Abstract
Objectives:
This paper explores the extent to which curriculum infusion (CI) impacted undergraduate students’ alcohol use, perceived peer alcohol use, use of protective behavioural strategies, academic performance and course engagement.
Design:
Two faculty members infused content on norms and protective behavioural strategies into selected academic courses.
Setting:
This study was conducted in a small liberal arts university in the northeastern part of the USA.
Methods:
A sample of 215 undergraduates from research design and statistics courses in the fall semester of 2013 and the spring and fall semesters of 2014 were selected. CI was experienced by 100 students and 115 students comprised the comparison group. Surveys were distributed at the end of each semester to these selected groups and the results were analysed.
Results:
Students who witnessed CI showed no improvement in alcohol use outcomes; however, there was a trend level effect for frequency of use. CI students reported higher expected final grades and similar levels of course engagement to comparison group students.
Conclusion:
CI is a topic that requires in-depth scientific research to assess its value on positively affecting college drinking consequences. From a pedagogical standpoint, CI was successful since it did not negatively impact on students’ engagement and it was associated with higher expected final grades.
Keywords
Introduction
US National data indicate that over 60% of students consumed alcohol in the past 30 days and that approximately 33% of students consumed five or more drinks at least once within the past 2 weeks (American College Health Association [ACHA], 2013). Additionally, 34% of university counselling centre directors reported an increase in the number of clients with alcohol problems compared to 5 years ago (Gallagher, 2013). Rates of alcohol use among college students are particularly concerning given that alcohol use is associated with deleterious consequences, such as injuries and assault (Hingson et al., 2009).
One prevention approach to address alcohol use among college students is curriculum infusion (CI). As defined by the Network for the Dissemination of Curriculum Infusion (NDCI, 1999), CI is the act of interweaving prevention information concerning alcohol and other drugs into course content. CI involves the dissemination of substance use prevention content to students through existing academic courses (NDCI, 1999). This approach has the advantage of involving faculty in prevention efforts and the potential to reach those students who may not typically attend campus-based prevention programmes or events (NDCI, 1999).
Existing studies have found support for the use of CI to affect substance use–related behaviours and attitudes. For example, one study found that the use of CI resulted in fewer alcohol-related consequences (e.g. driving under the influence, assault and self-injury) among college students (White et al., 2010). Lederman et al. (2007) found that the infusion of a norms-based simulation in undergraduate communication courses successfully changed students’ misperceptions about college drinking. Students’ knowledge and attitudes regarding alcohol and other drugs were impacted in a study that utilised CI, although effects were only found in small classes (Cordero et al., 2010). An additional study that infused content on information about campus resources into an undergraduate health course found a significant positive effect on awareness of campus resources (Riley et al., 2005).
However, more research is needed to establish the extent to which CI is an effective strategy to impact actual rates of alcohol and other drug use since almost no published literature exists to provide support for the use of this approach to impact rates of use. In fact, some researchers have found no effect on rates of use (White et al., 2010). In contrast, CI training for faculty has been cited as impacting binge drinking among college students when used in combination with other prevention approaches (Ziemelis et al., 2002), and CI was found to be one approach associated with favourable substance use prevention outcomes (Licciardone, 2003). Others have described CI as a highly promising educational approach (Vicary and Karshin, 2002).
Interestingly, little research exists that investigates the extent to which the use of CI as a pedagogical approach impacts students’ learning and engagement. Does this approach enhance or interfere with learning course content? Is engagement in the course increased or compromised when CI approaches are adopted? While more research is needed on the viability of CI as a prevention approach, research is also needed to better understand its impact on students’ learning.
This study seeks to add to the understanding of the effectiveness of CI. Specifically, it investigates the extent to which CI impacts rates of alcohol use among a sample of US college students. This study also explores the impact of CI on students’ perceptions of alcohol use and the use of protective behavioural strategies related to drinking.
In the light of the lack of research on CI as a pedagogical approach, the impact of CI on academic performance and course engagement is also investigated. If CI is found to promote students’ mastery of course material, the infusion of health-related information could be considered by academic institutions as a potential prevention approach that would simultaneously enhance learning.
Methods
Study design
A grant was received from the US Department of Mental Health and Addiction Services (DMHAS) to utilise CI as a pedagogic prevention approach. The project aims involved the infusion of content and the use of a CORE dataset in undergraduate courses to correct misperceived norms regarding alcohol use and expose students to information on prevention-related information.
The two participating faculty members selected courses (statistics and research design and analysis) that could accommodate this approach and the use of a dataset. Several collaborative informational sessions were held between participating faculty and the wellness centre staff to become knowledgeable about CI as an approach, to work on course development/revisions and to align research study evaluation efforts with objectives and planned infused course content. Periodic meetings were held to discuss the ongoing implementation of CI and to discuss any challenges.
Content on norms regarding alcohol use and information on prevention-related information was infused in both class lectures and assignments. A second focus was to expose students to information on protective behavioural strategies related to alcohol consumption (i.e. set a limit, eat before drinking and alternate drinks with water).
This study was a quasi-experimental posttest-only design with a comparison group. This design was chosen due to the inability to randomise which instructors could implement CI, the inability to pretest on key variables (e.g. course engagement, expected final grade in course) and to avoid timing issues related to alcohol use behaviours. This design also helped to avoid testing effects and to minimise response burden.
Participants
Study participants were 215 college undergraduate students enrolled in a mid-sized university in the northeast USA. Students ranged in age from 17 to 20 years with an average age of 19 years. Participants were predominately women (81%) and Caucasian (88%). The university as a whole (undergraduate and graduate students) comprises 67% female students and 19% minority ethnic students. Undergraduates comprised the majority of the sample with 38% first-year students, 44% second-year students, 17% third-year students and 1% fourth-year students. Students were enrolled in either a first-year college course (required for psychology majors/minors and open to all other majors) or a mid-level collegiate research design and analysis course (required for all psychology majors/minors). An official university institutional review board exemption approval was obtained prior to data collection. This was granted due to the fact the study did not involve the collection of identifying information and the focus being on the examination of an instructional technique.
The CI group comprised 100 students and the comparison group comprised 115 students. The comparison group was selected by the availability of professors teaching the same courses within the same academic year who were not doing any form of CI. An examination of equivalency between the two groups resulted in no significant differences regarding average age or percentage of White students. However, the proportion of male students and female students differed significantly between the two groups (χ2(1, N = 215) = 4.26, p = .04), with female students being more likely to be in the CI group. Similarly, year of study differed between the groups (χ2(1, N = 215) = 7.93, p = .05) with more first-year and second-year students likely to be in the CI group.
Data collection
Students were asked to complete a brief 7- to 10-minute online survey. The online survey was administered during the last week of classes by a counsellor from the university wellness centre who introduced the survey and provided the website link. Students were informed both verbally and in the online survey that participation was voluntary, confidential and would not impact upon their grade. Prior to entering the survey, students were given a description of its purpose, were made aware of the use of data and were asked for consent. The survey included basic demographic questions on gender, race/ethnicity and age, as well as questions regarding course, semester and current college performance (measured in grade point average [GPA]).
The survey included two items on alcohol use and two items on perceived alcohol use adapted from existing national surveys (e.g. Brief Alcohol Screening and Intervention for College Students [BASICS]). In the survey, a drink was defined as a 12-ounce bottle of beer, a 4-ounce glass of wine, a wine cooler, a 1.5-ounce shot of spirits or a mixed drink made with a 1.5-ounce shot. Alcohol use was assessed with an item regarding average consumption with the corresponding response set of ‘Never’, ‘Monthly or less’, ‘2–3 times/month’, ‘Once/week’, ‘2 times/week’, ‘3 times/week’ and ‘4 or more times/week’. A second item assessed alcohol use in terms of the number of drinks and included response choices: ‘None’, ‘1–2’, ‘3–4’, ‘5–6’, ‘7–8’, ‘9–10’ and ‘11+’. Perceived alcohol use was measured using two items that paralleled the above items; however, instead of focusing on use, the item stems asked about the perceived alcohol use of the average student at their university.
The use of protective behavioural strategies was assessed with six items (α = .81) adapted from commonly used scales, such as the National College Health Assessment (NCHA). A sample item is ‘If you drink, to what extent do you alternate between alcoholic and non-alcoholic drinks?’. Response choices included ‘Never’, ‘Rarely’, ‘Sometimes’, ‘Most of the time’, ‘Always’ and ‘N/A I don’t drink’.
Engagement in the course was assessed with two items. The first item, ‘How engaged are you in this course?’, had a four-point response set ranging from ‘Not at all engaged’ to ‘Extremely engaged’. The second item, ‘How engaged are you in this course compared to other courses you are taking this semester?’, had a five-point response set ranging from ‘Less engaged than in any of my other courses’ to ‘Much more engaged than in any of my other courses’. These two items were adapted from items used in past research (Handelsman et al., 2010).
The survey included one item that assessed student’s expected final grade in the course. Response choices included letter grades from A to F.
Results
A chi-square analysis was conducted to determine whether students in the CI group were more likely to drink less than students in the comparison group. The number of drinks on a typical night of drinking was dichotomised as none to moderate (four or less drinks) and heavy (five or more drinks). No significant difference was found regarding the proportion of none to moderate drinkers (χ2(1, N = 215) = .510, p = .48). Regarding the frequency of alcohol use, the proportion of participants that reported alcohol use less than three times per week was not significantly different for students in the CI group; however, there was a trend in that direction (χ2(1, N = 215) = 2.77, p = .09; see Table 1).
Chi-square analyses of alcohol use and perceived alcohol use among curriculum infusion group students compared to comparison group students.
CI: curriculum infusion.
Similar to the data on students’ number of drinks on a typical night, the perceived number of drinks on a typical night for the average student was dichotomised as none to moderate (four or less drinks) and heavy (five or more drinks). A chi-square analysis yielded no significant difference in proportions between the students in the CI group and the students in the comparison group (χ2(1, N = 215) = .035, p = .85). Similarly, the proportion of participants that reported perceived alcohol use less than three times per week was not significantly different for students in the CI group compared to that in the comparison group (χ2(1, N = 215) = 1.94, p = .16; Table 1).
An independent sample t test was conducted to determine whether the average score for the use of protective behavioural strategies was significantly different when comparing students in the CI group and students in the comparison group. A total score was calculated by adding six protective behavioural strategy items that were scored from 1 to 5 according to the frequency of use. The mean scores for the CI students (M = 21.99) and for the comparison students (M = 21.95) were almost identical (t(177) = .056, p = .96). An examination of the likelihood of using each strategy most of the time or always also resulted in no significant differences between the two groups.
A chi-square analysis was conducted to determine whether the students in the CI group were more likely to perform better academically than the students in the comparison group. The proportion of participants that reported an expected grade of B+ or higher was significantly different for students in the CI classes (χ2(1, N = 215) = 5.21, p = .02, Table 2).
Analyses of course engagement and academic performance among curriculum infusion group students compared to those among comparison group students.
CI: curriculum infusion.
Both groups, the students in the CI courses (M = 2.70) and the students in the comparison courses (M = 2.75) reported similar levels of engagement in the course (t(212) = −.469, p = .64). Similarly, students’ engagement in the course compared to other courses was not significantly different between students in CI group (M = 3.28) and students in comparison group (M = 3.12) (t(213) = 1.317, p = .19, Table 2).
Discussion
This study focused on several potential alcohol use-related and pedagogical outcomes of the use of CI in selected undergraduate courses. First, the extent to which CI impacted the amount and frequency of alcohol use was assessed. Similarly, the impact of CI on college students’ perceptions of the amount and frequency of alcohol use by peers was also examined. Despite the infusion of content and activities designed to address students’ perceived alcohol norms, there was no significant difference in the amount and frequency of alcohol use or in the perceived amount and frequency of alcohol use between students who experienced CI and those who did not. Results on the frequency of use were trend level in the direction of less frequent use among students in the CI group. However, the lack of an effect on students’ alcohol use is consistent with other studies (e.g. White et al., 2010).
A second aim of the study was to examine the extent to which the use of CI resulted in the use of protective behavioural strategies related to alcohol use. The students who received CI had similar levels of the use of these strategies than those who did not receive CI. No other published studies that investigated this outcome were found; however, one study found that the use of CI resulted in fewer alcohol-related consequences and related risk behaviour (White et al., 2010).
A third aim of the study was to investigate the impact of CI on academic performance and course engagement. The infusion of health-related materials in academic courses is an interesting prevention strategy, however, only to the extent that it does not negatively impact on students’ learning experiences. Interestingly, in this study, students from the CI group were more likely to report expecting a higher final grade than the students who did not experience CI. Expected grade was chosen as an outcome measure in order to avoid potential identifiers in the data and studies show that predicted grades have been related to final grades (e.g. Burns, 2007). This finding is a new contribution to the existing research on CI and represents an important pedagogical consideration. Does the utilisation of an interwoven theme facilitate knowledge acquisition? Does the inclusion of health content immediately relevant to students make course material more meaningful and thus more likely to be retained?
Students’ engagement in the course was also examined as a potential pedagogical outcome. No differences were found between students in the CI courses and students in the comparison courses on engagement in the course or engagement in the course compared to other courses. An interpretation of this finding could be twofold. It could be that CI as a pedagogical approach did not result in any new or different learning experience for the students compared to other classroom experiences. This lack of difference is unfortunate and somewhat contrary given the aforementioned results in which there is some support for CI as influential on academic performance. Alternately, the lack of a difference in course engagement between the two groups of students could be understood as a positive result. Specifically, the infusion of health-related content did not result in disengagement from the course or less engagement than in other courses. Students were not disengaged by health messaging and communication of alcohol-related norms from adult figures, and this content did not handicap the students’ experience of the course. Future research is needed to replicate these findings and better understand this outcome.
This study was characterised by some limitations. The use of a quasi-experimental posttest-only design was due to an inability to randomise participants or classes and other methodological considerations. Additionally, several key outcomes, such as course engagement, could not be assessed at pretest. Nonetheless, the use of this design resulted in differences between groups that complicate a clear understanding of the outcomes. Specifically, more female students and more first- and second-year students were in the CI group and thus unmeasured factors related to these characteristics could be influencing results. The absence of a pretest also prevents an understanding of baseline rates and an assessment of change over time. Future research using more rigorous designs is needed.
The courses used in this study resulted in a particular sample that may limit the generalisability of the findings, although the use of convenience samples is common among existing research on CI (e.g. Riley et al., 2005; White et al., 2010). The sample consisted of a large number of psychology majors/minors. Psychology students may be more knowledgeable about health-related content and perhaps engage in health-related behaviours at a different rate than other students who by choice of major have had less exposure to such content. Additionally, the students in the statistics courses were more likely to be eventual majors in one of the social sciences or health professions, as introductory level statistics is a common required course for those academic tracks.
The short time frame of the study and the use of a very brief survey also are limitations to this research. It may be that behaviour change, such as assessed with the survey items on use of alcohol and use of protective behavioural strategies, was difficult to capture within the time frame of a 16-week semester. Additionally, the survey used was necessarily very brief in order to increase the rate of response and minimise the use of class time.
A final limitation is the level of standardisation of the use of CI across the courses in the CI group. The two participating faculty members in the CI courses collaborated and agreed on the content to be infused and the use of a specific database was standardised across all CI courses. However, the delivery of that content varied across courses due to instructor style and course subject.
Conclusion
Future research that addresses the methodological challenges of evaluating this classroom-based, faculty-led prevention strategy is needed. Specifically, large-scale research utilising longer time frames and more comprehensive measures is needed to add to the understanding of the impact that CI can have on alcohol use and alcohol use–related behaviours and attitudes. More rigorous designs and the use of samples that are either very diverse or very homogeneous are also needed. Research is also needed to better understand what dose and what level of standardisation are needed to achieve desired outcomes and the extent to which dissemination at that level is feasible. An implication of this study is the importance of investigating pedagogical outcomes along with prevention-related outcomes.
Other research has found that CI can impact on important alcohol-related variables such as alcohol-related consequences (White et al., 2010) and students’ misperceptions about drinking (Lederman et al., 2007). These are important areas of focus for prevention. CI is also highly promising as a prevention approach due to the advantages of involving faculty in prevention efforts and the potential to reach those students who may not typically attend campus-based prevention programmes or events (NDCI, 1999). CI training for faculty was cited as impacting binge drinking among college students when used in combination with other prevention approaches (Ziemelis et al., 2002). Additional research on CI that helps to elucidate expected outcomes and best practices in terms of implementation will enable prevention researchers and college administrators to leverage these advantages to better address the ongoing issue of alcohol use among college students.
Footnotes
Acknowledgements
The authors would like to acknowledge Jennifer Genzen for her development of some of the alcohol use items used on the survey.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work is supported by the Department of Mental Health and Addiction Services (DMHAS) Connecticut Healthy Campus Initiative.