- Email: [email protected]
Alcohol self-control behaviors of adolescents
Addictive Behaviors 32 (2007) 590 – 597
Alcohol self-control behaviors of adolescents Tavis Glassman ⁎, Chudley (Chad) Werch, Edessa Jobli Addictive and Health Behaviors Research Institute, Department of Health Education and Behavior, University of Florida, 6852 Belfort Oaks Place, Jacksonville, FL 32216, USA
Abstract Purpose: The aims of the present study were to: (1) factor analyze a 13-item adolescent alcohol self-control behavior scale, (2) examine associations between frequency of self-control behavior use and alcohol consumption, and (3) to determine which self-control behaviors best predict alcohol use and consequences. Methods: A confidential standardized survey was used to collect data on participant's 30-day frequency, quantity, and heavy use of alcohol; alcohol-related consequences; and alcohol self-control behaviors. Results: A principal component factor analysis produced the following three components: Healthy Alternatives (α = .81), Self-regulation (α = .72), and Assertive Communication (α = .73). MANOVAs indicated strong associations between frequency of use of the three types of self-control behaviors and alcohol consumption (p values ≤ .001). Logistic regression analysis revealed that Self-regulation behaviors were the best predictor for all alcohol use measures and consequences (p values ≤ .001). Conclusion: Self-control behaviors differ in their ability to predict alcohol use and consequences. Self-regulation strategies emerged as the most consistent predictor of alcohol use patterns and consequences among adolescents, followed by Healthy Alternatives. © 2006 Elsevier Ltd. All rights reserved. Keywords: Self-control; Alcohol use
1. Introduction While considered a rite of passage by some, underage drinking poses a serious threat for a variety of reasons, including negative health outcomes, poor academic performance, and legal challenges (National Institute on Alcohol Abuse and Alcoholism Initiative on Underage Drinking, 2003; Williams & ⁎ Corresponding author. Tel.: +1 904 281 0726; fax: +1 904 296 1153. E-mail address: [email protected] (T. Glassman). 0306-4603/$ - see front matter © 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.addbeh.2006.06.003
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
591
Ricciardelli, 1999). Recent epidemiologic trends indicate adolescent alcohol use remains a challenge for the prevention field. According to the Youth Risk Behavior Surveillance System (YRBSS) data from 2003, nearly half of high school students (44.9%) had one or more drinks of alcohol in the last 30 days, and almost a third (28.3%) had five or more drinks on one or more occasions in the past 30 days (Center for Disease Control and Prevention, 2004). High school adolescents have been very resistant to intervention efforts (Foxcroft, Ireland, Lister-Sharp, Lowe, & Breen, 2003). Nevertheless, self-control skill instruction shows potential as one method of assisting in the prevention of alcohol and other drug misuse (Sussman, McCuller, & Dent, 2003). A study conducted by Carpenter, Lyons, and Miller (1985) indicated Behavioral Self Control Training (BSCT) resulted in significant decreases in quantity and frequency of drinking, and in peak blood alcohol levels, among Native American high school students, who identified as high risk for problem drinking. However, according to Carey and Maisto (1985), one shortcoming of the BSCT research is it fails to assess the use of self-control techniques that presumably account for the change in drinking behavior. In another BSCT-related study, the treatment and control group decreased their monthly heavy drinking days (Connors, Tarbox, & Failace, 1992). It appears both groups utilized self-control techniques; although, the control group used the techniques to a lesser extent than the treatment group. Moreover, a cross sectional study found that college students who used protective self-control behavioral strategies experienced fewer negative alcohol-related consequences than their peers who did not use such techniques or did so only on a limited basis (Martens et al., 2004). Thus, it appears that young people naturally use self-control strategies to reduce their risk when drinking alcohol. From a prevention perspective, it is important to know which of these self-control strategies elicit the greatest behavior impact on adolescent alcohol use. Although previous instruments have been developed to assess alcohol-related behavioral self-control (Collins & Lapp, 1992; Connors et al., 1992; Martens et al., 2004), we found none designed to measure alcohol self-control strategies used by high school adolescents. The present study used a 13-item measure of behavioral self-control strategies found in the Youth Alcohol and Health Survey (Werch, 2000). This scale measured self-control coping behaviors such as goal-setting, self-monitoring and use of alternative coping skills. These items were developed from two self-help program manuals (Miller & Munoz, 1982; Vogler & Bartz, 1982), which detailed commonly suggested alcohol-related behavioral self-control strategies (Werch, Carlson, Pappas, Edgemon, & DiClimente, 2002). The aims of the present study were to: (1) factor analyze a 13-item adolescent alcohol self-control behavior scale, (2) examine associations between frequency of self-control behavior use and alcohol consumption, and (3) to determine which self-control behaviors best predict alcohol use and consequences. Certain alcohol-related self-control strategies, and the frequency with which they are utilized may be important in preventing or reducing alcohol misuse and problems. To that end, this study may provide significant information for developing more efficacious adolescent prevention interventions in the future.
2. Method 2.1. Participants A total of 1284 students from a suburban high school in northeast Florida participated in the study. Recruitment occurred in fall 2002 (n = 604) and fall 2003 (n = 680). Participants reported the following
592
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
demographic data: Caucasian 49.6%, African American 21.2%, Multicultural 9.7%, Hispanic 7.9%, Asian 7.9%, Other 3%, and Native American .7%. Females represented 58% of the sample, with a mean age of 15.6 years (S.D. = 1.2 years). About 12% of subjects participated in the free or reduced cost lunch program. Two out of five students (40%) reported a family member with an alcohol or drug problem. The majority of fathers (73%) and mothers (57%) drank alcohol at least a few times a year. Over half (62%) of subjects reported receiving some form of alcohol or drug education during the past year. Finally, just under a third of the sample (31%) indicated they consumed alcohol within the last 30 days. 2.2. Measures Participating adolescents completed the Youth Alcohol and Health Survey (Werch, 2000), which took approximately 25 min to complete. The survey included a 13-item scale measuring self-control behaviors. These items originated from Miller and Munoz's (1982) seminal work on alcohol-related behavioral selfcontrol strategies. Werch and Gorman (1986) used this research to develop an extensive self-control inventory consisting of 50 plus self-control items which they administered to college students. Based on a factor analysis and theory, 13 items were selected to create a self-control scale. This scale is currently incorporated into the Youth Alcohol and Health Survey which includes items on alcohol use patterns, alcohol-related consequences, and other related substance issues. A number of prevention studies have used the survey, or some variation of it, to measure self-control and substance use behaviors (Werch et al., 1996, 2002; Werch, Carlson, Pappas, & DiClimente, 1996; Werch, Carlson, Pappas, Edgemon, & DiClimente, 2000). The self-control stem-item asked: “Have you used any of the following to help you stay away from using alcohol during the last year?” Sample-related branch items included “told others I was not going to drink”, “stayed away from or left places where drinking takes place”, and “used non-alcohol, healthy ways to deal with stress or nerves.” The dichotomous responses were “Yes” and “No.” This scale yielded an α coefficient of .88. Four measures of alcohol behaviors were also collected, including 30-day frequency, quantity, heavy use, and alcohol-related problems. The measure for 30-day frequency of alcohol use asked: “During the past 30 days, on how many days did you have at least one drink of alcohol?” Seven response categories ranged from “0 days” to “all 30 days.” The measure of alcohol quantity stated: “During the past 30-days, how much did you usually drink at one time?” Six response categories ranged from “I did not drink” to “5 or more drinks.” The item on heavy use read: “During the past 30-days, how many times have you had five or more drinks in a row?” Five responses ranged from “none” to “10 or more times.” Another item measured alcohol-related problems or consequences, with 13 response options. This item asked: “Have any of these things happened to you while you were drinking or drunk?” Related sample questions included “drove a car”, “an accident or injury”, and “trouble with police.” Responses included “yes”, “no”, and “I don't drink.” This scale had an α coefficient of .90. 2.3. Design and procedures The results presented are from the baseline surveys of two randomized controlled trial administered fall 2002 and 2003. A University Institutional Review Board approved the research protocols prior to implementing the study. All subjects submitted a signed parental consent form and a student assent form prior to participating in the study. Students received a nominal monetary incentive for their participation.
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
593
At the participating high school, trained research staff used standardized protocols to collect data, and ensure continuity in the research design. Participants received information concerning issues of confidentiality including use of code numbers on surveys, and assurance that no individual student data would be shared with anyone, including teachers, parents, or peers. To further protect confidentiality, students personally placed their surveys in folders immediately after completion. 2.4. Data analysis Descriptive statistics, including frequencies, percentages, means and standard deviations were conducted to describe the sample. A principal component factor analysis, utilizing a varimax rotation, was administered to determine the extent to which self-control items could be grouped. Individual self-control items falling within each resulting factor component were added to create self-control behavior categories. Each category was then stratified into three levels, representing increasingly greater frequency of selfcontrol utilization (i.e., none to low, moderate, and high). Multivariate analysis of variance (MANOVA) was conducted to examine mean differences of alcohol use across levels of self-control frequency. Lastly, a forward stepwise logistic regression analysis was generated to determine which self-control behavior categories best predicted the four alcohol use/problem measures. SPSS version 13.0 was used to conduct the aforementioned statistical analysis.
3. Results Related to the first study aim, we conducted a factor analysis on the 13-item alcohol self-control behavior scale. A principal component analysis and a varimax rotation produced a three component solution, which included the evaluation of the eigenvalue, variance and scree plot statistical values. After rotation, the three component solutions accounted for 55.3% of the total variance in the variables. Variables with a loading of < .500 were removed from the analysis. Consequently, “used healthy activities” (.468) and “thought about problems drinking can cause” (.429) were not included in the analysis. All correlation values among the three factors fell below r < .57 indicating high uniqueness from one another. Note: an additional factor analysis was conducted on only those students who indicated that they drank within the last 30 days. Results were almost identical to the original factor analysis reported in Table 1 which includes both drinkers and non-drinkers. Table 1 shows the results of the factor analysis, including the loadings and Cronbach's α scores. The first of the three component factors, labeled Healthy Alternatives, was comprised of four behaviors. Each of these items represents a behavioral substitute for alcohol use. Five items loaded highly to form the second factor, labeled Self-regulation. The two top loading items represent stimulus control strategies and the remaining three items characterize operant conditioning strategies. Lastly factor three, labeled Assertive Communication, includes two behaviors concerning strategies to communicate a desire not to drink. Table 2 shows the estimated marginal means of alcohol use measures by frequency of self-control strategies. The categories were divided into the three groups based on the range of the scores from the respective three factors. Each factor included a different number of self-control behaviors; thus, in order to make like comparisons the categories had to be collapsed. The labels of the categories, none to low, moderate, and high were logically labeled based on the distribution of scores. Greater use of Healthy Alternatives and Self-regulation strategies were significantly associated with lower mean alcohol
594
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
Table 1 Factor analysis of 13-item self-control strategies Component
Loadings
1) Healthy Alternatives: (Cronbach's α = .81) Used non-alcohol, healthy ways to feel at ease with people Used non-alcohol, healthy ways to deal with stress or nerves Used non-alcohol, healthy ways to feel good or high Choose non-alcoholic drink
.83 .82 .76 .54
2) Self-regulation: (Cronbach's α = .72) Stayed away from people who drink Avoided places where drinking takes place Rewarded self for not drinking Looked for more info about alcohol and health Punished self for drinking alcohol
.78 .74 .57 .51 .50
3) Assertive Communication: (Cronbach's α = .73) Said NO to an offer to drink alcohol Told others I was not going to drink
.83 .83
frequency, quantity, heavy use, and alcohol-related consequences (p values = .000). These associations where generally found at each increasing level of self-control frequency. A similar pattern emerges with the use of Assertive Communication strategies; however, these associations were limited to the three Table 2 Estimated marginal means of alcohol use measures by frequency of self-control strategies Self-control frequency None/Low
Moderate
Self-control categories/alcohol use measures
M
Healthy Alternatives 30-day frequency 30-day quantity 30-day heavy use Consequences Self-regulation 30-day frequency 30-day quantity 30-day heavy use Consequences Assertive Communication 30-day frequency 30-day quantity 30-day heavy use Consequences
F = 6.38; df = 8, 2496; p = .000 0.74 0.05 0.60 1.17 0.07 0.81 0.37 0.03 0.23 2.17 0.13 1.61 F = 10.40; df = 8, 2486; p = .000 0.84 0.04 0.46 1.28 0.06 0.65 0.39 0.03 0.18 2.02 0.11 1.66 F = 3.40; df = 8, 2524; p = .001 0.71 0.06 0.74 1.04 0.09 1.02 0.34 0.04 0.36 1.72 0.15 1.74
High scores = High risk. a Moderate is significantly different from None/Low. b High is significantly different from None/Low. c High is significantly different from Moderate.
S.E.
M
High S.E.
M
S.E.
p value
0.04 0.06 0.03 0.11
0.30 0.47 0.12 1.16
0.05 0.08 0.03 0.87
.000 a,b,c .000 a,b,c .000 a,b,c .000 a,b,c
0.05 0.07 0.03 0.12
0.29 0.42 0.11 1.17
0.05 0.08 0.03 0.14
.000 a,b,c .000 a,b,c .000 a,b .000 a,b,c
0.07 0.10 0.04 0.17
0.46 0.69 0.17 1.62
0.03 0.05 0.02 0.09
.000 b,c .001 b,c .000 b,c .77
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
595
Table 3 Logistic regression predicting alcohol use measures using self-control strategies a Alcohol use measures/ self-control strategies Frequency of alcohol use Healthy Alternatives Self-regulation Assertive Communication Quantity of alcohol use Healthy Alternatives Self-regulation Assertive Communication Heavy use Healthy Alternatives Self-regulation Assertive Communication Consequences/Problems Healthy Alternatives Self-regulation Assertive Communication
Odds ratio
Confidence interval Lower
Upper
Percent of cases p-value
Correctly classified 70
1.98
1.68
2.34
.000 70
2.01
1.71
2.37
.000
1.35 1.87
1.04 1.43
1.76 2.43
.02 .000
1.40 1.60 0.80
1.16 1.35 0.68
1.69 1.91 0.93
.000 .000 .004
86
61
Predictors used in the analyses: Healthy Alternatives: Use healthy ways to 1) feel at ease with people, 2) deal with stress, 3) feel good and high, 4) non-alcoholic drink, and 5) healthy activities. Self-regulation: 1) stay away from people who drink, 2) avoid places where drinking takes place, 3) reward self for not drinking, 4) looked for more info about alcohol and health, 5) punished self for drinking, and 6) thought about problems drinking can cause. Assertive Communication: 1) Said NO to an offer to drink, and 2) told others I was not going to drink. a High score = High risk.
alcohol use measures (not consequences), and only occurred at the highest level of use of Assertive Communication (p values ≤ .001). A forward stepwise logistic regression analysis was run to determine if the three self-control factors, Healthy Alternatives, Self-regulation and Assertive Communication could predict 30-day frequency, 30day quantity, 30-day heavy use, and consequences. The four dependent measures were dichotomized into yes/no responses. Table 3, for illustrative purposes, combines the statistically significant results of the four separate logistic regression analyses. Overall, Self-regulation strategies were found to be the best predictor of all four measures of alcohol use and consequences. Specifically, Self-regulation predicted 30day frequency (OR = 1.98, p = .000), 30-day quantity (OR = 2.01, p = .000), heavy use (OR = 1.87, p = .000), and alcohol consequences (OR = 1.60, p = .000). Healthy Alternatives strategies predicted heavy use (OR = 1.35, p = .02) and consequences (OR = 1.40, p = .000). The analysis resulted in 70%, 70%, 86%, and 61% cases correctly classified for frequency, quantity, heavy use, and consequences, respectively.
4. Discussion The aims of the present study were to: (1) factor analyze a 13-item adolescent alcohol self-control behavior scale, (2) examine associations between frequency of self-control behavior use and alcohol
596
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
consumption, and (3) to determine which self-control behaviors best predict alcohol use and consequences. Findings indicated that alcohol self-control behaviors used by adolescents factor into Healthy Alternatives, Self-regulation, and Assertive Communication strategies. All three types of selfcontrol behaviors are associated with alcohol consumption, with Healthy Alternatives and Selfregulation also related to alcohol problems. Lastly, self-control strategies differ in their ability to predict alcohol use and consequences, with Self-regulation behaviors emerging as the most consistent predictor of alcohol use patterns and negative consequences, followed by Healthy Alternative strategies. Based on the findings of this and prior studies, it appears that young people are naturally using selfcontrol strategies to restrain their drinking (Connors et al., 1992; Martens et al., 2004). Research indicates that roughly one quarter of adolescents who drink attempt to cut down or stop drinking each year (Wagner, Brown, Monti, Myers, & Waldron, 1999). Metrik and colleagues (2003) found that adolescents prefer behavioral self-management strategies as methods for reducing alcohol consumption. Adolescents may prefer to try to control their drinking using self-control behaviors than attend formal education, counseling, or endure restrictive policy and enforcement efforts (Greenfield, Guydish, & Temple, 1989; Metrik et al., 2003). Results from this and other related studies suggest that practitioners and researchers may want to incorporate specific types of behavioral self-control strategies into their prevention interventions (Werch & Gorman, 1986). Training in the use of stimulus–control self-regulation strategies, including staying away from people who drink and avoiding places where drinking occurs, may be particularly useful. Operant conditioning tactics such as rewarding and punishing one's self for reaching, or failing to reach, self-control goals also holds potential in self-regulating adolescent alcohol use. Further, encouraging adolescents to engage in Healthy Alternatives to cope with social, stress, or pleasure needs may reduce heavy alcohol use and related consequences among this age group. Assertive Communication strategies, such as telling others that one is not going to drink, or refusing offers to drink alcohol, appear to be the least useful type of self-control behaviors. Perhaps younger adolescents, who are less advanced in their readiness to drink alcohol, and less susceptible to alcohol use offers, may benefit more from these strategies than their older counterparts. Additional, research needs to be conducted examining which specific self-control strategies are most useful for various developmental levels of adolescence. The findings presented in this study should be interpreted carefully. First, only one high school participated in this study; consequently, generalizations concerning self-control behaviors may be somewhat limited. Second, this study was limited to a 13-item self-control measure. Additional selfcontrol strategies should be studied in the future for their potential to reduce alcohol use and problems, such as monitoring the number of drinks consumed, avoiding drinking games and high potency alcohol beverages. Third, a cause and effect relationship cannot be established from these results because of the cross sectional design used in this study. Nevertheless, this study suggests the use of certain self-control behaviors is related to and predictive of alcohol consumption patterns and problems among adolescents. Self-regulation and Healthy Alternatives strategies figure to be a particularly promising means of reducing alcohol misuse among adolescents. Based on the findings of this study, it appears that prevention efforts directed towards older adolescents should include selected self-control strategies. Future research efforts need to employ controlled trials to test the efficacy of self-control strategies alone and in combination with other prevention components for high school as well as college age adolescents.
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
597
Acknowledgements This manuscript was supported in part by grants from the National Institute on Alcohol Abuse and Alcoholism (Grant #AA9283), and the National Institute on Drug Abuse (Grant #DA018872 and #DA019172).
References Carey, K. B., & Maisto, S. A. (1985). A review of the use of self-control techniques in the treatment of alcohol abuse. Cognitive Therapy and Research, 9, 235−251. Carpenter, R. A., Lyons, C. A., & Miller, W. R. (1985). Peer-managed self-control program for prevention of alcohol abuse in American Indian high school students: A pilot evaluation study. International Journal of the Addictions, 2, 299−310. Center for Disease Control and Prevention. (2004). Surveillance summaries, May 21, 2004. MMWR, 53(SS-2). Collins, R. L., & Lapp, W. M. (1992). The temptation and restraint inventory for measuring drinking restraint. British Journal of Addiction, 87, 625−633. Connors, G. J., Tarbox, A. R., & Failace, L. A. (1992). Achieving and maintaining gains among problem drinkers: Process and outcome results. Behavior Therapy, 23, 449−474. Foxcroft, D. R., Ireland, D., Lister-Sharp, D. J., Lowe, G., & Breen, R. (2003). Longer-term primary prevention for alcohol misuse in young people: A systematic review. Addiction, 98, 397−411. Greenfield, T. K., Guydish, J., & Temple, M. T. (1989). Reasons students give for limiting drinking: A factor analysis with implications for research. Journal of Studies on Alcohol, 50, 108−115. Martens, M. P., Tayolor, 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. Metrik, J., Frissell, K. C., McCarthy, D. M., D'Amico, E. J., & Brown, S. A. (2003). Strategies for reduction and cessation of alcohol use: Adolescent preferences. Alcoholism, Clinical and Experimental Research, 27, 74−80. Miller, W. R., & Munoz, R. F. (1982). How to control your drinking. Albuquerque, NM: Univer. of New Mexico Press. National Institute on Alcohol Abuse and Alcoholism Initiative on Underage Drinking. (2003). Retrieved April 11, 2005, from http://www.niaaa.nih.gov/about/underage.htm#statistics Sussman, S., McCuller, W. J., & Dent, C. W. (2003). The associations of social self-control, personality disorders, and demographics with drug use among high-risk youth. Addictive Behaviors, 28, 1159−1166. Vogler, R. E., & Bartz, W. R. (1982). The better way to drink. New York: Simon and Schuster. Wagner, E. S., Brown, S. A., Monti, P., Myers, M. G., & Waldron, H. B. (1999). Innovations in adolescent substance abuse and prevention. Alcoholism, Clinical and Experimental Research, 23, 236−249. Werch, C. E. (2000). The youth alcohol and health survey. Jacksonsville, Fla: University of North Florida Center for Drug Prevention Research. Werch, C. E., Anzalone, D., Brokiewicz, L., Felker, J., Carlson, J., & Castellon-Vogel, E. (1996). An intervention for preventing alcohol use among inner-city middle school students. Archives of Family Medicine, 5, 146−152. Werch, C. E., Carlson, J., Pappas, D., & DiClimente, C. C. (1996). An intervention for prevention alcohol use among inner-city middle school students. Journal of School Health, 66, 335−338. Werch, C. E., Carlson, J., Pappas, D., Edgemon, P., & DiClimente, C. C. (2000). Effects of a brief alcohol preventive intervention for youth attending school sports physical examinations. Substance use and Misuse, 35, 421−432. Werch, C. E., Carlson, J., Pappas, D., Edgemon, P., & DiClimente, C. C. (2002). A brief alcohol preventive intervention for student athletes. The Prevention Researcher, 9, 4−5. Werch, C. E., & Gorman, D. R. (1986). Factor analysis of internal and external self-control practices for alcohol consumption. Psychological Reports, 59, 1207−1213. Williams, R. J., & Ricciardelli, L. A. (1999). Restrained drinking and cognitive control among adolescents. Adolescence, 34, 557−565.
Alcohol self-control behaviors of adolescents Tavis Glassman ⁎, Chudley (Chad) Werch, Edessa Jobli Addictive and Health Behaviors Research Institute, Department of Health Education and Behavior, University of Florida, 6852 Belfort Oaks Place, Jacksonville, FL 32216, USA
Abstract Purpose: The aims of the present study were to: (1) factor analyze a 13-item adolescent alcohol self-control behavior scale, (2) examine associations between frequency of self-control behavior use and alcohol consumption, and (3) to determine which self-control behaviors best predict alcohol use and consequences. Methods: A confidential standardized survey was used to collect data on participant's 30-day frequency, quantity, and heavy use of alcohol; alcohol-related consequences; and alcohol self-control behaviors. Results: A principal component factor analysis produced the following three components: Healthy Alternatives (α = .81), Self-regulation (α = .72), and Assertive Communication (α = .73). MANOVAs indicated strong associations between frequency of use of the three types of self-control behaviors and alcohol consumption (p values ≤ .001). Logistic regression analysis revealed that Self-regulation behaviors were the best predictor for all alcohol use measures and consequences (p values ≤ .001). Conclusion: Self-control behaviors differ in their ability to predict alcohol use and consequences. Self-regulation strategies emerged as the most consistent predictor of alcohol use patterns and consequences among adolescents, followed by Healthy Alternatives. © 2006 Elsevier Ltd. All rights reserved. Keywords: Self-control; Alcohol use
1. Introduction While considered a rite of passage by some, underage drinking poses a serious threat for a variety of reasons, including negative health outcomes, poor academic performance, and legal challenges (National Institute on Alcohol Abuse and Alcoholism Initiative on Underage Drinking, 2003; Williams & ⁎ Corresponding author. Tel.: +1 904 281 0726; fax: +1 904 296 1153. E-mail address: [email protected] (T. Glassman). 0306-4603/$ - see front matter © 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.addbeh.2006.06.003
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
591
Ricciardelli, 1999). Recent epidemiologic trends indicate adolescent alcohol use remains a challenge for the prevention field. According to the Youth Risk Behavior Surveillance System (YRBSS) data from 2003, nearly half of high school students (44.9%) had one or more drinks of alcohol in the last 30 days, and almost a third (28.3%) had five or more drinks on one or more occasions in the past 30 days (Center for Disease Control and Prevention, 2004). High school adolescents have been very resistant to intervention efforts (Foxcroft, Ireland, Lister-Sharp, Lowe, & Breen, 2003). Nevertheless, self-control skill instruction shows potential as one method of assisting in the prevention of alcohol and other drug misuse (Sussman, McCuller, & Dent, 2003). A study conducted by Carpenter, Lyons, and Miller (1985) indicated Behavioral Self Control Training (BSCT) resulted in significant decreases in quantity and frequency of drinking, and in peak blood alcohol levels, among Native American high school students, who identified as high risk for problem drinking. However, according to Carey and Maisto (1985), one shortcoming of the BSCT research is it fails to assess the use of self-control techniques that presumably account for the change in drinking behavior. In another BSCT-related study, the treatment and control group decreased their monthly heavy drinking days (Connors, Tarbox, & Failace, 1992). It appears both groups utilized self-control techniques; although, the control group used the techniques to a lesser extent than the treatment group. Moreover, a cross sectional study found that college students who used protective self-control behavioral strategies experienced fewer negative alcohol-related consequences than their peers who did not use such techniques or did so only on a limited basis (Martens et al., 2004). Thus, it appears that young people naturally use self-control strategies to reduce their risk when drinking alcohol. From a prevention perspective, it is important to know which of these self-control strategies elicit the greatest behavior impact on adolescent alcohol use. Although previous instruments have been developed to assess alcohol-related behavioral self-control (Collins & Lapp, 1992; Connors et al., 1992; Martens et al., 2004), we found none designed to measure alcohol self-control strategies used by high school adolescents. The present study used a 13-item measure of behavioral self-control strategies found in the Youth Alcohol and Health Survey (Werch, 2000). This scale measured self-control coping behaviors such as goal-setting, self-monitoring and use of alternative coping skills. These items were developed from two self-help program manuals (Miller & Munoz, 1982; Vogler & Bartz, 1982), which detailed commonly suggested alcohol-related behavioral self-control strategies (Werch, Carlson, Pappas, Edgemon, & DiClimente, 2002). The aims of the present study were to: (1) factor analyze a 13-item adolescent alcohol self-control behavior scale, (2) examine associations between frequency of self-control behavior use and alcohol consumption, and (3) to determine which self-control behaviors best predict alcohol use and consequences. Certain alcohol-related self-control strategies, and the frequency with which they are utilized may be important in preventing or reducing alcohol misuse and problems. To that end, this study may provide significant information for developing more efficacious adolescent prevention interventions in the future.
2. Method 2.1. Participants A total of 1284 students from a suburban high school in northeast Florida participated in the study. Recruitment occurred in fall 2002 (n = 604) and fall 2003 (n = 680). Participants reported the following
592
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
demographic data: Caucasian 49.6%, African American 21.2%, Multicultural 9.7%, Hispanic 7.9%, Asian 7.9%, Other 3%, and Native American .7%. Females represented 58% of the sample, with a mean age of 15.6 years (S.D. = 1.2 years). About 12% of subjects participated in the free or reduced cost lunch program. Two out of five students (40%) reported a family member with an alcohol or drug problem. The majority of fathers (73%) and mothers (57%) drank alcohol at least a few times a year. Over half (62%) of subjects reported receiving some form of alcohol or drug education during the past year. Finally, just under a third of the sample (31%) indicated they consumed alcohol within the last 30 days. 2.2. Measures Participating adolescents completed the Youth Alcohol and Health Survey (Werch, 2000), which took approximately 25 min to complete. The survey included a 13-item scale measuring self-control behaviors. These items originated from Miller and Munoz's (1982) seminal work on alcohol-related behavioral selfcontrol strategies. Werch and Gorman (1986) used this research to develop an extensive self-control inventory consisting of 50 plus self-control items which they administered to college students. Based on a factor analysis and theory, 13 items were selected to create a self-control scale. This scale is currently incorporated into the Youth Alcohol and Health Survey which includes items on alcohol use patterns, alcohol-related consequences, and other related substance issues. A number of prevention studies have used the survey, or some variation of it, to measure self-control and substance use behaviors (Werch et al., 1996, 2002; Werch, Carlson, Pappas, & DiClimente, 1996; Werch, Carlson, Pappas, Edgemon, & DiClimente, 2000). The self-control stem-item asked: “Have you used any of the following to help you stay away from using alcohol during the last year?” Sample-related branch items included “told others I was not going to drink”, “stayed away from or left places where drinking takes place”, and “used non-alcohol, healthy ways to deal with stress or nerves.” The dichotomous responses were “Yes” and “No.” This scale yielded an α coefficient of .88. Four measures of alcohol behaviors were also collected, including 30-day frequency, quantity, heavy use, and alcohol-related problems. The measure for 30-day frequency of alcohol use asked: “During the past 30 days, on how many days did you have at least one drink of alcohol?” Seven response categories ranged from “0 days” to “all 30 days.” The measure of alcohol quantity stated: “During the past 30-days, how much did you usually drink at one time?” Six response categories ranged from “I did not drink” to “5 or more drinks.” The item on heavy use read: “During the past 30-days, how many times have you had five or more drinks in a row?” Five responses ranged from “none” to “10 or more times.” Another item measured alcohol-related problems or consequences, with 13 response options. This item asked: “Have any of these things happened to you while you were drinking or drunk?” Related sample questions included “drove a car”, “an accident or injury”, and “trouble with police.” Responses included “yes”, “no”, and “I don't drink.” This scale had an α coefficient of .90. 2.3. Design and procedures The results presented are from the baseline surveys of two randomized controlled trial administered fall 2002 and 2003. A University Institutional Review Board approved the research protocols prior to implementing the study. All subjects submitted a signed parental consent form and a student assent form prior to participating in the study. Students received a nominal monetary incentive for their participation.
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
593
At the participating high school, trained research staff used standardized protocols to collect data, and ensure continuity in the research design. Participants received information concerning issues of confidentiality including use of code numbers on surveys, and assurance that no individual student data would be shared with anyone, including teachers, parents, or peers. To further protect confidentiality, students personally placed their surveys in folders immediately after completion. 2.4. Data analysis Descriptive statistics, including frequencies, percentages, means and standard deviations were conducted to describe the sample. A principal component factor analysis, utilizing a varimax rotation, was administered to determine the extent to which self-control items could be grouped. Individual self-control items falling within each resulting factor component were added to create self-control behavior categories. Each category was then stratified into three levels, representing increasingly greater frequency of selfcontrol utilization (i.e., none to low, moderate, and high). Multivariate analysis of variance (MANOVA) was conducted to examine mean differences of alcohol use across levels of self-control frequency. Lastly, a forward stepwise logistic regression analysis was generated to determine which self-control behavior categories best predicted the four alcohol use/problem measures. SPSS version 13.0 was used to conduct the aforementioned statistical analysis.
3. Results Related to the first study aim, we conducted a factor analysis on the 13-item alcohol self-control behavior scale. A principal component analysis and a varimax rotation produced a three component solution, which included the evaluation of the eigenvalue, variance and scree plot statistical values. After rotation, the three component solutions accounted for 55.3% of the total variance in the variables. Variables with a loading of < .500 were removed from the analysis. Consequently, “used healthy activities” (.468) and “thought about problems drinking can cause” (.429) were not included in the analysis. All correlation values among the three factors fell below r < .57 indicating high uniqueness from one another. Note: an additional factor analysis was conducted on only those students who indicated that they drank within the last 30 days. Results were almost identical to the original factor analysis reported in Table 1 which includes both drinkers and non-drinkers. Table 1 shows the results of the factor analysis, including the loadings and Cronbach's α scores. The first of the three component factors, labeled Healthy Alternatives, was comprised of four behaviors. Each of these items represents a behavioral substitute for alcohol use. Five items loaded highly to form the second factor, labeled Self-regulation. The two top loading items represent stimulus control strategies and the remaining three items characterize operant conditioning strategies. Lastly factor three, labeled Assertive Communication, includes two behaviors concerning strategies to communicate a desire not to drink. Table 2 shows the estimated marginal means of alcohol use measures by frequency of self-control strategies. The categories were divided into the three groups based on the range of the scores from the respective three factors. Each factor included a different number of self-control behaviors; thus, in order to make like comparisons the categories had to be collapsed. The labels of the categories, none to low, moderate, and high were logically labeled based on the distribution of scores. Greater use of Healthy Alternatives and Self-regulation strategies were significantly associated with lower mean alcohol
594
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
Table 1 Factor analysis of 13-item self-control strategies Component
Loadings
1) Healthy Alternatives: (Cronbach's α = .81) Used non-alcohol, healthy ways to feel at ease with people Used non-alcohol, healthy ways to deal with stress or nerves Used non-alcohol, healthy ways to feel good or high Choose non-alcoholic drink
.83 .82 .76 .54
2) Self-regulation: (Cronbach's α = .72) Stayed away from people who drink Avoided places where drinking takes place Rewarded self for not drinking Looked for more info about alcohol and health Punished self for drinking alcohol
.78 .74 .57 .51 .50
3) Assertive Communication: (Cronbach's α = .73) Said NO to an offer to drink alcohol Told others I was not going to drink
.83 .83
frequency, quantity, heavy use, and alcohol-related consequences (p values = .000). These associations where generally found at each increasing level of self-control frequency. A similar pattern emerges with the use of Assertive Communication strategies; however, these associations were limited to the three Table 2 Estimated marginal means of alcohol use measures by frequency of self-control strategies Self-control frequency None/Low
Moderate
Self-control categories/alcohol use measures
M
Healthy Alternatives 30-day frequency 30-day quantity 30-day heavy use Consequences Self-regulation 30-day frequency 30-day quantity 30-day heavy use Consequences Assertive Communication 30-day frequency 30-day quantity 30-day heavy use Consequences
F = 6.38; df = 8, 2496; p = .000 0.74 0.05 0.60 1.17 0.07 0.81 0.37 0.03 0.23 2.17 0.13 1.61 F = 10.40; df = 8, 2486; p = .000 0.84 0.04 0.46 1.28 0.06 0.65 0.39 0.03 0.18 2.02 0.11 1.66 F = 3.40; df = 8, 2524; p = .001 0.71 0.06 0.74 1.04 0.09 1.02 0.34 0.04 0.36 1.72 0.15 1.74
High scores = High risk. a Moderate is significantly different from None/Low. b High is significantly different from None/Low. c High is significantly different from Moderate.
S.E.
M
High S.E.
M
S.E.
p value
0.04 0.06 0.03 0.11
0.30 0.47 0.12 1.16
0.05 0.08 0.03 0.87
.000 a,b,c .000 a,b,c .000 a,b,c .000 a,b,c
0.05 0.07 0.03 0.12
0.29 0.42 0.11 1.17
0.05 0.08 0.03 0.14
.000 a,b,c .000 a,b,c .000 a,b .000 a,b,c
0.07 0.10 0.04 0.17
0.46 0.69 0.17 1.62
0.03 0.05 0.02 0.09
.000 b,c .001 b,c .000 b,c .77
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
595
Table 3 Logistic regression predicting alcohol use measures using self-control strategies a Alcohol use measures/ self-control strategies Frequency of alcohol use Healthy Alternatives Self-regulation Assertive Communication Quantity of alcohol use Healthy Alternatives Self-regulation Assertive Communication Heavy use Healthy Alternatives Self-regulation Assertive Communication Consequences/Problems Healthy Alternatives Self-regulation Assertive Communication
Odds ratio
Confidence interval Lower
Upper
Percent of cases p-value
Correctly classified 70
1.98
1.68
2.34
.000 70
2.01
1.71
2.37
.000
1.35 1.87
1.04 1.43
1.76 2.43
.02 .000
1.40 1.60 0.80
1.16 1.35 0.68
1.69 1.91 0.93
.000 .000 .004
86
61
Predictors used in the analyses: Healthy Alternatives: Use healthy ways to 1) feel at ease with people, 2) deal with stress, 3) feel good and high, 4) non-alcoholic drink, and 5) healthy activities. Self-regulation: 1) stay away from people who drink, 2) avoid places where drinking takes place, 3) reward self for not drinking, 4) looked for more info about alcohol and health, 5) punished self for drinking, and 6) thought about problems drinking can cause. Assertive Communication: 1) Said NO to an offer to drink, and 2) told others I was not going to drink. a High score = High risk.
alcohol use measures (not consequences), and only occurred at the highest level of use of Assertive Communication (p values ≤ .001). A forward stepwise logistic regression analysis was run to determine if the three self-control factors, Healthy Alternatives, Self-regulation and Assertive Communication could predict 30-day frequency, 30day quantity, 30-day heavy use, and consequences. The four dependent measures were dichotomized into yes/no responses. Table 3, for illustrative purposes, combines the statistically significant results of the four separate logistic regression analyses. Overall, Self-regulation strategies were found to be the best predictor of all four measures of alcohol use and consequences. Specifically, Self-regulation predicted 30day frequency (OR = 1.98, p = .000), 30-day quantity (OR = 2.01, p = .000), heavy use (OR = 1.87, p = .000), and alcohol consequences (OR = 1.60, p = .000). Healthy Alternatives strategies predicted heavy use (OR = 1.35, p = .02) and consequences (OR = 1.40, p = .000). The analysis resulted in 70%, 70%, 86%, and 61% cases correctly classified for frequency, quantity, heavy use, and consequences, respectively.
4. Discussion The aims of the present study were to: (1) factor analyze a 13-item adolescent alcohol self-control behavior scale, (2) examine associations between frequency of self-control behavior use and alcohol
596
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
consumption, and (3) to determine which self-control behaviors best predict alcohol use and consequences. Findings indicated that alcohol self-control behaviors used by adolescents factor into Healthy Alternatives, Self-regulation, and Assertive Communication strategies. All three types of selfcontrol behaviors are associated with alcohol consumption, with Healthy Alternatives and Selfregulation also related to alcohol problems. Lastly, self-control strategies differ in their ability to predict alcohol use and consequences, with Self-regulation behaviors emerging as the most consistent predictor of alcohol use patterns and negative consequences, followed by Healthy Alternative strategies. Based on the findings of this and prior studies, it appears that young people are naturally using selfcontrol strategies to restrain their drinking (Connors et al., 1992; Martens et al., 2004). Research indicates that roughly one quarter of adolescents who drink attempt to cut down or stop drinking each year (Wagner, Brown, Monti, Myers, & Waldron, 1999). Metrik and colleagues (2003) found that adolescents prefer behavioral self-management strategies as methods for reducing alcohol consumption. Adolescents may prefer to try to control their drinking using self-control behaviors than attend formal education, counseling, or endure restrictive policy and enforcement efforts (Greenfield, Guydish, & Temple, 1989; Metrik et al., 2003). Results from this and other related studies suggest that practitioners and researchers may want to incorporate specific types of behavioral self-control strategies into their prevention interventions (Werch & Gorman, 1986). Training in the use of stimulus–control self-regulation strategies, including staying away from people who drink and avoiding places where drinking occurs, may be particularly useful. Operant conditioning tactics such as rewarding and punishing one's self for reaching, or failing to reach, self-control goals also holds potential in self-regulating adolescent alcohol use. Further, encouraging adolescents to engage in Healthy Alternatives to cope with social, stress, or pleasure needs may reduce heavy alcohol use and related consequences among this age group. Assertive Communication strategies, such as telling others that one is not going to drink, or refusing offers to drink alcohol, appear to be the least useful type of self-control behaviors. Perhaps younger adolescents, who are less advanced in their readiness to drink alcohol, and less susceptible to alcohol use offers, may benefit more from these strategies than their older counterparts. Additional, research needs to be conducted examining which specific self-control strategies are most useful for various developmental levels of adolescence. The findings presented in this study should be interpreted carefully. First, only one high school participated in this study; consequently, generalizations concerning self-control behaviors may be somewhat limited. Second, this study was limited to a 13-item self-control measure. Additional selfcontrol strategies should be studied in the future for their potential to reduce alcohol use and problems, such as monitoring the number of drinks consumed, avoiding drinking games and high potency alcohol beverages. Third, a cause and effect relationship cannot be established from these results because of the cross sectional design used in this study. Nevertheless, this study suggests the use of certain self-control behaviors is related to and predictive of alcohol consumption patterns and problems among adolescents. Self-regulation and Healthy Alternatives strategies figure to be a particularly promising means of reducing alcohol misuse among adolescents. Based on the findings of this study, it appears that prevention efforts directed towards older adolescents should include selected self-control strategies. Future research efforts need to employ controlled trials to test the efficacy of self-control strategies alone and in combination with other prevention components for high school as well as college age adolescents.
T. Glassman et al. / Addictive Behaviors 32 (2007) 590–597
597
Acknowledgements This manuscript was supported in part by grants from the National Institute on Alcohol Abuse and Alcoholism (Grant #AA9283), and the National Institute on Drug Abuse (Grant #DA018872 and #DA019172).
References Carey, K. B., & Maisto, S. A. (1985). A review of the use of self-control techniques in the treatment of alcohol abuse. Cognitive Therapy and Research, 9, 235−251. Carpenter, R. A., Lyons, C. A., & Miller, W. R. (1985). Peer-managed self-control program for prevention of alcohol abuse in American Indian high school students: A pilot evaluation study. International Journal of the Addictions, 2, 299−310. Center for Disease Control and Prevention. (2004). Surveillance summaries, May 21, 2004. MMWR, 53(SS-2). Collins, R. L., & Lapp, W. M. (1992). The temptation and restraint inventory for measuring drinking restraint. British Journal of Addiction, 87, 625−633. Connors, G. J., Tarbox, A. R., & Failace, L. A. (1992). Achieving and maintaining gains among problem drinkers: Process and outcome results. Behavior Therapy, 23, 449−474. Foxcroft, D. R., Ireland, D., Lister-Sharp, D. J., Lowe, G., & Breen, R. (2003). Longer-term primary prevention for alcohol misuse in young people: A systematic review. Addiction, 98, 397−411. Greenfield, T. K., Guydish, J., & Temple, M. T. (1989). Reasons students give for limiting drinking: A factor analysis with implications for research. Journal of Studies on Alcohol, 50, 108−115. Martens, M. P., Tayolor, 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. Metrik, J., Frissell, K. C., McCarthy, D. M., D'Amico, E. J., & Brown, S. A. (2003). Strategies for reduction and cessation of alcohol use: Adolescent preferences. Alcoholism, Clinical and Experimental Research, 27, 74−80. Miller, W. R., & Munoz, R. F. (1982). How to control your drinking. Albuquerque, NM: Univer. of New Mexico Press. National Institute on Alcohol Abuse and Alcoholism Initiative on Underage Drinking. (2003). Retrieved April 11, 2005, from http://www.niaaa.nih.gov/about/underage.htm#statistics Sussman, S., McCuller, W. J., & Dent, C. W. (2003). The associations of social self-control, personality disorders, and demographics with drug use among high-risk youth. Addictive Behaviors, 28, 1159−1166. Vogler, R. E., & Bartz, W. R. (1982). The better way to drink. New York: Simon and Schuster. Wagner, E. S., Brown, S. A., Monti, P., Myers, M. G., & Waldron, H. B. (1999). Innovations in adolescent substance abuse and prevention. Alcoholism, Clinical and Experimental Research, 23, 236−249. Werch, C. E. (2000). The youth alcohol and health survey. Jacksonsville, Fla: University of North Florida Center for Drug Prevention Research. Werch, C. E., Anzalone, D., Brokiewicz, L., Felker, J., Carlson, J., & Castellon-Vogel, E. (1996). An intervention for preventing alcohol use among inner-city middle school students. Archives of Family Medicine, 5, 146−152. Werch, C. E., Carlson, J., Pappas, D., & DiClimente, C. C. (1996). An intervention for prevention alcohol use among inner-city middle school students. Journal of School Health, 66, 335−338. Werch, C. E., Carlson, J., Pappas, D., Edgemon, P., & DiClimente, C. C. (2000). Effects of a brief alcohol preventive intervention for youth attending school sports physical examinations. Substance use and Misuse, 35, 421−432. Werch, C. E., Carlson, J., Pappas, D., Edgemon, P., & DiClimente, C. C. (2002). A brief alcohol preventive intervention for student athletes. The Prevention Researcher, 9, 4−5. Werch, C. E., & Gorman, D. R. (1986). Factor analysis of internal and external self-control practices for alcohol consumption. Psychological Reports, 59, 1207−1213. Williams, R. J., & Ricciardelli, L. A. (1999). Restrained drinking and cognitive control among adolescents. Adolescence, 34, 557−565.