If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author at: University of Southern California, Keck School of Medicine, Department of Psychiatry and Behavioral Sciences, 2250 Alcazar Street, Suite 2200, Los Angeles, CA 90033, United States of America.
There are high rates of heavy drinking, risky sexual behaviors, and sexual violence among college students abroad
•
Few interventions exist for students abroad to prevent alcohol and sexual risk
•
Small effects were observed on drinking and risky sex outcomes favoring a brief online intervention for students abroad
•
Modifications to the developed intervention (e.g., booster sessions) may be necessary to observe larger, long-term effects on all outcomes
Abstract
Introduction
American college students who study abroad experience increases in their drinking behavior, concerning risky sexual behaviors, and high rates of sexual violence while abroad. Despite these concerns, institutions offer limited programming to students prior to departure to address these risks and no empirically supported interventions currently exist that are targeted toward preventing increased drinking, risky sexual behavior, and sexual violence abroad. To help address alcohol and sexual risk abroad, we designed a brief, single-session online predeparture intervention focused on risk and protective factors known to associate with alcohol and sexual risk abroad.
Methods
Using a sample of 650 college students from 40 home institutions, we conducted a randomized controlled trial of the intervention and tested its effects on drinking (drinks per week, binge drinking frequency, alcohol related-consequences), risky sexual behaviors, and sexual violence victimization during the trip abroad (first month, last month abroad) and one-month and three-months after return home.
Results
We observed small, but nonsignificant, intervention effects on drinks per week and binge drinking days during the first month abroad and three months after they had returned home to the United States, and small, significant effects on risky sexual behaviors during the first month abroad. The study found no observable effects at any time point for alcohol-related consequences or for sexual violence victimization abroad.
Conclusions
Though mostly nonsignificant, small initial intervention effects were promising in this first empirical test of an alcohol and sexual risk prevention program for study abroad students. However, students may need more intensive programming with booster sessions to experience lasting intervention effects during this particularly high-risk period.
More than 10 % of American college students study abroad in foreign countries at some point during their college career, with 347,099 students studying abroad from US institutions during the 2018/2019 academic year. Although studying abroad can be associated with multiple personal, cultural, and academic benefits for students who participate in these programs (
). Despite the known risks, not all college and universities offer predeparture programs focusing on these issues to help prepare students for potential hazards (
), and even if they do offer such programs, the majority of study abroad students report not receiving any alcohol and sexual risk prevention programming specific to the study abroad environment prior to leaving for trips abroad (
). Thus, empirically supported abroad-specific alcohol and sexual risk prevention programs are needed to help address heavy drinking among students, decrease risky sexual behaviors, and prevent incidents of sexual violence abroad.
1.1 Risk and protective factors for heavy drinking abroad
The number of drinks students drink per week more than doubles while students are studying abroad, with students reporting significant and lasting alcohol-related consequences, including academic disruptions, interpersonal conflicts, alcohol-related injuries, and health complications (
Hazardous alcohol use and cultural adjustment among U.S. college students abroad in Italy: Findings and recommendations for study abroad staff and researchers.
Not all those who wander are lost: Examining the impact of sojourner adjustment and drinking motives on alcohol consequences experienced by Americans studying in foreign countries.
). One modifiable risk factor for heavy drinking abroad is a student's misperception about peer drinking norms in the host country. Overestimations about the drinking behavior of other study abroad peers and beliefs that heavy drinking by local people is normative and acceptable are contributors to students' own drinking behavior abroad (
Demographic and predeparture factors associated with drinking and alcohol-related consequences for college students completing study abroad experiences.
). Providing college students with actual norms about peers' drinking in personalized normative feedback (PNF) interventions reduces their drinking and prevents alcohol-related consequences in college settings, including prior to known high-risk events like 21st birthday celebrations and spring break trips (
), and may be helpful to prevent escalation of heavy drinking behavior abroad.
Engagement with the culture while abroad, which is also known as sojourner adjustment, or the sociocultural and psychological adjustment of relatively short-term visitors to new cultures (
Hazardous alcohol use and cultural adjustment among U.S. college students abroad in Italy: Findings and recommendations for study abroad staff and researchers.
Examining the relationships between acculturation orientations, perceived and actual norms, and drinking behaviors of short-term American sojourners in foreign environments.
Not all those who wander are lost: Examining the impact of sojourner adjustment and drinking motives on alcohol consequences experienced by Americans studying in foreign countries.
). Those study abroad students who engage in aspects of positive sojourner adjustment, such as by spending quality time with local people, actively engaging with cultural experiences and events, developing and using foreign language skills, and temporarily identifying as a member of the host culture (
), are best protected from heavy alcohol use and problems. In contrast, those students most at-risk separate themselves from the host environment and spend most of their time with other Americans, perceive the abroad culture as very different than their home culture, feel out of place or homesick, and experience anxiety about interacting in the foreign environment (i.e., negative sojourner adjustment) (
Examining the relationships between acculturation orientations, perceived and actual norms, and drinking behaviors of short-term American sojourners in foreign environments.
Not all those who wander are lost: Examining the impact of sojourner adjustment and drinking motives on alcohol consequences experienced by Americans studying in foreign countries.
Helping students form accurate beliefs about drinking abroad and promoting positive sojourner adjustment during the abroad experience has real potential to prevent escalation of heavy and problematic drinking abroad. Yet only one known empirical evaluation exists of an intervention designed to prevent heavy drinking and alcohol-related consequences abroad. In a prior pilot study with 343 study abroad students, we found that providing PNF, as well as tips and strategies to help students increase engagement with the culture abroad (i.e., tips to increase their engagement in positive sojourner adjustment behaviors and limit their engagement in negative aspects of sojourner adjustment), prevented escalations in alcohol-related consequences abroad compared to students in a control condition, with the most pronounced prevention effects on alcohol-related consequences abroad observed in those who went abroad with already high levels of alcohol consequences at predeparture (
Brief online interventions targeting risk and protective factors for increased and problematic alcohol use among american college students studying abroad.
1.2 Risky sexual behavior and sexual violence abroad
Though there is promise for an alcohol-focused prevention program for study abroad students, no empirical studies exist of interventions designed to address both alcohol use and sexual risk abroad, including limiting risky sexual behavior (e.g., sex with multiple casual partners) and preventing incidents of sexual violence. Small cross-sectional studies have shown that upwards of 50 % of students report risky sexual behavior abroad (
). Our own longitudinal study of 2630 students studying abroad from 65 different US colleges and universities showed that although risky sexual behavior slightly decreased overall from predeparture levels for students while abroad, these behaviors were still prominent, and impulsive sexual behaviors (e.g., number of sexual partners, regretted sexual encounters) increased while abroad (
). Our own large multisite study indicated rates of 21 % for sexual violence during the period they spend abroad (i.e., nonconsensual sexual touching, attempted and completed rape) among male and female students (
). These rates are alarming given the brief nature of study abroad. That is, most study abroad students are only abroad for between one month to one semester/one quarter (∼15 weeks) (
). Moreover, the limited social, family, and home school supports available abroad could impact students' ability to cope with sexual violence if they experience it abroad. In addition, sexual violence abroad is often perpetrated by local people in the home country (
), making interventions that help American students engage in safe behaviors while abroad essential, as interventions geared toward reducing perpetration among host country individuals are less practical.
Further, risks for both risky sexual behaviors and sexual violence are amplified when students are drinking heavily, with on-campus research demonstrating heavy alcohol use increases potential for risky sexual behavior, and is associated with increased risk for both sexual assault victimization and perpetration (
). By addressing heavy drinking abroad, and helping students understand the connection between unwanted sexual outcomes and heavy drinking, potential exists for interventions to address both alcohol and sexual risks for students abroad.
1.3 The current study
Based on our successful pilot alcohol intervention work with college students at one US institution (
Brief online interventions targeting risk and protective factors for increased and problematic alcohol use among american college students studying abroad.
), we developed an intervention that incorporated empirically informed content based on risk and protective factors specific to alcohol and sexual risk for study abroad students (
). Using an online modality available to be viewed on computers and mobile phones, we designed the intervention to increase the accessibility of predeparture programming to students across a large number of US home institutions and be applicable to students traveling to popular study abroad host countries. It was designed to appeal to students, with peers delivering the content through video clips and interactive components to engage students' attention, all during a brief format. In the current study, we tested the intervention in a randomized controlled trial with 650 students from 40 home colleges and universities and evaluated outcomes of drinking, risky sexual behaviors, and sexual violence victimization.
Drinking frequency may escalate for students while abroad, perhaps because drinking may actually be more normative abroad than in the United States (e.g., wine with dinner in European countries) and is more available to those under 21 who cannot purchase alcohol legally at home in the United States. However, focusing on problematic drinking (such as binge drinking or experiences of alcohol-related consequences) is perhaps more important as it has the potential to ruin students' experiences abroad. Thus, we evaluated drinking outcomes of total drinks per week, binge drinking days, and alcohol-related consequences, and hypothesized that although we would see an increase in these three assessed drinking behaviors abroad, those increases would be lesser for intervention participants. We also tested the intervention's effects on preventing risky sexual behaviors, as well as preventing any experiences of sexual violence during the abroad trip. We hypothesized that risky sexual behavior would increase less for intervention participants and that intervention participants would be less likely to experience sexual violence abroad than control participants. Though the intervention was geared toward preventing outcomes during study abroad, we also included two assessment points post-return to the United States (first month home, third month home) to explore any potential lasting effects of the intervention. These analyses were exploratory, given that no prior study has evaluated lasting intervention effects on students upon returning home.
2. Methods
2.1 Participants
The study recruited students through emails sent to prospective study abroad students from the study abroad personnel at 40 different US colleges and universities. The study team selected a larger pool of institutions from two Institute of International Education lists (top 50 U.S. institutions with greatest number of students studying abroad and top 50 institutions with the largest percentage of their student population studying abroad during their college experience). The former list was primarily large universities, while the latter was primarily smaller colleges and universities. Forty of these institutions agreed to assist in the recruitment for the study and sent an email to their students with a description of the opportunity to participate in an independent research study for students who were preparing to study abroad during the next semester/quarter. Interested students clicked on a link and completed a brief screening questionnaire to determine eligibility based on: (1) age between 18 and 25 years old; (2) studying abroad in Australia, China, Costa Rica, France, Germany, Ireland, Italy, Japan, Mexico, South Africa, Spain, or the United Kingdom; and (3) studying abroad for anywhere between 8 and 25 weeks (approximately one semester/quarter). Eligibility criteria were set to enroll students who studied in the 12 most popular study abroad locations for the length of the majority of study abroad programs (
Two cohorts of prospective study abroad students were recruited. Cohort 1 consisted of students leaving for study abroad trips in fall 2019, completing trips by December 2019 (N = 650). As described in the protocol for the RCT (
; clinicaltrials.gov identifier NCT03928067), we intended to recruit approximately 1200 participants total. Thus, we enrolled an additional 673 participants into a second cohort of participants, who left for study abroad trips in December 2019/January 2020. The COVID-19 pandemic greatly reduced the length of these study abroad students' trips as many were abroad for a month or less. All programs were canceled, and students were sent back to the United States, precluding us from continuing the study or evaluating outcomes for this cohort. Thus, we focus the current analyses of the intervention's effect on the one cohort of students who completed their study abroad experiences as planned.
Of the 1254 students who expressed interest in the study and requested to participate in the first cohort, 875 were eligible and invited to complete the baseline survey. These invited participants were then randomized based on four criteria to ensure a similar number of types of students per cell for sex (male, female), age (under 21, 21 and older), location of study abroad trip (Europe, non-Europe), and school size (fewer than 10,000 undergraduate students, 10,000 undergraduate students or more), with a projected even number of students per condition (intervention, control). Seventy-four percent of the invited and randomized participants (N = 650) completed the baseline survey and continued into the study, which resulted in 334 in the intervention condition and 316 in the control condition. Table 1 provides descriptive information. Randomization was successful for all demographics and baseline outcomes, except for male sex, where the intervention condition contained a higher percentage of male participants. Given the notable imbalance between the intervention and control groups on sex, we implemented a set of sensitivity analyses that utilized propensity score weights to balance the groups on key baseline confounders (age, sex, race, and baseline outcomes values) and refit our models, described below, controlling for these weights. Results did not differ meaningfully from the unweighted models, suggesting that the lingering sex imbalance did not bias our results.
Table 1Baseline characteristics of analytic sample by treatment arm.
p-Value comparing control vs. intervention baseline values with t-test for continuous characteristics, Fisher's exact test for binary characteristics, and Chi-square test for characteristics with >2 categories. Bolded p-values are p < .05
% or mean
Covariates
Male
20.3
16.5
24.0
.02
Race/ethnicity
.73
Asian/Pacific Islander
10.6
10.1
11.1
Latinx
16.2
14.9
17.4
White
63.4
65.5
61.4
Black/American Indian/Alaska Native/Multiracial
9.9
9.5
10.2
Age 21 or over
22.6
23.7
21.6
.51
Member of fraternity or sorority
35.9
39.7
32.3
.06
Grade point average
.54
2.1–2.5
0.9
0.6
1.2
2.6–3.0
7.0
8.3
5.7
3.1–3.5
37.0
36.6
37.4
3.6–4.0
55.1
54.5
55.7
Study abroad country
.69
Australia
2.9
2.5
3.3
China
2.5
2.5
2.4
Costa Rica
1.4
1.9
0.9
France
9.4
8.9
9.9
Germany
5.5
5.7
5.4
Ireland
5.7
4.4
6.9
Italy
27.1
27.9
26.4
Japan
1.7
2.2
1.2
Mexico
0.3
0.3
0.3
South Africa
0.2
0.3
0.0
Spain
24.5
26.6
22.5
United Kingdom
18.9
16.8
21.0
Completed alcohol or drug prevention programming specific to risk abroad predeparture
53.0
49.4
56.5
.08
Attended sexual assault prevention program specific to sexual assault risk abroad predeparture
92 % of the sample was abroad between 13 and 18 weeks.
15.7 (1.6)
15.8 (1.7)
15.6 (1.6)
.14
Outcomes at baseline
Total drinks per week
4.6 (5.9)
4.6 (6.0)
4.6 (5.9)
.94
Binge drinking
1.7 (2.8)
1.6 (2.8)
1.7 (2.9)
.64
Alcohol consequences
4.8 (5.1)
4.9 (5.3)
4.6 (5.0)
.40
Risky sex events
5.1 (6.7)
5.3 (6.8)
4.9 (6.6)
.48
Sexual violence before trip
49.1
49.0
49.1
1.00
^ p-Value comparing control vs. intervention baseline values with t-test for continuous characteristics, Fisher's exact test for binary characteristics, and Chi-square test for characteristics with >2 categories. Bolded p-values are p < .05
a 92 % of the sample was abroad between 13 and 18 weeks.
All procedures were approved by the local Institutional Review Board, and all procedures were conducted online with links to confidential surveys sent by email. After the baseline survey, participants completed the intervention or were asked to spend 20 to 30 min reviewing their home school's study abroad website for information about safety abroad, including the policies the school had for alcohol use abroad and resources for sexual violence. This request to review information represented the standard practice of most schools for alcohol and sexual violence information prior to going abroad. One month into their study abroad trip, participants were asked to complete a follow-up survey (first month abroad survey) to assess drinking and risky sex outcomes in the past 30 days. Approximately two months later, when students had just a few days remaining before their return to the United States, they were asked to complete a survey of alcohol and risky sexual behavior outcomes in the past 30 days, as well as their experience of any sexual violence over the course of the study abroad trip (last month abroad survey). One and three months after returning home to the United States, they completed surveys of their alcohol and risky sexual behavior outcomes in the past 30 days (first month return survey, third month return survey). Participants received a $20 Amazon gift card for each of the surveys they completed. Surveys took approximately 20 min to complete, and we verified that no participant completed the survey in an impossible length of time (e.g., under 5 min, which would have suggested they were not carefully reading the survey items). Completion rates for these surveys ranged from 86 % to 89 %. After each survey, participants received a list of home school–specific resources, including contact information for their sexual violence prevention coordinators, numbers to report or seek help for sexual violence, and information about mental health and counseling services at their institution and abroad.
2.1.1 Intervention condition
The intervention was named TREK to help facilitate recognition of the study for participants. It is described in detail in prior work (
Brief online interventions targeting risk and protective factors for increased and problematic alcohol use among american college students studying abroad.
). The 20- to 30-minute intervention was based on three modules aimed at preventing escalation of heavy drinking abroad and increasing protective behaviors to reduce risky sexual behaviors and sexual victimization abroad. First, we included evidence-based components known to reduce heavy alcohol use among college students during high-risk events, mainly, the presentation of country-specific drinking norms, which targeted students' predeparture misperceptions of drinking in their host country by both other study abroad student peers and local young adults. Students estimated the drinking of their peer reference groups and the study presented them with graphical depictions of actual moderate norms from 2650 study abroad students peers collected in prior work (
), as well as descriptions of moderate drinking norms within the host country. The study gave them a description of norms theory, including examples of how drinking perceptions can be inaccurate and influence one's own behavior. Second, participants received tips and strategies for engaging with the culture while they were abroad (i.e., sojourner adjustment), such as how to meet local people, best practices for foreign language acquisition, where to find out about local cultural events, and strategies to deal with culture shock abroad. Third, the intervention contained discussions of protective strategies to be used abroad to reduce sexual risk and sexual violence victimization, such as avoiding unwanted sexual experiences abroad (e.g., having a plan with the help of friends to get out of a potentially unsafe situation), engaging in a safe sexual relationships abroad (e.g., use of condoms, ensuring consent was received), and seeking help if one experienced sexual violence or a negative sexual consequence abroad (e.g., list of home campus and national resources). Content specifically stated that sexual violence is never the victim's fault.
The intervention contained a mix of videos and text that could be viewed on a computer, tablet, or phone. A female narrator led participants through the program, with video testimonials of students who had studied abroad in a prior year discussing aspects of drinking norms (e.g., how local people view alcohol as a complement to a meal rather than as a means to get drunk), drinking culture (e.g., how peers that spent most of their time in bars with other American students risked missing out on meaningful social and cultural experiences), aspects of positive sojourner adjustment (e.g., practicing local language skills with peers and locals), aspects of negative sojourner adjustment (e.g., managing anxious feelings or homesickness), and strategies for protecting oneself from risky or unwanted sexual situations (e.g., safe sex practices if choosing to have a sexual relationship abroad, going out with peers who can help them out of an unsafe situation). Drinking data presented were specific to the host country to which students were traveling, whereas sojourner adjustment and safety components were more general to the study abroad experience.
2.2 Measures
2.2.1 Demographics and covariates
Participants completed baseline measures of age, sex, race/ethnicity, study abroad location, current grade point average (GPA), and membership in a fraternity or sorority. Participants also indicated which home institution they attended in the U.S., which was categorized as large (over 10,000 undergraduate students) or small (under 10,000 undergraduate students) for descriptive purposes. Participants also responded to a question about whether they had completed any alcohol or drug prevention program during college (including a class, a meeting in their dorm, or an online program) that was specific to drinking risk abroad (yes/no response). Last, the intervention asked participants whether they had completed any sexual assault program (a class, meeting, or online program) specific to sexual assault risk abroad (yes/no response).
2.2.2 Drinking outcomes
We assessed drinking in the past 30 days at all-time points with the Daily Drinking Questionnaire (DDQ;
), which asked participants about their typical drinking on each day of a typical week in the past 30 days (total drinks per week). Frequency of binge drinking days in the past 30 days was assessed at each time point with an item asking about the number of days the participant drank five or more drinks on one occasion (for males) or four or more drinks on one occasion (for females). At each time point, the 24-item Brief Young Adult Alcohol Consequences Questionnaire (B-YAACQ) assessed whether participants experienced each of 24 different alcohol-related consequences in the past 30 days (
Toward efficient and comprehensive measurement of the alcohol problems continuum in college students: The Brief Young Adult Alcohol Consequences Questionnaire.
). Participants indicated whether they experienced each consequence (e.g., passing out, blacking out, conflicts with peers) or not, yielding a sum score of the breadth of consequences from 0 to 24.
2.2.3 Risky sexual behavior outcome
The study assessed risky sexual behavior (e.g., number of casual partners, limited use of condoms) in the past 30 days with the Sexual Risk Survey (SRS), a 23-item measure designed and validated with college student samples (
). The study modified the SRS to assess past 30-day behavior and included at each time point. The SRS yielded an overall score, with open-ended responses that were recoded into ordinal categories, as specified by the SRS scoring instructions, for generating that overall score (
The study assessed any experience of sexual violence victimization prior to or during college at baseline, and any experience of sexual violence victimization during the abroad trip was assessed on the last month abroad survey. We used a modification of the Sexual Experiences Survey (SES) (
) to assess 12 gender- and sex-specific items that reflected five aspects of sexual violence victimization: nonconsensual sexual contact, sexual coercion, completed sexual assault by force, alcohol- and drug-facilitated sexual assaults, and attempted sexual assault. Specific wording of items can be found in our prior work (
). The sexual violence outcome reflected experience of any sexual violence abroad (i.e., a “yes” response to at least one of the 12 items).
2.3 Analytic plan
Drinking and risky sexual behavior outcomes were count data, and preliminary analyses suggested that we use negative binomial models, as variances were larger than means for outcomes at all-time points. Thus, we conducted four mixed effects negative binomial repeated measures models for the four primary count outcomes of drinks per week, binge drinking days, alcohol-related consequences, and risky sexual behavior. We conducted analyses using SAS Proc Glimmix with nested random effects for individual within school, specifying a negative binomial distribution and using Laplace estimation method. Sexual violence victimization was a dichotomous outcome, and we conducted a mixed effect logistic model using SAS Proc Glimmix with random effects for school. The primary predictor of interest in these models was the interaction between an intervention indicator and categorical time (survey), thus, we used postestimation contrasts to present the intervention effect at each time point (Table 2).
Table 2Intervention effects at each time point for the drinking and risky sexual events outcomes.
Intervention effect at each time point
Outcome
Log-odds
SE
RR
RR LCL
RR UCL
Total drinks per week
Effect at first month abroad
−0.09
0.07
0.92
0.79
1.06
Effect at last month abroad
0.00
0.08
1.00
0.86
1.16
Effect at first month return
−0.01
0.08
0.99
0.85
1.16
Effect at third month return
−0.10
0.08
0.91
0.78
1.06
Binge drinking days per week
Effect at first month abroad
−0.16
0.10
0.85
0.70
1.04
Effect at last month abroad
−0.07
0.10
0.93
0.77
1.13
Effect at first month return
−0.04
0.11
0.96
0.78
1.18
Effect at third month return
−0.18
0.10
0.84
0.68
1.03
Alcohol consequences
Effect at first month abroad
−0.08
0.09
0.92
0.77
1.11
Effect at last month abroad
0.02
0.09
1.02
0.85
1.21
Effect at first month return
0.06
0.09
1.06
0.89
1.27
Effect at third month return
0.02
0.09
1.02
0.85
1.22
Risky sexual events
Effect at first month abroad
−0.34
0.16
0.71
0.52
0.98
Effect at last month abroad
−0.09
0.16
0.92
0.67
1.25
Effect at first month return
−0.02
0.16
0.98
0.72
1.34
Effect at third month return
0.00
0.15
1.00
0.74
1.36
Note. SE = standard error. RR = rate ratio. LCL = lower confidence level. UCL = upper confident level. Effect significant at a traditional p < .05 level is bolded.
All models controlled for sex, race/ethnicity (White, Latinx, Asian/Pacific Islander, Other), age (21 and older, under 21), weeks abroad, fraternity or sorority membership, GPA, whether the study abroad host country was in Europe, and the baseline value of the outcome. Then, the study added additional control covariates to different outcome models depending on the outcome of interest. Drinking outcomes further controlled for whether students received any home institution programming specific to drinking risk abroad (i.e., prior to our program being tested), and the risky sex and sexual violence outcomes further controlled for whether the student attended any home institution sexual assault prevention programming that was specific to sexual assault risk abroad. We note that baseline measures for sexual violence victimization were represented by an indicator of any sexual violence victimization prior to going abroad (i.e., in college or precollege). Negative binomial models produced rate ratios (RR), which are the exponentiated log odds estimates and represent the ratio of incidence rates in the intervention group versus the control groups. An RR of approximately 1 indicated no effect, whereas an RR of <1 indicated a lower risk for intervention participants, and an RR >1 indicated a higher risk for intervention participants. We focus on interpreting the magnitude of the RR intervention effects and the evidence provided by the 95 % confidence intervals. Figures are included to show the means of each outcome by condition at each time point to facilitate interpretation.
3. Results
3.1 Drinking outcomes
As expected, participants increased drinking from baseline levels, with students increasing total drinks per week (see Fig. 1) and number of binge drinking days (see Fig. 2) in the past 30 days by nearly double during their first month abroad. This heavy drinking stabilized throughout the remainder of the study abroad trip, and returned to near baseline levels upon return home.
Table 2 contains our estimated intervention effects at each follow-up time point. For drinks per week during the first month abroad, the intervention group's rate of drinking was 0.92 times (95 % CI = 0.79, 1.06) lower than the rate of drinking in the control group, with those in the control group drinking 9.2 drinks per week on average and intervention participants drinking approximately one fewer drink per week (8.4 drinks per week). At the third month return survey, the intervention group's rate of drinking was 0.91 times (95 % CI = 0.78, 1.06) lower than the rate of drinking in the control group (6.4 drinks per week for control, 5.8 drinks per week for intervention). For binge drinking days during the first month abroad, the intervention group's rate of binge drinking was 0.85 times (95 % CI = 0.70, 1.04) lower than the rate in the control group, with those in the control group binge drinking on 3.7 days during the first month abroad and intervention participants binge drinking on approximately 3.1 days during the first month. On the third month return survey, the intervention group's rate of binge drinking was 0.84 times (95 % CI = 0.68, 1.03) lower than the rate of binge drinking in the control group (2.5 binge drinking days for control, 2.1 binge drinking days for intervention).
For alcohol-related consequences, participants showed overall decreases in the number of consequences at first month abroad, reducing consequences by approximately one and a half to two consequences during that time (see Fig. 3). They then increased their consequences back to predeparture levels during the last month abroad, with decreases during the return months. RRs and associated 95 % confidence intervals presented in Table 2 suggested similar levels of consequences across intervention and control participants.
Fig. 3Alcohol-related consequences outcome in the past 30 days.
Fig. 4 shows means for the risky sexual behavior outcome. Participants were relatively similar in their reports of risky sexual behavior at baseline, during the last month abroad, and upon return home. Though control participants also reported fewer risky sexual behaviors during the first month abroad than at baseline, there was a statistically significant intervention effect, such that those in the intervention group engaged in a pattern of sexual risk behaviors that was a 0.71 step below the risk behaviors of the control group (95 % CI = 0.52, 0.98; see Table 2).
That is, intervention participants reported significantly less risky sexual behavior during the first month abroad compared to controls. The RRs and associated 95 % confidence intervals for all other time points suggested similar levels of risky sexual behaviors across intervention and control participants.
Regression results for sexual violence suggested similar levels of sexual violence across intervention and control participants, with an intervention effect of RR = 0.84, SE = 0.21 (CI = 0.55, 1.26). Between one-fifth and one-quarter of participants experienced sexual violence during their trip abroad, with just about 3 % more control participants (or approximately 6 participants) experiencing sexual violence abroad than intervention participants (24.2 % of the control; 21.2 % in the intervention).
4. Discussion
This study represents the first empirical test of a brief, online, predeparture intervention designed to prevent heavy drinking, risky sexual behaviors, and sexual violence among American college students studying abroad. We evaluated effects of the intervention on drinking, risky sex, and sexual violence outcomes over the course of study abroad trips as well as explored lasting effects upon return home to the United States, with a large sample of 650 students studying abroad in 12 countries from 40 home colleges and universities in the United States. Although we did not find statistically significant effects, effect sizes during the first month abroad and the last month after return home suggested that intervention participants drank fewer drinks and engaged in binge drinking less frequently than control participants. The only statistically significant effect of the intervention was for risky sexual behaviors during the first month abroad, where intervention participants reported significantly fewer sexual risk behaviors compared to control participants. The study found no effects of the program on any outcomes during the final month of study abroad or the first month post-return from studying abroad. No observable effects of the intervention occurred on prevention of sexual violence victimization abroad.
Drinking increased for all participants while they were abroad, with the largest increases occurring during the first month abroad. The lack of statistically significant findings between conditions during this escalation somewhat mirror the small effects typically observed in brief alcohol interventions with college students (
), who are often not motivated to change their drinking and do not seek out alcohol treatments. Thus, from a public health point of view, and for a group that currently receives very limited prevention programming prior to going abroad (
), differences of about one drink less per week and about one-half day less of binge drinking per month could be potentially meaningful. However, as this study and others demonstrate, drinking escalates drastically for students during that first month abroad, with these patterns then continuing over the course of the trip (
). Thus, approaches that can achieve substantial effects on drinking behavior and that can temper escalation during this first month may be crucial to help students establish moderate levels of drinking during the entire period abroad. Prevention of this escalation would be important, in particular, to limit students' experiences of alcohol-related consequences.
We saw an overall decrease in risky sexual behavior for both groups during the first month abroad, but students who received the intervention reported engaging in even less risky sexual behavior during that first month—a small and significant effect. Similar to drinking behavior, preventing risky sexual behavior in the first month is important for establishing patterns during the study abroad trip. Effects did not last; however, and the study found no differences in the last month abroad. Still, prevention of potential risky sexual behaviors during the first month abroad (e.g., condomless sex, sex with multiple partners) has many implications, as students are first getting to know each other and creating reputations for themselves with a new peer group. Focusing on adapting to the new foreign environment and establishing friendships without the added potential stressors of managing consequences from risky sexual behaviors (e.g., sexually transmitted infections, regretted behavior) could help students to have a positive overall experience abroad.
Rates of sexual violence were similar for students in both conditions, with rates of sexual violence comparable to samples of study abroad students who did not receive any intervention (
). Sexual violence abroad is concerning and the intervention's inability to prevent occurrence of sexual violence could be attributed to multiple factors. First, though the content of the intervention included aspects directed at perpetrators (e.g., ensuring that consent is received), the content was primarily targeting potential victims (e.g., having a plan to get out of a potentially unsafe situation). The focus on potential victims places unfortunate burden on potential victims and if the intervention were directed at preventing perpetration, we may have seen better effects. Still, research shows that while perpetrators can be other students, they are often local people (
), and intervening to prevent perpetration by members of a foreign country could be difficult. Including institution-level policy changes to supplement the individual intervention (e.g., providing better supervision of students, requiring safer housing options) may be necessary to help prevent sexual violence victimization among students. Also, drinking increased while abroad, even among those who received the intervention, which may have placed these students at greater risk, as studies have shown that heavy drinking is associated with sexual violence victimization abroad (
). Last, though the intervention encouraged students to intervene if peers were in a potentially dangerous situation, such content was limited. Bystander interventions have been shown to be helpful on campus (
) and a greater focus on content related to helping others out of potentially dangerous situations could have helped to achieve stronger effects.
Though we did not observe many significant effects across drinking, risky sex, and sexual violence outcomes during the post-abroad follow-up months, the exploratory effects of the intervention on drinking during the third month return are interesting. As the intervention content was completely focused on the abroad context, we did not hypothesize that the study would find lasting effects. It may be that aspects of the intervention with respect to drinking translated to their peer context at home. For example, students had learned that they may be inaccurate in their perceptions of peer drinking and they may have begun observing their peers back on campus, recognizing that their peers do not drink as much as they previously thought. As another example, students in the intervention learned tips for engaging with the culture more while they were abroad, and they may have adopted some of those practices and brought them back to the United States, such as drinking moderately at dinner instead of in bars and clubs or engaging in conversations with new people rather than sticking with the same friends and drinking heavily. These ideas would need to be tested further in follow-up studies that could also examine long-term effects of the intervention.
Similar to prior studies—which showed that schools offered limited alcohol and sexual information on their study abroad websites and that most students said they did not receive alcohol or sexual violence prevention programming prior to departure abroad (
)—we found that only about half of the participants in our sample recalled receiving any alcohol or sexual risk programming from their institution prior to departure (53 % received alcohol programming, 46 % received sexual violence programming). Even accounting for receipt of such programing in our analyses, we found large increases in drinking behavior abroad and concerningly high rates of sexual violence. Too few targeted, effective resources exist for study abroad students. Colleges and universities sending students abroad should provide students with accessible prevention programing to address heavy drinking and risky sexual behavior while abroad. Although the current intervention represents an accessible program that can be easily available to students across institutions, further work should determine the best timing of the intervention and whether short booster sessions once students have acclimated to their country and new peer context may help to reduce heavy drinking and sexual risk behavior.
4.1 Limitations
Results should be interpreted in light of methodological limitations. Although the sample is large and represents students from 40 different home institutions in the United States across 22 states and Washington D.C., the study abroad offices at the institutions opted in to allow their prospective study abroad students to participate, which may represent sampling bias. In addition, by not randomly sampling from the study abroad student population and by restricting our eligibility criteria, the sample was under-representative of male and White students compared to the actual study abroad student population (
). We also tested the intervention with students traveling to 12 specific countries to personalize the drinking norms content. These common host countries are skewed European, yet the included countries represent 60 % of all students abroad, and we included representation of the most popular Asian and Latin American host countries (
). Further, though we made some intervention content country-specific (peer normative drinking presentation), the tips and strategies to promote sojourner adjustment were generalized. Sojourner adjustment is possibly something that differs across countries, as each country has unique cultural aspects that could be specifically addressed.
In an effort to standardize the assessment periods, we also focused on students who were abroad for about three months, which also represents the majority of students completing programs, but neglects the increasing number of students studying abroad for one month or less. Given the limited duration of their trips abroad and inability to fully absorb the culture during that time, the sojourner adjustment aspects of the intervention may not be appropriate (i.e., immersion into a culture may be impractical if abroad for just a week or two). Moreover, though attrition was limited, the use of longitudinal self-report measures collected via online survey could be associated with response bias. However, research suggests that self-report data are reliable and confidential surveys enhance reliability (
). As mentioned, we lost about half of the preplanned sample due to the COVID-19 outbreak abroad, which may have limited our ability to detect significant intervention effects and explore differences in effects across more than one cohort of students. Last, we focused on sexual violence victimization outcomes. Although the intervention contained information targeting potential perpetrators (e.g., getting and receiving consent), we expected to be underpowered to report effects given low base rates of admitted perpetration in the sample.
4.2 Conclusions and future directions
Like 21st birthday celebrations and spring break trips (
), study abroad represents a high risk event for college students during which heavy drinking, risky sexual behavior, and sexual violence often occur (
), which allows for an extended period where harms can occur. To date, few schools offer programming prior to departure, and if they do, programs are not evidence-based or are not received by students.
The study team developed the current intervention utilizing risk and protective strategies shown to be associated with alcohol and sexual risk for American students abroad. Future work should refine the current intervention, perhaps with booster sessions or redesigned content. Booster sessions, likely delivered during the initial weeks abroad when drinking spikes and then perhaps again throughout the trip, may be necessary to keep drinking and risky sex from increasing during the rest of the study abroad trip. As the predeparture period may be hectic with planning for the trip, ensuring academic credits will transfer, and saying goodbye to loved ones, delivering (or re-delivering) the intervention once abroad may give students more time to absorb the information once they are settled abroad and have time to focus on intervention content. This delivery timing may be particularly important for the drinking norms presentation. It is likely critical for students to be able to confirm the veracity of the newly presented drinking norms about this novel environment by observing the behavior of peers and local people within the new cultural environment. By using the tips and strategies offered around sojourner adjustment, primarily through meeting local people and participating in cultural events in the new environment, students can see themselves that drinking is actually quite moderate abroad.
Research also suggests that heavier drinkers in college may self-select into study abroad programs (
Heavier drinking american college students may self-select into study abroad programs: An examination of sex and ethnic differences within a high-risk group.
), making this a group already at high risk even before entering an environment with known heightened risk. Thus, study abroad students may need more than a brief self-directed online approach to experience sustained effects on drinking, risky sexual behavior, and sexual violence. These students may need more intensive programming to better prepare them for their abroad experiences, perhaps receiving programming both before and during their abroad experience. Online self-directed programs such as our developed intervention allow for more students to receive programming with the use of less staff resources, which greatly expands access to evidence-based care to a population of students in great need. Findings provide an important first look at how a brief online intervention may affect drinking and risky sexual behavior in the first month of study abroad, and suggest that the intervention could be enhanced with booster sessions once students are engaged abroad and able to better absorb and apply the content. Rigorous evaluation of modifications to this intervention and other developed programs and policies will be essential.
Data available for research purposes upon request to the first author.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
This work was funded by a grant from the National Institute on Alcohol Abuse and Alcoholism (R01AA025909, “Online Intervention to Prevent Risky Behaviors During College Student Study Abroad Experiences”) awarded to Eric R. Pedersen. The authors wish to thank the 3C Institute for online survey hosting, Andy Langdon at Good Pictures for filming the intervention content, Michael Woodward for assistance with data collection and retention efforts, and the study abroad directors who asssisted us with recruitment by sharing our study opportunity with their prospective students.
References
Abbey A.
Alcohol's role in sexual violence perpetration: Theoretical explanations, existing evidence and future directions.
Toward efficient and comprehensive measurement of the alcohol problems continuum in college students: The Brief Young Adult Alcohol Consequences Questionnaire.
Hazardous alcohol use and cultural adjustment among U.S. college students abroad in Italy: Findings and recommendations for study abroad staff and researchers.
Examining the relationships between acculturation orientations, perceived and actual norms, and drinking behaviors of short-term American sojourners in foreign environments.
Heavier drinking american college students may self-select into study abroad programs: An examination of sex and ethnic differences within a high-risk group.
Brief online interventions targeting risk and protective factors for increased and problematic alcohol use among american college students studying abroad.
Not all those who wander are lost: Examining the impact of sojourner adjustment and drinking motives on alcohol consequences experienced by Americans studying in foreign countries.
Demographic and predeparture factors associated with drinking and alcohol-related consequences for college students completing study abroad experiences.
00001-2&id=gr1.jpg){kind=link}
00001-2&id=gr2.jpg){kind=link}
00001-2&id=gr3.jpg){kind=link}
00001-2&id=gr4.jpg){kind=link}