MOST of Us ® Newsletter Article

Untitled Page

In Their Own Words:
Student Characterizations of Protective Behaviors to Prevent Alcohol Harm

Charles K. Atkin
Sandi W. Smith
Katherine Klein
Edward Glazer
Dennis Martell

Michigan State University

Abstract

A considerable number of college students engage in heavy consumption of alcohol and often suffer negative consequences from it. While the most important goal is to persuade students to reduce their alcohol consumption, a secondary goal is to identify and persuade them to use protective behaviors that reduce the likelihood of harm from heavy drinking. This research endeavor was designed to augment the standard list of 9 protective behaviors that are identified on the National Health College Assessment (NCHA) instrument that is often employed by researchers and health practitioners. Forty-three separate protective behaviors were listed in response to an open ended question. One possible higher order scheme of dimensions is offered here, as is a discussion of the findings

In Their Own Words:

Student Characterizations of Protective Behaviors to Prevent Alcohol Harm
With the considerable number of college students who frequently consume large quantities of alcohol (Wechsler et al., 2002) there is substantial emphasis placed on reducing the dangerous consequences of heavy drinking at college campuses across the nation. Recently, researchers have examined protective behaviors that students engage in to reduce their likelihood of harm from heavy drinking. These studies have supplied participants with a list of protective behaviors and asked them to indicate which of them they use. The present study measures the frequency with which students used a standard list of protective behaviors that have been generated previously. To augment the previous literature, this study aims to generate a more complete list of protective behaviors by using an open-end technique. A representative sample of students was asked to report their own protective behaviors, and the responses were coded into an array of categories. These categories are then grouped into a set of higher-order dimensions that emerged from the behaviors generated by the students.

High Risk Alcohol Consumption.

National surveys continue to reveal that college students put themselves at risk with the amount and frequency of their alcohol consumption (Knight et al., 2002; O�Malley & Johnston, 2002; Wechsler et al., 2002). Compared to all drinkers, the highest proportion of excessive drinking occurs among those 18 to 20 years of age (Serdula, Brewer, Gillespie, Denny & Mokdad, 2004).This is because approximately 90% of alcohol consumed by those under the age of 21 occurs during heavy episodic drinking (Office of Juvenile Justice and Delinquency Prevention, 2001).This is particularly problematic as heavy drinking may cause negative consequences, most of which are not long term such as liver damage, but instead result from a heavy episode of drinking (Neighbors, Oster-Aaland, Berstrom & Lewis, 2006). Consequences from excessive drinking may include injuries, alcohol poisoning, sexually transmitted diseases, unintended pregnancy, children born with fetal alcohol syndrome, stroke, and neurological damage (Centers for Disease Control and Prevention [CDC], 2006). Many of these consequences could be avoided by using protective behaviors when consuming alcohol.

Despite the large number of students who engage in heavy drinking and the numerous negative consequences from their alcohol consumption, Lewis and Thombs (2005) found that most students perceived no risk from drinking. Lewis and Thombs (2005) also found that, for the most part, perceptions of risk had negligible correlations with measures of alcohol involvement after controlling for other variables.

Protective Behaviors

Because many students have not altered their alcohol consumption in order to reduce risks associated with heavy drinking, it is important to examine protective behaviors, especially behaviors that are not centered on alcohol consumption, which may be used to protect against these harms. Martens et al., (2004) define protective behavioral strategies as �behaviors that individuals can engage in while drinking alcohol in order to limit negative alcohol-related consequences� (p. 390). There is evidence that the use of protective behaviors reduces the number of consequences from heavy drinking (Martens et al., 2004; Benton et al., 2004). Therefore, if students are going to drink, it is important that they protect themselves from possible consequences.

Health communicators can help to prevent potential negative consequences of alcohol consumption by knowing which protective behaviors students utilize, and this information can be used to generate health messages based on several theoretical underpinnings. For example, knowing which behaviors students frequently engage in can be used to create social norms campaign messages (Haines & Spear, 1996, Perkins & Berkowitz, 1986, Perkins & Craig, 2006) which are based on behaviors that are engaged in by the majority. Alternatively, fear appeals (Witte, 1992; Witte, 1994; Witte, Meyer & Martell, 2001) can be created centered on negative consequences that can befall students if they do not use certain protective behaviors. In designing campaigns, formative evaluation data identifying promising but infrequently employed protective behaviors is useful in selecting underutilized behaviors to be promoted to student audiences.

In addition to implications for prevention campaigns, certain modes of protection feature interpersonal communication and coordination with peers. Individuals may rely on friends to help regulate their alcohol consumption through verbal or environmental intervention, to work together in arranging for safe transportation, and to provide safe companionship in high-risk situations such as a wild party or a late-night walk home.

Due to the importance of protective behaviors in avoiding harmful alcohol related consequences, studies have begun to identify certain protective behaviors that are used by students (Benton, et al, 2004; Haines, Barker & Rice, 2006; Martens et al., 2005). See Table 1: Previous Studies and Protective Behaviors for a list of protective behaviors from previous studies. Many of the protective behaviors that have been identified center around alcohol consumption (e.g. Benton et al., 2004; Martens et al., 2005; Haines et al., 2006). All three of these studies have identified protective behaviors such as �alternating alcohol with nonalcoholic beverages,� �avoiding drinking games,� and �determining, in advance, not to exceed a set number of drinks� (Benton et al., 2004; Martens et al., 2005; Haines et al., 2006). Studies typically identify only two or three behaviors that are not directly related to alcohol consumption, such as �only drinking in safe environments� or �hanging out with trusted friends� (Benton et al., 2004).

A more complete list of protective behaviors students use might be generated because typically the investigators have supplied students with a list of protective behaviors and asked them to rate their likelihood of use rather than asking students to generate a list of protective behaviors that they actually use. Although providing students with a list provides an initial basis for looking at protective behaviors, there should be greater breadth if the students themselves were asked to comprehensively describe the behaviors that they use. Therefore, the main research question of this study is to identify, in addition to the protective behaviors other researchers have previously found (i.e., Benton et al., 2004; Martens et al., 2005; Haines et al., 2006), what other protective behaviors students engage in to avoid harmful consequences from consuming alcohol. This study, therefore, seeks to extend the work of previous authors by discovering additional protective behaviors that students engage in.

RQ1: What additional protective behaviors do students identify that they use to avert harm from alcohol consumption beyond the set of behaviors previously identified by other researchers?

Method

Student Sample

A web based survey (N = 1110) was conducted in the spring of 2006 at a large Midwestern university. The student services office drew a representative sample of the total student population for this study. The demographic characteristics of the sample closely matched the university�s student body: the proportion of males in the survey was 46% which is identical to the population; the distribution by year in school was also quite similar (22% Freshmen in the sample compared to 26% in the population, 23% Sophomore in the sample and 22% in the population, 23% Junior in the sample and 25% in the population, and 32% Senior in the sample versus 27% in the population, and the average age was 20.6 years in the sample vs. 20.3 years among all students.. The ethnic comparisons were also similar: Caucasian (84% sample vs. 82% population), African American (6% vs. 8%), Hispanic (5% vs. 3%), Native American (1% vs. 1%), and Asian Pacific Islander (8% vs. 6%). These percentages sum to over 100% as participants could chose more than one category if they identified with more than one ethnic group.
Measures

NCHA Protective Behaviors: The National Collegiate Health Assessment (NCHA) protective behaviors of alternating non-alcoholic drinks, setting limits, choosing not to drink, using designated driver, eating before or during drinking, having friends track drinks, personally tracking drinks, pacing drinking, and consuming alcohol look-alikes were presented to the respondents. Each behavior was assessed with a 5 point scale where 1=never and 5=always; those who did not drink alcohol in the past year were excluded from the analysis. The nine items were summed into an index of Protective Behavior Use with a mean of 26.18 (SD = 6.64).

Open-ended Protective Behaviors. Just after the NCHA list of protective behavior items, students were asked an open-ended question which stated, �What are some of the things that you do to protect yourself from harm when you party?� In response to this question, students were able to type multiple responses without limits of space. The total number of items mentioned by each respondent was summed to create an index of Open-Ended Protective Behavior Use ranging from 0 to 6 with a mean of 1.01 (SD = .99).

Open- Ended Coding Procedures
Previously established protective behaviors categories from Haines et al. (2006) were used as a basis for the coding scheme to analyze the open ended responses. Additional categories were also created if they appeared frequently in the participant�s responses. A complete listing of 43 different behaviors mentioned by respondents can be found Table 2: Coding Scheme and Frequencies Table 2b: Coding Scheme and Frequencies in order of their frequency along with notations of whether they are similar to categories found in the other three coding schemes noted above.

Four research assistants were trained as coders to analyze participant responses. After all coders completed the first 10% of responses, Guetzgow�s U and Cohen�s K were calculated to assess reliability. Guetzgow�s U was .008 indicating high agreement in unitizing responses. Cohen�s Kappa was .81 indicating that the research assistants coded the open ended responses reliably. The coders then resolved any differences that they had when coding. After determining that there was acceptable unitizing and coding reliability, the coders continued to code the rest of the open ended responses individually.
The 43 behaviors generated via this technique were summed into the index of Open-End Protective Behavior Use described above.

Results

Research Question 1

The research question asked what additional protective behaviors students engage in to avoid harmful consequences from alcohol consumption. Based on responses to the open-ended measure, approximately 65% of all study participants reported that they used one or more protective behaviors above and beyond the standard set measured in the NCHA instrument. Among this subset of students who reported a protective behavior, 1.53 was the mean number of different protective behaviors they listed.

Table 2 displays all of the frequencies for each of the 43 protective behaviors that were identified. The most common responses were staying with the same group of friends all night (30%), assuming personal responsibility to drink moderately, keep control and be careful (10%), and watching your drinks (7%)

Seven of the items that were mentioned by at least 2% of student drinkers involve interpersonal communication or coordination: Stay with the same group of friends, party only in a safe environment, arrange for an escort when traveling, drink with people that you know, don�t take alcohol from unfamiliar person, carry a cell phone, and watch out for companions. Two other communication-related items appeared on the NCHA list: Use a designated driver, and have a friend let you know when you have had enough. A majority of respondents reported that they employ these two listed behaviors always, usually or sometimes; 85% cited the designated driver item, and 51% cited the friend item.

The responses generated by the open-end items are quite distinct from the original nine behaviors; the nine-item index is only slightly related to the supplemental behaviors in the open-end index (r= .06, ns). Table 3: Percentages, Means and Standard Deviations for Key Variables provides the percentages, means, and standard deviations for the key variables in the study

Discussion

These findings revealed that the two most commonly cited protective behaviors in open-end responses were staying with the same group of friends and assuming personal responsibility; neither item was included in previous research on protective behaviors. The third most common item, watching drinks, captures the spirit of the Benton et al. (2004) �make one�s own drinks� and the Martens et al. (2005) �know where your drink has been at all times.�

This study also suggested that it is possible there are several higher order categories of protective behaviors as well. Please see Table 4: Nine Categories of Open-Ended Protective Behaviors Table 4b: Nine Categories of Open-ended Protective Behaviors for one preliminary scheme for categorizing protective behaviors. The three most common categories were drinking in a controlled environment (38%), safe traveling after drinking (15%), and minimizing intoxication (12%). The most popular higher order category, drinking in a controlled environment, included the following protective behavior items: party only in a familiar, safe, comfortable environment; avoid excessive drinking situations and companions; stay with the same group of friends all night; be aware of your surroundings; avoid partying or bars; and don't party with strangers or by yourself, only drink with people that you know and trust. None of these protective behaviors were used by NCHA or by researchers mentioned in Table 1. Therefore, this category encompasses a distinctive new subset of protective behaviors; indeed, the very low correlation of these new behaviors with the original list of behaviors indicates that these results tap into an array of protective practices that are quite different in nature Future research needs to measure the new protective behaviors uncovered here in such a way that factor analyses can be performed and a higher order scheme such as the one provided here can be confirmed.

What emerges from these findings is that there are some commonly used protective behaviors that previous studies did not examine. This study was unique because it asked students to name the protective behaviors that they use without supplying them with a list of behaviors. The expanded array of new protective behaviors provides a better understanding of the breadth of ways that students protect themselves in drinking contexts.

These findings have important implications for future social norms campaigns. Often normative campaigns solely focus on informing students about typical alcohol consumption by their peers. This study points to the idea that protective behaviors can play an important role in safeguarding students from harm due to alcohol consumption. Normative campaigns may seek to include messages about protective behaviors on their campus. Normative campaigns usually focus on positive messages and this is another way to encourage positive behavior among students. Further, normative campaigns are often unable to target high-risk drinking groups because behaviors that are normative in that group may involve high levels of alcohol consumption. Focusing on protective behaviors may be an alternate way to reach high-risk populations.

Unlike previous studies, this investigation allows participants to name the protective behaviors that they use. Other researchers have given participants a list of potential protective behaviors that they might use. Costa, Jessor and Turbin (1999) examined adolescent problem drinking with seventh, eighth and ninth grade students and identified other types of protective behaviors which are associated with problem drinking. Costa et al.�s (1999) measures of protective behaviors included positive orientation to school, attitude intolerance of deviance, positive relations with adults, perceived regulatory controls, friends as models for conventional behavior, pro-social activities, and positive orientation to health. These measures appear to be more internal characteristics which are already present as opposed to active measures taken to protect against alcohol. They do, however, represent interesting constructs that may further elucidate which protective measures students use and for what reasons. In other words, some of the protective behaviors measured by Costa et al. (1999) may still be influential for college students and may be related to specific protective behaviors employed in the present study.

These findings suggest several key implications for improving campaigns to prevent risky drinking problems. The social norms strategy requires promotion of behaviors that are normative (i.e., performed by a substantial majority of a particular population). Of the original nine NCHA protective behaviors, two to four of the practices garner a high enough percentage to be potentially normative. If the criterion is the percentage who report that they perform the protective behavior �always� or �usually,� then eating before/during drinking (82%) and using the designated driver (72%) meet the test. If the minimum frequency is expanded to also include �sometimes� performing the behavior, then there are four protective behaviors that meet the normative criterion: eating (97%) and designated driver (85%), plus keeping track of drinks (72%) and choosing not to drink (72%).

Among the many new behaviors generated via open-end questioning, one appears to be a promising candidate for a norms message: staying with the same group of friends all night. Fully 30% mention this one without a cue; if this practice is specified on a checklist, it is likely to meet the normative threshold. This percentage figure can be determined in a subsequent survey of students which includes the �same group of friends� item on a listing of protective behaviors.

This study also advances a related normative strategy involving multiple behaviors packaged in the same message. In addition to creating messages built around a single behavior, designers can devise combinations of protective behaviors where the percentage figure cited in a message is based on prevalence of performing at least one of multiple behaviors (e.g., �82% of students protect themselves by practicing one or more of the following behaviors when partying�). In the past, this approach has typically featured disparate combinations (e.g., using designated driver, pacing, and drinking look-alikes). Indeed, the original NCHA list yields few combinations that appear to coherently hang together. By expanding the array of behaviors and organizing the individual items into categories, this study provides a basis for creating stronger normative messages. Moreover, it is likely that the audience can better process and retain a set of multiple behaviors that are conceptually similar rather than a collection of unrelated behaviors.

Finally, this study demonstrates the importance of interpersonal communication in reducing drinking problems. Beyond self-protective actions, the research investigation identifies a number of protective behaviors involving interpersonal processes that may be fairly difficult to perform effectively. Campaigns can augment persuasive appeals with educational material that teaches audiences how to solicit, arrange, provide, and accept various forms of interpersonal protections in social settings; this may be supplemented with message content seeking to bolster self-efficacy in performing the communication behaviors.

Limitations and Future Directions

As mentioned briefly above, measurement of the new protective behaviors uncovered here via interval level analyses would allow researchers to perform a factor analysis to determine psychometrically if higher order categories do exist for the protective behaviors that students reported here.
Despite the limitations, this study is groundbreaking and has important implications for future campaigns. This study was based on a large and representative sample at a typical state university campus. This sampling feature serves to increase the generaliziability of the findings to other large campuses. Future studies should look to replicate the findings of this study in order to understand the drinking behaviors of undergraduates at their university and to design messages to protect the health and safety of those undergraduates.

References

American College Health Association (2003). National College Health Assessment.

Retrieved July 6, 2007 from http://www.acha-ncha.org/docs/sample_ncha.pdf.

Benton, S.L., Schmidt, J.L., Newton, F.B., Shin, D., Benton, S.A., & Newton, D.W.

(2004). College student protective strategies and drinking consequences. Journal of Studies on Alcohol, 65, 115-121.

Centers for Disease Control and Prevention (2006). Quick stats binge drinking. Retrieved on February 28, 2007 from http://www.cdc.gov/alcohol/quickstats/binge_drinking.htm

Costa, F.M., Jessor, R. & Turbin, M.S. (1999). Transition into adolescent problem drinking: The role of psychosocial risk and protective factors. Journal of Studies on Alcohol, 60, 480-490.

Haines, M.P., Barker, G., & Rice, R.M. (2006). The personal protective behaviors of college student drinkers: Evidence of indigenous protective norms. Journal of American College Health, 55, 69-75.

Haines, M. & Spear, S.F. (1996). Changing the perception of the norm: A strategy to decrease binge drinking among college students. Journal of American College Health, 45, 134-140.

Hanson, J.A. & Benedict, J.A. (2002). Use of the Health Belief Model to examine older adults' food-handling behaviors. Journal of Nutrition, Education and Behavior, 34, S25-S30.

Janz, N., Champion, V., & Strecher, V. (2002). The health behavior model. In K. Glanz,

B. K. Rimer, & F. M. Lewis (Eds.), Health behavior and health education: Theory, research, and practice (pp. 99�120). San Francisco: Jossey-Bass.

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

Lewis, T.F. & Thombs, D.L. (2005). Perceived risks and normative beliefs a explanatory models for college student alcohol involvement: An assessment of a campus with conventional alcohol control policies and enforcement practices. National Association of Student Personnel Administrators (NAPSA) Journal, 42, 202-222.

Martens, M.P., Ferrier, A.G., Sheehy, M.J., Corbett, K., Anderson, D.A. & Simmons, A. (2005). Development of the protective behavioral strategies survey. Journal of Studies on Alcohol, 66, 698-705).

Martens, M.P., Taylor, K.K., Damann, K.M., Page, J.C., Mowry, E.S. & Cimini, M.D. (2004). Protective behavioral strategies when drinking alcohol and their relationship to negative alcohol-related consequences in college students. Psychology of Addictive Behaviors, 18, 390 � 393.

Neighbors, C., Oster-Aaland, L., Berstrom, R.L. & Lewis, M.A. (2006). Event and context-specific normative misperception and high-risk drinking: 21^st birthday celebrations and football tailgating. Journal of Studies on Alcohol, 67, 282 � 289.

Nexoe, J., Kragstrup, J. & Sogaard, J. (1999). Decision on influenza vaccination among the elderly. Scandinavian Journal of Primary Health Care, 17, 105 � 110.

Office of Juvenile Justice and Delinquency Prevention. Drinking in America: Myths,

Realities, and Prevention Policy (PDF�103K). Pacific Institute for Research and Evaluation in support of the OJJDP Enforcing the Underage Drinking Laws Program. U. S. Department of Justice. November 2001.

O'Malley, P. M., & Johnston, L. D. (2002). Epidemiology of alcohol and other drug use among American college students. Journal of Studies on Alcohol, 63(Suppl. 14), 23-39)

Perkins, H. W. & Berkowitz, A. D. (1986). Perceiving the community norms of alcohol use among students: Some research implications for campus alcohol education programming. International Journal of the Addictions, 21, 961-976

Perkins, H.W. and Craig, D.W. (2006). A successful social norms campaign to reduce alcohol misuse among college student-athletes. Journal of Studies on Alcohol

Pielak, K.L. & Hilton, A. (2003). University students immunized and not immunized for measles: A comparison of beliefs, attitudes, and perceived barriers and benefits. Canadian Journal of Public Health, 94, 193-196.

Rosenstock, I.M., Strecher, V.J. & Becker, M.H. (1988). Social learning theory and the health belief model. Health Education & Behavior, 15, 175 � 183.

Rumpf, H., Hapke, U. & John, U. (1998) Previous help seeking and motivation to change< drinking behavior in alcohol-dependent general hospital patients. General Hospital Psychiatry, 20, 115 � 119.

Serdula, M.K., Brewer, R.D., Gillespie, C., Denny, C.H. & Mokdad, A. (2004). Trends in alcohol use and binge drinking, 1985-1999. Results of a multi-state survey. American Journal of Preventive Medicine, 26, 294-298.

Skinner, C.S., Strecher, V.J. & Hospers, H. (1994). Physicians' recommendations for mammography: do tailored messages make a difference? American Journal ofPublic Health, 84, 43 � 49.

Wecshler, H., Lee, J.E., Kuo, M., Seibring, M., Nelson, T.F., & Lee. H. (2002). Trends inJournal of American College Health, 50, 203-217.

Witte, K. (1992). Putting the fear back into fear appeals: The extended parallel processCommunication Monographs, 59, 329 � 349.

Witte, K. (1994). Fear control and danger control: A test of the extended parallel process model (EPPM). Communication Monographs, 61, 113�34.

Witte, K., Meyer, G., & Martell, D. (2001). Effective health risk messages: A \theoretically-based, step-by-step, how-to guide on developing persuasive communications that work. Newbury Park, CA: Sage.

Acknowledgments
This research was supported by grants from the AB Foundation and the United States Department of Education to Dennis Martell and Sandi W. Smith as the Principal Investigators. The authors would like to thank Larry Hembroff and Karen Clark for their assistance with the survey process, Carolyn LaPlante and Alex Mayer for their assistance with coding, and Rebecca Allen and Andrew Poole for their hard work on the campaign.