research decreasing heavy drinking in first-year students ... · diano al. doumas, christina m....

Research

Decreasing Heavy Drinking in First-YearStudents: Evaluation of a Web-Based

Personalized Feedback Program AdministeredDuring Orientation

Diana M. Doumas, Christina M. Kane.Tabitha B. Navarro,and Jennifer Roman

This study evaluated the effectiveness of a web-based personalized normative feedback program, electronicCheck-Up to Go (e-CHUG). in decreasing heavy drinking among Ist-year university students. Results in-dicated high-risk students receiving the e-CHUG program during Ist-year orientation activities reportedsignificantly greater reductions in heavy drinking and alcohol-related consequences than did students in anassessment-only control group at a 3-month follov^-up. Recommendations for integrating e-CHUG intoorientation activities are discussed.

Heavy drinking represents a significant problem on college and uni-versity campuses in the United States, with over 30% of studentsmeeting criteria for a diagnosis of alcohol abuse (Knight et al.,

2002). National survey data have indicated 80% to 85% of U.S. college anduniversity students reported drinking (O'Malley & Johnston, 2002) and 40%to 45% reported a heavy drinking episode at least once in the 2 weeks priorto the survey (Wechsler ct ah, 2002). Heavy drinking is also associated withmultiple social problems, such as arguing vvith friends, unplanned sexual ac-tivitx', drinking and driving, getting into trouble with the law, and academicditficulties (Abbey, 2002; Cooper, 2002; Perkins, 2002; Vik, Carrello, Täte,& Field, 2000; Wechsler, Lee, Kuo, & Lee, 2000), as well as severe conse-quences such as unintended injuries, assault, and death (Hingson, Heeren,Winter, & Wechsler, 2005).

The National Institute for Alcohol Abuse and Alcoholism (NIAAA; NationalAdvisory Council on Alcohol Abuse and Alcoholism, Task Force on CollegeDrinking, 2002) has identified Ist-year students as a high-risk group forheavy drinking relative to the general student population. First-year studentsdrink more drinks and engage in heavy drinking episodes more frequently incomparison to upperclassmen (Turrisi, Padella, & Wiermsa, 2000). This high-risk status afforded to 1 st-year students has been attributed to several factors,

Diano Al. Doumas, Christina M. Kane, Tabitha B. Navarro, and Jennifer Roman, Department of Counselor Educa-

tion. Boise State University. Correspondence concerning this article should be addressed to Diana M. Doumas.

Department of Counselor Education, boise State University, 1910 University Drive, Roise, ID 83725-1721 (e-mail:

[email protected]).

© 2011 by the American Counseling Association. All rights reserved.

lournal of College Counseling • Spring 2011 • Volume 14 5

including an increase in freedom, decrease in social control, and increase instress experienced in higher education relative to high school (Arnett, 2005).Student drinking is also affected by perceptions of peer drinking (Perkins,2002), and research indicates this influence is strongest during the 1st year(Turrisi et al., 2000). These studies support the importance of providingevidence-based programs targeting heavy drinking among lst-year students.

Recent reviews of the literature support the efficacy of brief interventionsusing motivational interviewing and personalized normative feedback forreducing high-risk drinking in university and college students (Carey, Scott-Sheldon, Carey, & DeMartini, 2007; Larimer & Cronce, 2007). Innovativeapproaches to implementing brief motivational interventions have also beendeveloped, with a growing number of controlled studies indicating thatweb-based personalized feedback programs are effective in reducing drinkingand alcohol related consequences (Bersamin, Paschall, Fearnow-Kenney, &Wyrick, 2007; Chiauzzi, Green, Lord, Thum, & Goldstein, 2005; Doumas &Anderson, 2009; Doumas & Haustveit, 2008; Kypri et al., 2004; Neighbors,Larimer, & Lewis, 2004; Walters, Vader, & Harris, 2007). Although recentreviews of the literature have indicated that feedback—whether deliveredin person, by mail, or online—can be effective (Larimer & Cronce, 2007;Walters & Neighbors, 2005), web-based interventions may be particularlyuseful for lst-year students because online programming has the potentialboth to reach a wide audience and to be an engaging medium for studentswho enjoy "surfing the net."

Providing research to support the effectiveness of web-based alcohol pro-grams has become increasingly important with the proliferation of both thenumber of colleges and universities using web-based programs to addresscampus alcohol use and the number of these programs currently available onthe market. Although the literature supports the efficacy of providing personal-ized normative feedback electronically (Larimer & Cronce, 2007), only one ortwo program evaluations have been published supporting the effectiveness ofany specific program. An example of this type of online personalized feedbackprogram is electronic Check-Up to Go (e-CHUG), which was developed bycounselors and psychologists at San Diego State University' to help motivatestudents to examine their drinking. Although e-CHUG has been adopted byover 550 colleges and universities nationwide (San Diego State UniversityResearch Foundation, n.d.), to date, only two studies have been publishedspecifically evaluating the efficacy of e-CHUG in reducing heavy drinking inlst-year students (Doumas & Anderson, 2009; Walters et al., 2007).Walters et al. (2007) evaluated the efficacy of e-CHUG over 16 weeks in a

volunteer sample of 106 lst-year university students reporting heaw episodicdrinking. Students were randomly assigned to either e-CHUG or an assessment-only condition. Results at an 8-week follow-up assessment indicated that suidentsin the e-CHUG group reduced their drinks per week and peak blood alcoholcontent relative to the control condition, but at the 16-week follow-up therewere no longer differences between the two groups. Additionally, although

Journal ot College Counseling • Spring 2011 • Volume 14

alcohol-related consequences declined, there were no differences between thetwo groups on alcohol-related consequences at either the 8-week or 16-weekassessments. Similarly, Doumas and Anderson (2009) examined the efficacy ofe-CHUG in reducing heavy drinking in lst-year students using a randomizedcontrcjlled design. Doumas and Anderson, however, evaluated the effectiveness ofe-CHUG incorporated into the curriculum of a lst-year seminar otïered duringthe spring semester. Results at a 3-month follow-up assessment indicated studentsclassified as high-risk drinkers participating in e-CHUG reduced their weeklydrinking quantity, frequency of drinking to intoxication, and number of alcohol-related consequences relative to high-risk students in an assessment-only controlcondition. Results indicated that incorporating e-CHUG into lst-year studentactivities is a promising strategy tor reducing heavy drinking among such students.

Because e-CHUG is being used at so many colleges and universities and isoften administered during lst-year orientation for matriculating students, itis important to evaluate e-CHUG as a potential evidence-based strategy. Theaim ofthe current study is to extend the literature by conducting a programevaluation to examine the effectiveness of administering e-CHUG duringlst-year students' orientation. To achieve our aims, two orientation sectionswere randomly assigned to either the e-CHUG group or an assessment-onlycontrol group. We also classified students as high-risk or low-risk drinkersusing reports of binge drinking at the baseline assessment. We hypothesizedthat lst-year students classified as high-risk drinkers receiving e-CHUG wouldreport greater reductions in heavy drinking and alcohol-related consequencescompared with those in the control condition. We also hypothesized thatthere would be no differences in drinking measures between the two groupsfor lst-year students classified as low-risk drinkers.

Method

Participants and Procedure

Participants were recruited from two lst-year summer orientation sectionsat a large metropolitan university in the Northwest. The two orientationsections were randomly assigned by coin toss to either the e-CHUG groupor assessment-only control group. All lst-year students enrolled in the twosections and present during the baseline assessment ( N = 350) were given anopportunit)' to participate in the study. Enrollment in lst-year orientationwas voluntary and the program was offered as part of lst-year orientationactivities. Participants were offered a chance to win a $100 VISA gift cardas compensation for their participation. All participants were treated accord-ing to established American Psychological Association (2010) and AmericanCounseling Association (2005) ethical standards, and the research was ap-proved by the university's institutional review board.

At baseline, 350 students (35% men, 65% women) were present and participatedin the study. Of these, 167 (48%) were in the orientation section assigned to

Journal of College Counseling • Spring 2011 • Volume 14

the e-CHUG condition and 183 (52%) were in the orientation section assignedto the assessment-only control condition. Ages of the students ranged from17 to 19 years (M = 18.0, SD = 0.45). Ninety percent were Caucasian, 4%Hispanic, 3% Asian American, and 3% other. A series of chi-square analysesand t tests confirmed there were no differences between the two groups ingender, age, ethnicity, or any of the drinking variables at baseline.

All procedures were completed by participants during Ist-year orientationactivities. Members of the research team (the four authors) joined orientationleaders to facilitate the administration of the baseline assessment and e-CHUGprogram. Students completed an online survey at the baseline assessmentduring orientation (mid-July) and were sent online surveys by e-mail for the3-month follow-up assessment (late October). During baseline data collec-tion, students were assigned a personal code. This code was used to identifybaseline and follow-up responses for each student and to calculate responserates from baseline to follow-up assessments. All participants were informedof the nature of the study, risks and benefits of participation, and informationregarding the voluntary nature of participation. The e-CHUG program wasadministered during orientation for the intervention group. Students in thecontrol condition were sent information for accessing the e-CHUG programafter completion of the follow-up assessment survey. Botli baseline and follow-up assessment surveys took approximately 15 minutes to complete.

Instruments

Alcohol consumption. Recommendations by the NIAAA Task Force includeassessing patterns of alcohol consumption in addition to the average numberof drinks consumed (NIAAA, 2003). We included three measures to assessalcohol consumption: peak drinking quantity, frequency of drinking to in-toxication, and typical weekly drinking quantity. These variables were selectedto reflect the quantity of drinking on the heaviest drinking days on collegecampuses, the frequency of high-risk drinking behavior, and t)'pical drinkingbehavior. Peak drinking, drinking to intoxication, and weekly drinking arestandard measures of alcohol consumption and are based on widely used itemsin the higher education literature (e.g., Chiauzzi et al., 2005; Doumas &Anderson, 2009; Doumas & Haustveit, 2008; Doumas, McKinley, & Book,2009; Neighbors et al., 2004; Walters et al., 2007).

First, peak drinking quantity was assessed by the question "What is thehighest number of drinks that you have consumed on any given night in thepast 3 months?" Second, frequency of drinking to intoxication was assessedby the question "During the past 30 days (about 1 month), how many timeshave you gotten drunk, or very high from alcohol?" This item was rated ona 6-point scale with the anchors Ö, 1 to 2, 3 to 4, 5 to 6., 7 to 8, or more than 9times. Third, quantity of weekly drinking was assessed using a modified ver-sion of the Daily Drinking Questionnaire (DDQ; Collins, Parks, & Marlatt,1985). This item asks participants to indicate how much they typically drink.

Journal ot College Counseling • Spring 2011 • Volume 14

"Given that it is a typical week, please write the number of drinks you prob-ably would have each day." A response scale is provided for each day of theweek (e.g., Monday , Tuesday ). Weekly drinking was calculatedby combining the reports for the 7 days of the week. Previous research withadolescents and adults demonstrates nonsignificant correlations between thesedrinking measures and indexes of social desirability, reasonably high test-retestreliability estimates with coefficients from r = .85 to .90, and good conver-gence between these items and indexes of drinking quantity and frequencywith coefficients of r= .70 or higher (Turrisi, 1999; Turrisi & Jaccard, 1991).

Alcohol-related consequences. Alcohol-related consequences were assessed usingthe Rutgers Alcohol Problem Index (RAPI; White & Labouvie, 1989). TheRAPI is a 23-item self-administered screening tool used to measure problemdrinking. Participants are asked the number of times in the past 3 monthsthey experienced each of 23 consequences as a result of drinking. Responsesare measured on a 5-point scale ranging from never to more than 10 times.A total consequence score is created by summing the 23 items. The RAPIassesses both traditional physical consequences and consequences presumedto occur at higher rates in a college or university student population. TheRAPI has good internal consistency (Neal & Carey, 2004) and test-retestreliability (Miller et al., 2002), and is correlated significantly with severaldrinking variables (White & Labouvie, 1989). Cronbach's alpha for the cur-rent sample was a = .78.

Classification of hißh-risk versus low-risk drinkers. Following the HarvardSchool of Public Health College Alcohol Study (CAS), binge drinking wasdefined as having five or more drinks in a row for males (four or more forfemales; Wechsler, Davenport, Dowdall, Moeykens, & Castillo, 1994). Thisitem was used as an indicator of high-risk drinking and was used to createa risk variable, with participants indicating one or more occasions of bingedrinking in the 3 months at the baseline assessment classified as high-risk drink-ers. The 5/4 binge drinking measure has been widely used and supported asan appropriate threshold to identify high-risk drinkers (Wechsler & Nelson,2001, 2006) and identified as a dangerous level of drinking ("NIAAA Councilapproves definition of binge drinking," 2004). Although other studies haveused the binge criteria at rime frames of 1 week (Chiauzzi et al., 2005), 2weeks (Doumas & Anderson, 2009), and 1 month (Bersamin et al., 2007;Kypri et al., 2004; Walters et al., 2007), we selected a 3-month time frameto capture longer standing drinking patterns and to provide a less stringentcriteria for high-risk drinking, thus including more students in this group.Using this measure, 39% of the participants were classified as high-risk drink-ers and 61% were classified as low-risk drinkers.

Intervention

Students in the intervention group were directed to take e-CHUG, a Na-tional Association of Student Personnel Administrators (NASPA) recognized,

Journal of College Counseling • Spring 2011 • Volume 14

evidenced-based, online alcohol intervention and personalized feedback tooldeveloped by counselors and psychologists at San Diego State University. Thisbrief web-based program is designed to reduce high-risk drinking by providingpersonalized feedback and normative data regarding drinking and the risks as-sociated with drinking. The program is commercially available and is managedby the San Diego State University Research Foundation. Further details aboutthe program, procedures and costs for subscribing to the program, and sup-porting research are provided on the program website (http://www.e-chug.com/). The program is customized for the participating university, includingproviding normative data for the specific university population, referrals for thelocal community, and designing the website using university colors and logos.

The personalized feedback program takes approximately 30 minutes to com-plete. Students first complete an online assessment consisting of basic demo-graphic information (e.g., sex, age, weight, living situation, class standing) andinformation on alcohol consumption, drinking behavior, and alcohol-relatedconsequences. Immediately following the assessment, individualized graphedfeedback is provided in the following domains: Summary of quantity' and fi^equencyof drinking including graphical feedback, such as the number of cheeseburgersthat are equivalent to alcohol calories consumed, graphical comparison of one'sown drinking to U.S. adult and college-drinking norms, estimated risk status tornegative consequences associated with drinking and risk status for problematicdrinking based on the participant's Alcohol Use Disorder Identification Testscore, genetic risk, tolerance, approximate financial cost of drinking in the pastyear, normative feedback comparing one's perception of peer drinking to actualuniversity drinking normative data, and referral information for local agencies.

Results

Overall, 82 (23.4%) of the 350 participants completed the 3-month toUow-upassessment. For the final sample, 44% (n = 36) were in the e-CHUG group{n = 29 low risk; n = 7 high risk) and 56% (w = 46) were in the controlgroup (« = 35 low risk; w = 11 high risk). There was no difference in the rateof attrition across the e-CHUG and control groups, x"(l, N = 350) = 0.62,p = .43. In addition, chi-square and í tests revealed no differences in demo-graphic variables between the participants who completed the follow-upassessment and those who did not. Participants who did not completethe follow-up assessment, however, reported higher levels of peak drink-ing, r(348) = 3.58, p < .001, Cohen's d = .44; drinking to intoxication,i(348) = 2.60, ^ < .01, Cohen's d = .32; weekly drinking, i(348) = 3.22,p < .001, Cohen's d = .41; and alcohol-related consequences, i(348)= 2.22, p < .05, d = .29, than did those who completed the 3-monthsurvey, although effect sizes were small. Only the 82 participants whocompleted both pretest and posttest assessments were included in thefollowing analyses.

10 Journal of College Counseling • Spring 2011 • Volume 14

Alcohol Consumption

Three repeated measures factorial analyses of variance (ANOVAs) were con-ducted to examine differences between the e-CHUG group and the controlgroup from baseline to the 3-month follow-up assessment for peak drinkingquantity, frequency of drinking to intoxication, and weekly drinking quantity.The three independent variables were time (baseline, 3-month follow-up),group (e-CHUG, control), and risk status (high-risk; low-risk). Means foreach ofthe dependent variables by group and risk status are shown in Table 1.

Differences in

TABLE 1

Alcohol Consumption and Alcohol-RelatedConsequences by Study Condition

Condition and Time

e-CHUGBaseline3 months

ControlBaseline3 months







Note. e-CHUG = electronic

High(n =

M

Peak

9.33.9

7.17.9

Risk18)

SD

Drinking

4.42.3

2.66.1

Risk

and Risk

status

Low Risk("

M

Quantity

0.61.0

0.40.7

Drinking to Intoxication

1.40.9

1.31.5

1.90.7

0.81.2

0.10.3

0.10.1

Weekly Drinking Quantity

5.63.7

5.25.7

7.74.4

4.15.7

0.11.0

0.10.3

= 64)

SD

1.02.1

0.91.5

0.40.8

0.40.4

0.43.8

0.31.1

Alcohol-Related Consequences

4.60.9

3.62.2

Check-Up to

4.71.6

3.32.4

Go.

0.40.4

0.20.6

1.22.0

0.80.3

Status

Total(W

M

2.61.6

2.02.4

0.40.4

0.40.4

1.20.6

1.31.6

1.20.5

1.00.6

Sample= 82)

SD

• '

4.11.6

3.24.4

1.00.7

0.70.9

3.92.6

3.03.4

2.82.0

2.21.5

Journal ot College Counseling • Spring 2011 • Volume 14 II

Results of the ANOVAs indicated a significant interaction effect for theTime X Group x Risk Status for peak drinking, f ( l , 78) = 16.41, p < .001,r| = .17, and for drinking to intoxication, P(l, 78) = 5.59, p < .02, r\^ = .07.Findings indicate that high-risk students in the e-CHUG group reduced theirpeak drinking quantity and frequency of drinking to intoxication significantlymore than did those in the control group. Because of the small sample sizein the high-risk groups, we elected to use effect size calculations rather thansignificance testing to examine post hoc between-group pair-wise comparisonsfor drinking measures at the 3-month follow-up between the e-CHUG groupand the control group. For high-risk students, effect size calculations indicateda large effect size between the means for the students in the e-CHUG groupand control group for peak drinking, Cohen's d = .81, and a medium effectsize for drinking to intoxication, Cohen's d = .58. In contrast, for low-riskstudents, calculations indicated small effect sizes between the means for stu-dents in the e-CHUG and control group for both peak drinking, Cohen's d= .17, and drinking to intoxication, d = .32. Calculations ofthe differencesbetween baseline and 3-month follow-up means in Table 1 indicate thathigh-risk students in the e-CHUG group reduced their peak drinking by58% compared with an 11% increase in the control group. Similarly, high-risk students in the e-CHUG group reduced their drinking to intoxicationby 65% compared with a 15% increase in the control group (see Figure 1).

In contrast, the ANOVA for weekly drinking was not significant, althoughresults for the Time x Group x Risk Status followed a similar pattern, i^(l,78) = 2.23, p = .14, X]^ = .03. Although not statistically significant, calcula-tions of the differences between baseline and 3-month follow-up means inTable 1 indicate that high-risk students in the e-CHUG group reduced theirweekly drinking by approximately 34% compared with increases of 10% inthe control group (see Figure 1).

Alcohol-Related Consequences '

A repeated measures factorial ANOVA was conducted to examine differencesin alcohol-related consequences between the e-CHUG group and the controlgroup from baseline to the 3-month follow-up assessment. The three inde-pendent variables were time (baseline, 3-month follow-up), group (e-CHUG,control), and risk status (high-risk, low-risk). Means for alcohol-related con-sequences by group and risk status are shown in Table 1.

Results ofthe ANOVA indicated a significant interaction effect for the TimeX Group X Risk Status for alcohol-related consequences, F{1, 78) = 4.63, p <.03, T] = .06. Findings indicate that high-risk students in the e-CHUG groupreported fewer alcohol-related consequences than those in the control group.Ef fect size was used for post hoc between-group pair-wise comparisons foralcohol-related consequences at the 3-month follow-up between the e-CHUGgroup and control group. For high-risk students, effect size calculations in-dicated a large effect size between the means for students in the e-CHUG


: High-Risk e-CHUG Baseline: High-Risk Control 3 Months

= Low-Risk e-CHUG Baseline= Low-Risk Controi 3 Months

8 -

6 —

4 —

2 -

2 —

1.5 —

1 —

0 . 5 -

5 -

4 —

Peak Drinking

Drinking to Intoxication

Alcohol-Related Consequences

FIGURE 1

Changes in Drinking and Alcohol-Related Consequences by Study

Condition and Risk Status

Note. e-CHUG = electronic Check-Up to Go.

lournal ot College Counseling • Spring 2011 • Volume 14 13

group and control group, Cohen's d= .61. In contrast, for low-risk students,calculations indicated a small effect size between the means for students inthe e-CHUG and control group, Cohen's d = .24. Calculations of the dif-ferences between baseline and 3-month follow-up means in Table 1 indicatethat high-risk students in the e-CHUG group reported an 80% reduction intheir alcohol-related consequences compared with a 39% reduction in thecontrol group (see Figure 1).

Discussion

The aim of this study was to evaluate the effectiveness of a web-based per-sonalized feedback program (e-CHUG) in decreasing heavy drinking andalcohol related consequences among lst-year university students. Althoughe-CHUG has been adopted by over 550 colleges and universities nationwide(San Diego State University Research Foundation, n.d.), to date, there are onlytwo published studies providing evidence to support the efficacy of e-CHUGin lst-year students (Doumas & Anderson, 2009; Walters et al., 2007). Thisis the first study to examine e-CHUG as part of lst-year orientation activi-ties for matriculating students. Thus, this study adds to the growing body ofliterature supporting the efficacy of web-based normative feedback programsand provides evidence for the effectiveness of e-CHUG in particular.

Findings confirmed the hypothesis that the reductions in heavy drink-ing and alcohol-related consequences in the e-CHUG group would besignificantly greater than reductions in the assessment-only control groupfor high-risk drinkers, whereas changes in drinking for lst-year studentsin the low-risk group would be similar across the intervention and controlgroups. High-risk lst-year students in the e-CHUG group reported a 58%reduction in peak drinking, 65% reduction in frequency of drinking tointoxication, and 34% reduction in weekly drinking compared with 11%,15%, and 10% increases, respectively, in the control group at the 3-monthfollow-up. Additionally, although both groups reported a decrease inalcohol-related consequences, high-risk students in the e-CHUG groupreported an 80% reduction compared with a 39% reduction in the controlgroup. It is important to note that for high-risk students, hea\7 drinkingdecreased 34% to 65% in the e-CHUG group compared with a 10% to 15%increase in drinking in the control condition. This indicates that lst-yearstudents in the e-CHUG group reduced their drinking despite the naturaltrajectory of an increase in heavy drinking seen in the lst-year studentswho did not receive the program. Additionally, students in the e-CHUGcondition reported a reduction in alcohol-related consequences that weredouble that of students in the control group.

Results of this study are consistent with research indicating web-based per-sonalized feedback programs are effective in reducing heavy drinking (Bersaminet al., 2007; Chiauzzi et al., 2005; Doumas & Anderson, 2009; Doumas &


Haust\eit, 2008; Doumas et al., 2009; Kypri et al., 2004; Neighbors et al.,2004; Walters et al., 2007) and alcohol-related consequences (Doumas &Anderson, 2009; Kypri et al., 2004; Neighbors et al., 2004) among collegeand university students. In addition, the finding that e-CHUG was mosteffective for Ist-year students classified as high-risk drinkers parallels the lit-erature examining web-based feedback. Specifically, the majority of researchexamining web-based programs in college and universit)' students has dem-onstrated etlicacy in students identified as high-risk drinkers (Chiauzzi et al.,2005; Doumas & Anderson, 2009; Doumas & Haustveit, 2008; Kypri et al.,2004; Neighbors et al., 2004; Walters et al., 2007) or indicates that reduc-tions in drinking are greater in high-risk students (Bersamin et al., 2007) andpersistent binge drinkers (Chiauzzi et al., 2005). ; • , '

Implications for College and University Counseling

Results of this study have important implications for prevention and interven-tion efforts aimed at reducing drinking and alcohol-related consequences forIst-year college and university students. First, 39% of this sample was classifiedas high-risk drinkers, indicating more than one third of the matriculating stu-dents in this sample reported binge drinking at least once in the past 3 monthsprior to the baseline survey. Additionally, 1 st-year students in the control groupactually increased their drinking over the course of the fall term. Coupled withprior research indicating Ist-year students increase their alcohol use over theacademic year (Borsari, Murphy, & Barnett, 2007) and are at risk for alcohol-related consequences throughout the academic year (Doumas & Anderson,2009), counselors need to remain aware that drinking may become heavier andmore alcohol-related consequences may occur as the academic year progresses.

Additionally, web-based programs such as e-CHUG should be incorpo-rated into a comprehensive strategy including community, campus envi-ronment, and individual-level programs. For example, Sullivan and Risler(2002) suggested college counselors develop a multisystemic interventionapproach that includes social marketing, risk reduction, gender-specificrecovery groups, and brief motivational interventions. Additionally, en-vironmental strategies such as community and campus alcohol policiestargeting responsible drinking need to be implemented. Although web-based personalized feedback is cost-effective and easy to disseminate tolarge groups of students, this strategy may not be effective for all students.For example, although web-based programs may be effective for studentscited for campus alcohol policy violations in the short term (Doumas etal., 2009), recent research indicates that in-person brief interventions maybe more effective than written or computer-based feedback for mandatedstudents over longer periods of time (Carey, Henson, Carey, & Maisto,2009; Doumas & Rudeen, 2009; White, Mun, Pugh, & Morgan, 2007).Thus, web-based feedback programs should be viewed as part of a largeroverall campus strategy to reduce heavy drinking.

Journal of College Counseling • Spring 2011 • Volume 14 15

LiiTiitations and Directions for Future Research

Although this study provides additional empirical support for web-basedpersonalized feedback programs in general, and e-CHUG in particular,there are several limitations. First, the high rate of attrition significantly af-fects the generalizability ofthe results. Although attrition rates were similaracross study conditions, suggesting attrition was not related to the particularstudy condition, students who dropped out of the study reported higherlevels of alcohol consumption and alcohol-related consequences at base-line than did students who completed the study. This differential dropoutpattern resulted in fewer students in the high-risk group than would havebeen expected. Because web-based interventions have been shown to bemore effective with high-risk students (Bersamin et al., 2007; Chiauzzi etal., 2005; Doumas & Anderson, 2009; Doumas & Haustveit, 2008; Kypriet al., 2004; Neighbors et al., 2004; Walters et al., 2007), the differentialattrition of these students relative to low-risk students may have dampenedour effects. Similarly, because attrition resulted in small sample sizes in thehigh-risk groups, statistical power was decreased, affecting our ability toconduct post hoc analyses of statistical significance and thus leading to thereport of effect sizes only.

Next, information in this study was obtained through self-report. Althoughself-report potentially leads to biased or distorted reporting, university studentsmay not be motivated to misrepresent their alcohol use because heavy drinkingis perceived as normal in the college and college setting (Borsari & Muellerieile,2009). Self-reported alcohol use is common practice in studies evaluating web-based interventions for college and university students, and results of a recentmeta-analysis support this usage, indicating that the reliability of self-reporteddrinking in college students is good, with little bias reported between participantand collateral reports (Borsari & Muellerieile, 2009). Future research, however,could include more objective outcomes (e.g., sanctions for campus alcoholpolicy violations, retention, or grade point average) in addition to self-reportmeasures of drinking behavior and associated consequences.

An additional limitation is related to our recruitment strategy. First-yearstudents were recruited from two lst-year orientation sections. Studentsenrolled in orientation may be different from general college or universitylst-year students, because orientation was voluntary and may have attracted aparticular type of student. Additionally, because of logistics in implementingthis evaluation as part ofthe lst-year orientation, orientation sections ratherthan lst-year students were randomly assigned to the two conditions. Lackof random assignment may lead to initial differences between groups, whichmay affect the validity of findings. Examination of demographic variables anddrinking measures revealed no baseline differences between the two groups,thus mitigating the problem of nonrandom assignment. In future studies,however, random assignment of individual students rather than orientationsections should be conducted.


Finally, the duration ofthe 3-month follow-up was fairly short. Althougheffects of web-based feedback programs have been shown to last for up to6-months in university students (Neighbors et al., 2004), a recent dismantlingstudy indicated that at a 6-month follow-up, motivational interviewing withpersonalized normative feedback was more effecrive in reducing universitystudent drinking than was web-based feedback alone, and there were nodifferences between the web-based feedback alone and assessment-onlyconditions (Walters, Vader, Harris, Field, & Jouriles, 2009). Additionally,results from a recent meta-analysis suggested that intervention effects maydecline after 6 months in college and university students (Carey et al.,2007). Therefore, future research should examine the efficacy of web-basedprograms implemented for lst-year students across a longer period of timeand preferably beyond the 1st year.

Conclusions

Despite prevention efforts, lst-year students remain a high-risk populationfor heavy drinking and alcohol-related consequences on college and universitycampuses. Additionally, although web-based alcohol prevention programs havebeen adopted by hundreds of colleges and universities across the country, thereare no more than one or two studies using a randomized controlled designpublished on any specific web-based program. Results of this study add to thegrowing body of literature suggesting that providing web-based personalizedfeedback to lst-year students is a promising strategy for decreasing heavy drink-ing in this high-risk population. This study also provides additional evidencefor the efficacy of e-CHUG in reducing heavy drinking and alcohol-relatedconsequences and is the first to demonstrate the effectiveness of e-CHUGadministered to matriculating students during orientation.

Because of the low cost, ease of dissemination, and growing empirical evi-dence associated with web-based personalized normarive feedback, this typeof program is ideal for colleges and universities having limited resources,needing to target large numbers of students, or wanting to provide studentsunlimited program access across the academic year. Direcrions for futureresearch include examining the impact of web-based feedback programsover longer follow-up periods, with larger samples and with other high-risk groups such as student athletes, fraternity and sorority members, andmandated students. Furthermore, additional measures of alcohol-relatedconsequences (e.g., campus alcohol policy violadons) and academic success(e.g., retention or grade point average) could be included as more objectivemeasures of effectiveness.

References

Abbey, A. (2002). Alcohol related sexual assault: A common problem among college students.Journal of Studies on Alcohol, VoL Supplement 14, 118-128.

American Counseling Association. (2005). ACA eode ofethies. Alexandria, VA: Author.


American Psychological Association. (2010). Ethical principles ofpsycholqgists and code of conduct:2010 amendments. Retrieved from http://www.apa.org/ethics/code/index.aspx

Arnett, J. J. (2005). The developmental context of substance abuse in emerging adulthood.Journal of Dniß Issues, 35, 235-254.

Bersamin, M., Paschall, M. I., Fearnow-Kenncy, M., & Wyrick, D. (2007). Effectiveness of aweb based alcohol misuse and harm prevention course among high- and low risk students.Journal of American College Health, .55, 247-254.

Borsari, B., & Muellerleile, P. (2009). Collateral reports in the college setting: A meta-analyticintegration. Alcoholism: Clinical and Experimental Research, 33, 826-838.

Borsari, B., Murphy, J. G., & Barnett, N. P. (2007). Predictors of alcohol use during the firstyear of college: Implications for prevention. Addictive Behaviors, 32, 2062-2086.

Carey, K. B., Henson, J. M., Carey, M. P., & Maisto, S. A. (2009). Computer versus in personintervention for students violating campus alcohol policy. Journal of Constating and ClinicalPsychology, 77, 74-87.

Carey, K. B., Scott-Sheldon, L. A. J., Carey, M. P., & DeMartini, K. S. (2007). Individual-levelinterventions to reduce college student drinking: A meta-analytic review. Addictive Behaviors,32, 2469-2494.

Chiauzzi, E., Green, T. C , Lord, S., Thum, C, & Goldstein, M. (2005). My Student Body:A high-risk drinking prevention web site for college students. Journal of American CollegeHealth, 53, 263-274.

Collins, R. L., Parks, G. A., & Marlatt, G. A. (1985). Social determinants of alcohol consump-tion: The effects of social interaction and model status on the self-administration of alcohol.Journal of Consulting and Clinical Psychology, 53, 189-200.

Cooper, M. L. (2002). Alcohol use and risky sexual behavior among college students and youth:Evaluating the evidence. Journal of Studies on Alcohol, Vol. Supplement 14, 101-117.

Doumas, D. M., & Anderson, L. (2009). Reducing alcohol use in first year university students:Evaluation of a web-based personalized feedback program. Journal of College Counseling, 18,18-32.

Doumas, D. M., & Haustveit, T. (2008). Reducing hea\7 drinking in intercollegiate athletes:Evaluation of a web-based personalized feedback program. The Sport Psychologist, 22, 212-228.

Doumas, D. M., McKinley, L., & Book, P. (2009). Evaluation of two web-based alcohol in-terventions for mandated college students. Journal of Substance Abuse Treatment, 36, 65-74.

Doumas, D. M., & Rudeen, C. (2009, June). Decreasing drinking in mandated students: Web-based vs. in-person feedback. Poster session presented at the meeting of Research Society forAlcoholism, San Diego, CA.

Hingson, R. W., Heeren, T., Winter, M., & Wechsler, H. (2005). Magnitude of alcohol-relatedmortality and morbidity among U.S. college students ages 18-24: Changes from 1998-2001.Annual Review of Public Health, 26, 259-279.

Kjiight, J. R., Wechsler, H., Kuo, M., Seibring, M., Weitzman, E. R., & Schuckit, M. (2002). Alcoholabuse and dependence among U.S. college students. Journal of Studies on Alcohol, 63, 263-270.

Kypri, K., Saunders, J. B., Williams, S. M., McGee, R. O., Langley, J. D., Cashell Smith, M. L.,& Gallagher, S. J. (2004). Web-based screening and brief intervention for hazardous drinking:A double-blind randomized controlled trial. Addiction, 99, 1410-1417. doi:10.111 l/j.1360-0443.2004.00847.x

Larimer, M. E., & Oonce, J. M. (2007). Identification, prevention, and treatment revisited:Individual-focused college drinking prevention strategies 1999-2006. Addictive Behaviors, 32,2439-2468. doi:10.1016/j.addbeh.2007.05.006

Miller, E. T., Neal, D. J., Roberts, L. J., Baer, J. S., Cressler, S. O., Metrik, J., & Marlatt, G.A. (2002). Test-retest reliability' of alcohol measures: Is there a difference between Internet-based assessment and traditional methods? Psychology of Addictive Behaviors, 16, 56-63.doi:10.1037/0893-164X.16.1.56

National Advisory Council on Alcohol Abuse and Alcoholism, Task Force on College Drinking.(2002). A call to action: Changing the culture of drinking at U.S. colleges (NIH Publica-tion, No. 02-5010). Retrieved from http://www.collegedrinkingprevention.gov/media/TaskForceReport.pdf

18 Journal üt College Counseling • Spring 2011 • Volume 14

National Institute on Alcohol Abuse and Alcoholism. (2003 ). Task Force on Recommended AlcoholQtiestions: National Council on Alcohol Abuse and Alcoholism Recommended sets of alcohol con-sumption questions-October IS-16,2003. Retrieved from http://ww\v,niaaa.nih.go\'/resources/researchresources/taskforce.htni

National Institute of Alcohol Abuse and Alcoholism Council approves definition of bingedrinking. (2004, Winter). National Institute of Alcohol Abuse and Alcoholism Newsletter.Retrieved from http://pubsniaaanih.gov/publications/newsletter/winter2004/newslet-ter_number3.pdf

Neal, D. J., & Carey, K. B. (2004). Developing discrepancy within self-regulation theory: Use ofpersonalized normative feedback and personal strivings with heavy-drinking college students.Addictive Behaviors, 29, 281-297.

Neighbors, C , Larimer, M. E., & Lewis, M. A. (2004). Targeting niisperceptions of descriptivedrinking norms: EtHcacy of a computer-delivered personalized normative feedback intervention.Journal of Consulting and Clinical Psychology, 72, 434-447.

O'Malley, R M., & Johnston, L. T). (2002). Epidemiology of alcohol and other drug use amongAmerican college students. Journal of Studies on Alcohol, Vol. Supplement 14, 23-29.

J erkins, H. W. (2002). Social norms and the prevention of alcohol mi.suse in collegiate contexts.Journal of Studies on Alcohol, Vol. Supplement 14, 164-172.

San Diego State University Research Foundation, (n.d.). eCHECKUP TO GO. Retrieved fromhttp://www.echeckuptogo.com/usa/

Sullivan, M., & Risler, E. (2002). Understanding college alcohol abuse and academic per-formance: Selecting appropriate intervention strategies. Journal of College Coumelin^, 5114-123.

Turrisi, R. (1999). Cognitive and attitudinal factors in the analysis of alternatives to binge drink-ing. Journal of Applied Social Psychology, 29, 1.S12-1535.

Turrisi, R., & Jaccard, J. (1991 ). Judgment processes relevant to drunk driving. Journal of Ap-plied Social Psychology, 21, 89-118.

Turrisi, R., Padella, K. K., & Wiersma, K. A. (2000). College student drinking: An examinationof theoretical models of drinking tendencies in freshmen and upperclassmen. Journal of Studieson Alcohol, 61, 598-602.

Vik, P. W., C:arrello, P., Täte, S. R., & Field, C. (2000). Progression of con.sequences amongheaw-drinking college students. Psycholoßy of Addictive Behaviors, 14, 91-101.

Walters, S. T., & Neighbors, C. (2005). Feedback interventions for college alcohol misuse: What,why, and for whom? Addictive Behaviors, 30, 1168-1182.

Walters, S. T., Vader, A. M., & Harris, T. R. (2007). A controlled trial of web-based feedbackfor hea\'y drinking college students. Prevention Science, 8, 83-88.

Walters, S. T., Vader, A. M., Harris, T. R., Field, C. A., & Jouriles, E. N. (2009). Dismantlingmotivational interviewing and feedback for college drinkers: A randomized clinical trial. Journalof Consulting and Clinical Psychology, 77, 64-73.

Wechsler, H., Davenport, A., Dowdall, G., Moeykens, B., & Castillo, S. (1994). Health andbehavioral consequences of binge drinking in college: A national survey of students at 140campuses. Journal of the American Medical Association, 272, 1672-1677.

Wechsler, H., Lee, J. E., Kuo, M., & Lee, H. (2000). College binge drinking in the 1990s: Acontinuing problem. Results of the Harvard School of Public Health 1999 College AlcoholStudy. Journal of American College Health, 48, 199-210.

Wechsler, H., Lee, J. E., Kuo, M., Seibring, M., Toben, N., & Lee, H. (2002). Trends in collegebinge drinking during a period of increased prevention efforts. Journal of American CollegeHealth, 50, 203-217.

Wechsler, H., & Nelson, T. F. (2001). Binge drinking and the American college student: What'sfive drinks? Psychology of Addictive Behaviors, 15, 287-291.

Wechsler, H., & Nelson, T. F. (2006). Relationship between level of consumption and harmsin assessing drink cut-points for alcohol research: Commentary on "Many College FreshmenDrink at Levels Far Beyond the Binge Threshold" by White et al. Alcoholism: Clinical andExperimental Research, 30, 922-927. doi:10.1111/j.l530-0277.2006.00124.x


White, H. R., & Labouvie, E. W. (1989). Towards the assessment of adolescent problem drink-ing. Journal of Studies on Alcohol, 50, 30-37.

White, H. R., Mun, E. Y., Pugh, L. A., & Morgan, T. J. (2007). Long-term effects of briefsubstance-use interventions for mandated college students: Sleeper effects of an in-person per-sonal feedback intervention. Alcoholism: Clinical and Experimental Research, 31, 1380-1391.doi:10.1111/j.l530-0277.2007.00435.x


Copyright of Journal of College Counseling is the property of American Counseling Association and its content

may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express

written permission. However, users may print, download, or email articles for individual use.

research decreasing heavy drinking in first-year students ... · diano al. doumas, christina m....

Documents