes than observed in this study. In addition, "success stories" at particular FIPSE-funded institutions may have been obscured by the overall findings. For example, at one institution, a highly successful student competition to develop an innovative alcohol education strategy involved 1200 students and realized a number of positive outcomes (36). Another campus attributed improvements in perceived risk of AODs and self-reported use of alcohol and marijuana to a self-regulation theory-based approach to primary and secondary prevention (17). At a public residential campus, a media campaign designed to change student perceptions of the amount of binge drinking decreased the number of students who perceived binge drinking as the norm and resulted in reduced self-reported binge drinking (37). This social norms approach continues to be promoted in the development of college AOD prevention programs (38), although recent findings suggest the need for more specific tailoring of these programs to such factors as the stage of initiating heavy drinking (39). The social norms approach also is part of a larger "proactive" view of AOD prevention programming that aims to impact the college environment, thereby complementing the "reactive" response to students needing assistance with AOD-related problems (40).
Tobacco use has been identified as a serious problem among college students (41). In contrast to the generally limited impact that FIPSE funding had on short-term outcomes relative to AOD use and its consequences, FIPSE students reported decreased annual, 30-day, and daily prevalences of tobacco use. In addition to the theories and success stories noted above, perhaps there are lessons to be learned by a closer inspection of FIPSE programs aimed at tobacco use in higher education. Finally, the results of this study also suggest that AOD prevention programmers and researchers should examine more thoroughly the use of curriculum infusion, administrative responses, and faculty and community activities in higher education.
Table 1. Baseline characteristics of and standardized changes
among students at FIPSE-funded AOD prevention
programs. *, ([dagger])
Variable Core survey Baseline
item characteristics
([double dagger]) Mean SD
Demographic characteristics
Classification (y of 1 2.0 0.7
postsecondary education)
Gender (% male) 5 42.9 11.4
Ethnic origin (% nonwhite) 3 17.7 16.9
Marital status (% married) 4 12.7 12.3
Work status (0 = not working, 7 35.3 13.6
100 = working full-time)
Student status (% attending 11 87.7 14.7
full-time)
Residence (% residing 6 41.7 30.7
on-campus)
Perception of the campus environment (% responding "yes")
Campus has AOD policies 12a 79.0 16.5
Campus AOD policies 12b 44.5 19.1
are enforced
Campus has an AOD 12c 42.1 15.5
prevention program
Campus is concerned about 12d 74.9 12.8
prevention of AOD use
Student is involved in campus 12e 7.3 4.0
AOD prevention efforts
Student desires availability 10 67.0 14.8
and use of alcohol at parties
in and around campus
Student desires availability 10 12.4 7.9
and use of drugs at parties
in and around campus
AOD use
Alcohol 17b 14.7 5.1
(% of days used in last y)
Binge drinking (no. of 14 1.3 0.6
episodes in last two wk)
Tobacco 17a 18.8 8.3
(% of days used in last y)
Marijuana 17c 2.5 2.0
(% of days used in last y)
Cocaine 17d 0.1 0.3
(% of days used in last y)
Consequences of AOD use (no. of episodes in last y)
Had a hangover 21a 3.1 1.1
Performed poorly on a test or 21b 0.6 0.3
important project
Been in trouble with police, 21c 0.2 0.2
residence hall, or other
college authorities
Damaged property, pulled fire 21d 0.2 0.1
alarm, etc.
Got into an argument or fight 21e 0.9 0.4
Got nauseated or vomited 21f 1.5 0.5
Driven a car while under 21g 1.4 0.5
the influence
Missed a class 21h 1.1 0.6
Been criticized by a 21i 0.9 0.3
known person
Had a memory loss 21k 0.9 0.4
Did something that was 211 1.2 0.4
later regretted
Been arrested for DWI or DUI 21m 0.03 0.03
Been hurt or injured 21s 0.3 0.2
Variable Standardized changes
(effect size)
Mean SE P
Demographic characteristics
Classification (y of 0.050 0.064 0.44
postsecondary education)
Gender (% male) 0.004 0.014 0.78
Ethnic origin (% nonwhite) 0.073 0.015 <0.001
Marital status (% married) 0.027 0.030 0.37
Work status (0 = not working, 0.016 0.024 0.51
100 = working full-time)
Student status (% attending -0.059 0.033 0.08
full-time)
Residence (% residing 0.033 0.049 0.50
on-campus)
Perception of the campus environment (% responding "yes")
Campus has AOD policies 0.019 0.023 0.41
Campus AOD policies 0.052 0.027 0.06
are enforced
Campus has an AOD 0.201 0.033 <0.001
prevention program
Campus is concerned about -0.005 0.018 0.78
prevention of AOD use
Student is involved in campus 0.031 0.014 0.03
AOD prevention efforts
Student desires availability -0.020 0.016 0.21
and use of alcohol at parties
in and around campus
Student desires availability 0.055 0.015 <0.001
and use of drugs at parties
in and around campus
AOD use
Alcohol 0.000 0.015 0.99
(% of days used in last y)
Binge drinking (no. of 0.013 0.016 0.42
episodes in last two wk)
Tobacco 0.038 0.012 0.002
(% of days used in last y)
Marijuana 0.112 0.034 0.001
(% of days used in last y)
Cocaine 0.222 0.068 0.002
(% of days used in last y)
Consequences of AOD use (no. of episodes in last y)
Had a hangover -0.020 0.015 0.19
Performed poorly on a test or 0.046 0.015 0.003
important project
Been in trouble with police, 0.048 0.024 0.05
residence hall, or other
college authorities
Damaged property, pulled fire 0.049 0.023 0.04
alarm, etc.
Got into an argument or fight 0.015 0.016 0.35
Got nauseated or vomited 0.012 0.016 0.46
Driven a car while under -0.020 0.012 0.10
the influence
Missed a class 0.072 0.033 0.03
Been criticized by a 0.029 0.013 0.03
known person
Had a memory loss 0.035 0.015 0.02
Did something that was -0.011 0.015 0.47
later regretted
Been arrested for DWI or DUI 0.068 0.018 <0.001
Been hurt or injured 0.029 0.019 0.13
* Baseline characteristics were based on 41,567 preprogram student
responses to the Core Alcohol and Drug Survey, by using the 82
participating institutions as the units of analysis. Standardized
changes were based on the corresponding institutional pre- and
post-program measures.
([dagger]) FIPSE denotes Fund for the Improvement of
Postsecondary Education; AOD, alcohol and other drug:
DWI, driving while intoxicated; DUI, driving while
under the influence.
([double dagger]) Responses to Core Alcohol and Drug
Survey items. [From Ref. (4)].
Table 2. Baseline AOD use and adjusted prevention program
outcomes at FIPSE institutions. *, ([dagger])
Baseline AOD use Adjusted prevention
program outcomes
Variable OR 95% CI OR 95% CI
Alcohol use
Annual prevalence 0.70 0.62-0.79 1.20 1.16-1.25
([double dagger])
30-day prevalence 0.92 0.81-1.04 1.16 1.11-1.21
([double dagger])
Daily prevalence 0.11 0.08-0.16 1.24 0.93-1.66
([double dagger])
Binge drinking prevalence 0.92 0.85-0.99 1.02 0.99-1.05
([double dagger])
Tobacco use
Annual prevalence 1.18 1.09-1.28 0.93 0.91-0.96
30-day prevalence 1.33 1.17-1.52 0.96 0.92-1.00
Daily prevalence 0.84 0.71-0.99 0.87 0.82-0.92
Marijuana use
Annual prevalence 0.83 0.76-0.90 1.05 1.02-1.09
30-day prevalence 0.97 0.83-1.14 1.10 1.04-1.16
Daily prevalence 0.29 0.18-0.46 1.61 1.25-2.08
Cocaine use([subsection])
Annual prevalence 0.91 0.75-1.09 1.25 1.15-1.36
30-day prevalence 1.23 0.70-2.21 1.24 1.00-1.54
* Baseline AOD use was based on a comparison of students at FIPSE
institutions with Monitoring the Future referents. Prevention program
outcomes were based on comparisons of postprogram AOD use relative to
preprogram use, adjusted for contemporaneous two-year trends in AOD
use in the Monitoring the Future study. [From Ref. (1).]
([dagger]) AOD denotes alcohol and other drug; FIPSE, Fund for the
Improvement of Postsecondary Education; OR, odds ratio; CI, confidence
interval.
([double dagger]) Annual prevalence was defined as any use in the last
year; 30-day prevalence was any use in the last 30 days; daily
prevalence was use in each of the last 30 days; binge drinking
prevalence was defined as any episode of five or more drinks at a
sitting within the last 2 weeks.
([subsection]) Data for daily prevalence of cocaine use were not
available.
Table 3. Alcohol and other drug prevention program
elements at FIPSE institutions according to
dimension. *, ([dagger])
No. of program elements
Program dimension Minimum Maximum Mean SD
Student activities 1 26 16.4 4.7
Development and distribution 0 10 4.0 2.0
of program materials
Curriculum infusion 0 5 1.8 1.1
Administrative response 1 9 5.1 2.0
Faculty and community activities 0 3 2.1 0.9
Early intervention 0 9 3.8 2.3
Total for all dimensions 10 52 33.2 8.0
* FIPSE denotes Fund for the Improvement of Postsecondary Education.
([dagger]) The maximum possible elements within each dimension were
student activities. 29; development and distribution of program
materials, 14; curriculum infusion, 7; administrative response,
10; faculty and community activities, 3; and early intervention,
9. Means (SD) were based on the 82 FIPSE institutions.
Table 4. Outcomes associated with AOD prevention programs at FIPSE
institutions according to dimension. *, ([dagger])
AOD prevention program dimension
Development and
Student distribution of
activities program materials
Dependent variable [beta] [+ or -] SE [beta] [+ or -] SE
Perceptions of the
campus environment
Student desires 0.001 [+ or -] 0.004 0.013 [+ or -] 0.009
availability and use
of alcohol at
parties in and
around campus
Student desires 0.006 [+ or -] 0.004 0.001 [+ or -] 0.008
availability and use
of drugs at parties
in and around campus
AOD use
Alcohol 0.003 [+ or -] 0.004 0.016 [+ or -] 0.008
([double dagger])
Binge drinking 0.002 [+ or -] 0.004 0.007 [+ or -] 0.009
Tobacco 0.005 [+ or -] 0.003 0.002 [+ or -] 0.006
Marijuana 0.000 [+ or -] 0.009 -0.021 [+ or -] 0.018
Cocaine 0.032 [+ or -] 0.018 0.011 [+ or -] 0.037
([section])
Consequences of AOD use
Had a hangover 0.004 [+ or -] 0.004 0.012 [+ or -] 0.008
Performed poorly on a 0.005 [+ or -] 0.004 0.008 [+ or -] 0.008
test or important
project
Been in trouble with 0.002 [+ or -] 0.006 0.012 [+ or -] 0.013
police, residence
hall, or other
college authorities
Damaged property, 0.002 [+ or -] 0.006 0.019 [+ or -] 0.012
pulled fire alarm,
etc.
Got into an argument -0.001 [+ or -] 0.004 0.012 [+ or -] 0.009
or fight
Got nauseated or 0.001 [+ or -] 0.004 0.017 [+ or -] 0.009
vomited ([doble dagger])
Driven a car while 0.003 [+ or -] 0.003 0.008 [+ or -] 0.006
under the influence
Missed a class 0.005 [+ or -] 0.008 0.025 [+ or -] 0.017
Been criticized by a 0.002 [+ or -] 0.003 0.009 [+ or -] 0.007
known person
Had a memory loss 0.001 [+ or -] 0.004 0.009 [+ or -] 0.008
Did something that was -0.003 [+ or -] 0.004 0.014 [+ or -] 0.008
later regretted ([section])
Been arrested for DWI 0.001 [+ or -] 0.005 -0.003 [+ or -] 0.010
or DUI
Been hurt or injured -0.003 [+ or -] 0.005 0.019 [+ or -] 0.010
([section])
AOD prevention program dimension
Curriculum Administrative
infusion response
Dependent variable [beta] [+ or -] SE [beta] [+ or -] SE
Perceptions of the
campus environment
Student desires -0.009 [+ or -] 0.016 0.000 [+ or -] 0.009
availability and use
of alcohol at
parties in and
around campus
Student desires -0.039 [+ or -] 0.015 -0.011 [+ or -] 0.008
availability and use ([double dagger])
of drugs at parties
in and around campus
AOD use
Alcohol -0.013 [+ or -] 0.015 -0.019 [+ or -] 0.008
([double dagger])
Binge drinking -0.018 [+ or -] 0.017 -0.007 [+ or -] 0.009
Tobacco -0.022 [+ or -] 0.012 -0.003 [+ or -] 0.007
([section])
Marijuana -0.009 [+ or -] 0.035 -0.009 [+ or -] 0.019
Cocaine -0.016 [+ or -] 0.070 -0.015 [+ or -] 0.038
Consequences of AOD use
Had a hangover -0.027 [+ or -] 0.015 -0.017 [+ or -] 0.008
([section]) ([double dagger])
Performed poorly on a -0.007 [+ or -] 0.015 -0.013 [+ or -] 0.008
test or important
project
Been in trouble with -0.021 [+ or -] 0.025 -0.019 [+ or -] 0.014
police, residence
hall, or other
college authorities
Damaged property, -0.012 [+ or -] O.024 -0.009 [+ or -] 0.013
pulled fire alarm,
etc.
Got into an argument -0.004 [+ or -] 0.016 0.004 [+ or -] 0.009
or fight
Got nauseated or -0.032 [+ or -] 0.017 -0.009 [+ or -] 0.009
vomited ([section])
Driven a car while -0.024 [+ or -] 0.012 -0.006 [+ or -] 0.007
under the influence ([double dagger])
Missed a class -0.046 [+ or -] 0.032 -0.041 [+ or -] 0.018
([dagger])
Been criticized by a -0.014 [+ or -] 0.013 -0.008 [+ or -] 0.007
known person
Had a memory loss -0.022 [+ or -] 0.015 -0.014 [+ or -] 0.008
Did something that was -0.013 [+ or -] 0.015 -0.008 [+ or -] 0.008
later regretted
Been arrested for DWI 0.015 [+ or -] 0.018 0.005 [+ or -] 0.010
or DUI
Been hurt or injured -0.024 [+ or -] 0.020 -0.015 [+ or -] 0.011
AOD prevention program dimension
Faculty and
community Early
activities intervention
Dependent variable [beta] [+ or -] SE [beta] [+ or -] SE
Perceptions of the
campus environment
Student desires -0.008 [+ or -] 0.020 -0.001 [+ or -] 0.008
availability and use
of alcohol at
parties in and
around campus
Student desires -0.014 [+ or -] 0.018 0.006 [+ or -] 0.007
availability and use
of drugs at parties
in and around campus
AOD use
Alcohol -0.027 [+ or -] 0.018 0.012 [+ or -] 0.007
([section])
Binge drinking -0.019 [+ or -] 0.021 0.006 [+ or -] 0.008
Tobacco 0.013 [+ or -] 0.015 -0.010 [+ or -] 0.006
(section])
Marijuana -0.014 [+ or -] 0.043 -0.008 [+ or -] 0.016
Cocaine 0.005 [+ or -] 0.086 0.003 [+ or -] 0.033
Consequences of AOD use
Had a hangover -0.017 [+ or -] 0.019 0.000 [+ or -] 0.007
Performed poorly on a -0.023 [+ or -] 0.019 -0.001 [+ or -] 0.007
test or important
project
Been in trouble with -0.029 [+ or -] 0.030 0.011 [+ or -] 0.012
police, residence
hall, or other
college authorities
Damaged property, -0.025 [+ or -] 0.029 0.004 [+ or -] 0.011
pulled fire alarm,
etc.
Got into an argument -0.017 [+ or -] 0.020 0.002 [+ or -] 0.008
or fight
Got nauseated or -0.018 [+ or -] 0.020 0.000 [+ or -] 0.008
vomited
Driven a car while -0.010 [+ or -] 0.015 0.000 [+ or -] 0.006
under the influence
Missed a class -0.055 [+ or -] 0.040 0.017 [+ or -] 0.015
Been criticized by a -0.021 [+ or -] .016 0.006 [+ or -] O.006
known person
Had a memory loss -0.023 [+ or -] 0.018 0.011 [+ or -] 0.007
([section])
Did something that was 0.004 [+ or -] 0.018 0.011 [+ or -] 0.007
later regretted
Been arrested for DWI 0.021 [+ or -] 0.022 0.002 [+ or -] 0.009
or DUI
Been hurt or injured -0.009 [+ or -] 0.024 0.015 [+ or -] 0.009
([section])
* Based on multiple linear regression with effect size for the outcome
as the dependent variable and the numbers of elements within each
program dimension as the independent variables. Units of measurement
of the dependent variables were the same as presented in Table 1.
Regression results for each dimension were independent of the
other five dimensions.
([dagger]) FIPSE denotes Fund for the Improvement of Postsecondary
Education: AOD, alcohol and other drug: DWI, driving while
intoxicated; DUI, driving while under the influence.
([double dagger]) P [less than or equal to] 0.05.
([section]) 0.06 [less than or equal to] P [less than or equal
to] 0.10.
ACKNOWLEDGMENTS
This study was partially supported by grants (P183F20013 and P183F40001) from the Fund for the Improvement of Postsecondary Education, U.S. Department of Education, Washington, D.C.
REFERENCES
(1.) Johnston LD, O'Malley PM, Bachman JG. National Survey Results on Drug Use from the Monitoring the Future Study, 1975-1994. Volume II: College Students and Young Adults. Rockville, MD: U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health, National Institute on Drug Abuse, 1996.
(2.) Johnston L, O'Malley P, Bachman J. National Survey Results on Drug Use from the Monitoring the Future Study, 1975-1998. Volume II: College Students and Young Adults. Rockville, MD: U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health, National Institute on Drug Abuse, 1999.
(3.) Presley CA, Meilman PW, Lyerla R. Alcohol and Drugs on American College Campuses: Use, Consequences, and Perceptions of the Campus Environment. Volume I: 1989-91. Carbondale, IL: Core Institute, 1993.
(4.) Presley CA, Meilman PW, Lyerla R. Alcohol and Drugs on American College Campuses: Use, Consequences, and Perceptions of the Campus Environment. Volume II: 1990-92. Carbondale, IL: Core Institute, 1995.
(5.) Presley CA, Meilman PW, Cashin JR, Lyerla R. Alcohol and Drugs on American College Campuses: Use, Consequences, and Perceptions of the Campus Environment. Volume III: 1991-93. Carbondale, IL: Core Institute, 1996.
(6.) Presley CA, Meilman PW, Cashin JR. Alcohol and Drugs on American College Campuses: Use, Consequences, and Perceptions of the Campus Environment. Volume IV: 1992-94. Carbondale, IL: Core Institute, 1996.
(7.) Weehsler 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. J Am Med Assoc 1994; 272:1672-1677.
(8.) 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: results of the Harvard School of Public Health College Alcohol Study. J Am Coll Health 1998; 47:57-68.
(9.) Wechsler H, Lee JE, 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 Coil Health 2000; 48:199-210.
(10.) 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.
(11.) Jones SE, Oeltmann J, Wilson TW, Brener ND, Hill CV. Binge drinking among undergraduate college students in the United States: implications for other substance use. J Am Coil Health 2001; 50:33-38.
(12.) U.S. Department of Education. Drug-free schools and campuses; final regulations (34 CFR Part 86). Fed Regist 1990; 55:33579-33601.
(13.) Center for Substance Abuse Prevention. Prevention Primer: An Encyclopedia of Alcohol, Tobacco, and Other Drug Prevention Terms. Rockville, MD: U.S. Department of Health and Human Services, 1993.
(14.) Brandon PR. State-level evaluation of school programs funded under the Drug-Free Schools and Communities Act. J Drug Educ 1992; 22:25 36.
(15.) Palmer CJ, Gehring DD, Guthrie VL. Student knowledge of information mandated by the 1989 Amendments to the Drug Free Schools and Communities Act. NASPA J 1992; 30:30-38.
(16.) Werch CE, Pappas DM, Castellon-Vogel EA. Drug use prevention efforts at colleges and universities in the United States. Subst Use Misuse 1996; 31:65 80.
(17.) Miller WR, Toscova RT, Miller JH, Sanchez V. A theory-based motivational approach for reducing alcohol/drug problems in college. Health Educ Behav 2000; 27:744-759.
(18.) Ziemelis A, Bucknam RB, Elfessi AM. Prevention efforts underlying decreases in binge drinking at institutions of higher education. J Am Coil Health 2002; 50:238-252.
(19.) Presley CA, Harrold R, Scouten E, Lyerla R, Meilman PW. Core Alcohol and Drug Survey: User's Manual. 5th ed. Carbondale, IL: Core Institute, 1994.
(20.) Meilman PW, Cashin JR, McKillip J, Presley CA. Understanding the three national databases on collegiate alcohol and drug use. J Am Coil Health 1998; 46:159-162.
(21.) Presley CA, Meilman PW, Lyerla R. Development of the Core Alcohol and Drug Survey: initial findings and future directions. J Am Coil Health 1994; 42:248-255.
(22.) Rosner B. Fundamentals of Biostatistics. Boston, MA: Duxbury Press, 1982:27-31.
(23.) Orwin RG. Evaluating coding decisions. In: Cooper H, Hedges LV. eds. The Handbook of Research Synthesis. New York: Russell Sage Foundation, 1994:139-162.
(24.) Hawkins JD, Nederhood B. Evaluation design. In: Handbook for Evaluating Drug and Alcohol Prevention Programs: Staff/Team Evaluation of Prevention Programs (STEPP). Rockville, MD: U.S. Department of Health and Human Services, Public Health Service, Alcohol, Drug Abuse and Mental Health Administration, Office for Substance Abuse Prevention, 1987.
(25.) Glass GV, McGaw B, Smith ML. Meta-Analysis in Social Research. Newbury Park, CA: Sage Publications, Inc., 1981.
(26.) Osburn HG, Callender J. A note on the sampling variance of the mean uncorrected correlation in meta-analysis and validity generalization. J Appl Psychol 1992; 77:115-122.
(27.) U.S. Census Bureau. Statistical Abstract of the United States: 1997. 117th ed. Washington, DC: Department of Commerce, 1997:181 - 184.
(28.) Clapp JD, Stanger LS, Burke AC. Life after funding: a strategy for survival of a program to prevent substance abuse. J Am Coil Health 1996; 45:79-81.
(29.) Kleinbaum DG, Kupper LL, Morgenstern H. Typology of epidemiologic research. In Epidemiologic Research: Principles and Quantitative Methods. New York, NY: Van Nostrand Reinhold Company Inc., 1982:62-95.
(30.) Perneger TV. What's wrong with Bonferroni adjustments. Br Med J 1998: 316:1236-1238.
(31.) Thompson JR. Re: "Multiple comparisons and related issues in the interpretation of epidemiologic data. "Am J Epidemiol 1998:147:801 - 806.
(32.) Goodman SN. Multiple comparisons, explained. Am J Epidemiol 1998; 147:807-812.
(33.) Savitz DA, Olshan AF. Describing data requires no adjustment for multiple comparisons: a reply from Savitz and Olshan. Am J Epidemiol 1998; 147:813-814.
(34.) Thompson JR. A response to "describing data requires no adjustment for multiple comparisons." Am J Epidemiol 1998; 147:815.
(35.) Browner WS, Newman TB. Are all significant P values created equal? The analogy between diagnostic tests and clinical research. J Am Med Assoc 1987; 257:2459-2463.
(36.) Morritz T, Seehafer RW, Maatz-Majestic E. A student competition to develop an innovative alcohol education strategy. J Am Coil Health 1993: 41:283-286.
(37.) Haines M, Spear SF. Changing the perception of the norm: a strategy to decrease binge drinking among college students. J Am Coll Health 1996; 45:134- 140.
(38.) Perkins HW, Meilman PW, Leichliter JS, Cashin JR, Presley CA. Misperceptions of the norms for the frequency of alcohol and other drug use on college campuses. J Am Coll Health 1999; 47:253-258.
(39.) Werch CE, Pappas DM, Carlson JM, DiClemente CC, Chally PS, Sinder JA. Results of a social norm intervention to prevent binge drinking among first-year residential college students. J Am Coll Health 2000; 49:85-92.
(40.) Bucknam RB. The other side of the coin. J Am Coil Health 1994; 42:305 308.
(41.) Wechsler H, Rigotti NA, Gledhill J, Lee H. Increased levels of cigarette use among college students: a cause for national concern. J Am Med Assoc 1998; 280:1673-1678.
John C. Licciardone, D.O., M.S. *
Department of Family Medicine, University of North Texas Health Science Center, Fort Worth, Texas, USA
* Correspondence: John C. Licciardone, D.O., M.S., Department of Family Medicine, University of North Texas Health Science Center, 3500 Camp Bowie Blvd., Fort Worth, TX 76107, USA; E-mail: jlicciar@hsc.unt.edu.
COPYRIGHT 2003 Marcel Dekker, Inc.
COPYRIGHT 2004 Gale Group
|
