Elsevier

Addictive Behaviors

Volume 37, Issue 2, February 2012, Pages 179-186
Addictive Behaviors

Self-medication or social learning? A comparison of models to predict early adolescent drinking

https://doi.org/10.1016/j.addbeh.2011.09.016Get rights and content

Abstract

The current study examined the effects of social anxiety, depressive symptoms, and alcohol expectancies of social behavior change on alcohol involvement to determine whether the self-medication and/or social learning models predicted drinking behavior in a sample of over 400 eighth grade students. Middle school students completed confidential surveys that assessed current alcohol use and expectancies as well as negative affectivity including social anxiety and depressive symptoms. Consistent with the self-medication hypothesis, depressive symptoms predicted more frequent and heavier alcohol use as well as solitary drinking. The social learning model was supported by a negative association between social anxiety and quantity/frequency of drinking and less drinking at parties, and a positive association between alcohol expectancies and all drinking outcomes. Additionally, social anxiety moderated the association between expectancies and alcohol use. These findings suggest that self-medication and social learning processes may both play a role in predicting early adolescent alcohol use and the contexts in which youths drink.

Highlights

►We predict drinking in middle school with self-medication and social learning models. ►Consistent with the SMH, depression predicted more heavy and solitary drinking. ►Consistent with a social learning model, social anxiety predicted less drinking. ►Social anxiety moderated the association between expectancies and alcohol use. ►Both models may play a role in early adolescent alcohol use and drinking contexts.

Introduction

Early adolescent alcohol use is a predictor of young adult alcohol dependence, other mental health problems (Grant and Dawson, 1997, Palmer et al., 2009), and delinquency (Dawkins, 1997, Gruber et al., 1996). Understanding the mechanisms that increase risk for early onset and progression of drinking is important for targeting effective prevention strategies. While negative affectivity including social anxiety and depression is consistently associated with increased alcohol use by early adulthood (e.g., Fergusson et al., 2009, Grant et al., 2004, Kushner et al., 2000), the role of these two types of internalizing symptoms in early alcohol involvement is unclear. Some studies have found that negative affect and depression symptoms predict alcohol use in adolescence (e.g., Colder and Chassin, 1997, Crum et al., 2008, Wu et al., 2008), and Zimmermann et al. (2003) found that baseline social phobia predicted the onset of hazardous alcohol use for youth. However, other studies have found internalizing symptoms to be negatively correlated with alcohol use over time in community samples of adolescents (Shedler and Block, 1990, Stice et al., 1998). In addition, social anxiety has been shown to provide some protection against the initiation of alcohol and other drug use (Myers, Aarons, Tomlinson, & Stein, 2003). The self-medication model and the social learning model are both theories that have been proposed to explain processes involved in the initiation, progression, and maintenance of alcohol use. The current study compares these two models by examining social anxiety, depression, and alcohol expectancies to predict alcohol use characteristics and drinking contexts among middle school students.

The self-medication hypothesis (Carrigan and Randall, 2003, Khantzian, 1985) is a motivation focused theory of substance use which hypothesizes that individuals use alcohol and other drugs to alleviate or cope with negative affect (Sher & Trull, 1994). There is evidence to support this model of adolescent substance use. Depression is a known risk factor for adolescent substance abuse (Burke et al., 1994, Neighbors et al., 1992), and self-report studies indicate that adolescents use substances after experiencing emotional distress (Brook et al., 1989, Hussong et al., 2008). Negative affect has also been found to significantly predict increase in substance use over time as youth transition from middle to late adolescence (Mason, Hitch, & Spoth, 2009). Social anxiety in particular may pose a salient risk for alcohol use because of the desire to reduce anxiety in social situations, such as parties or social gatherings, where drinking is likely to occur at this age. However, to date there have been no studies that have examined the drinking contexts of socially anxious youth in early adolescence. Studies of adolescents in treatment for psychiatric problems, substance use disorders, or both tend to find that social anxiety is a risk for alcohol involvement (e.g., Clark et al., 1995, Deas-Nesmith et al., 1998, Hovens et al., 1994). In these studies, social anxiety preceded substance involvement 65–100% of the time and the average time between onset of social anxiety and substance involvement was two years. By contrast, studies of community samples of adolescents have found either no relationship between social anxiety and substance involvement, or a protective relationship, in which social anxiety was seen to decrease the risk for subsequent substance use (e.g., Kandel et al., 1997, Myers et al., 2003). Together, these studies suggest that the self-medication model may be more useful in explaining processes involved in the maintenance or progression of problematic adolescent alcohol use that is associated with other forms of psychopathology rather than processes related to initiation or early adolescent drinking behaviors.

According to social learning theory, adolescent alcohol use is a learned behavior acquired through exposure to alcohol use models and social interactions that provide access and reinforcement for drinking (Petraitis, Flay, & Miller, 1995). The consistent strong association between adolescent drinking and their friends' drinking (e.g., Bauman and Ennett, 1996, Curran et al., 1997) is consistent with social learning, in which alcohol use is being modeled by one's peers. The social avoidance behavior and peer neglect experienced by socially anxious youth may decrease exposure to peer modeling of alcohol use as well as access to alcohol and interpersonal reinforcement at social events in which adolescents most typically drink. Adolescents with social anxiety often avoid participating in extracurricular activities and social events (Albano, Chorpita, & Barlow, 2003) and are invited to friends' homes and parties significantly less often than nonanxious youth (Spence, Donovan, & Brechman-Toussaint, 1999). In addition, anxious youth, including those with generalized distress, fear of social evaluation or new situations, may self-select peers who are less likely to model risk-taking behavior including early adolescent alcohol use (Clark, Thatcher, & Cornelius, 2008). For example, Fergusson and Horwood (1999) found that children in the highest decile of anxious and withdrawn symptoms at age ten were less likely to associate with peers who used alcohol and drugs at age fifteen compared to less anxious/withdrawn children.

Learning theories posit that through exposure to models of drinking behavior, cognitions about alcohol effects are developed, and these cognitions (expectancies) have a direct influence on drinking decisions and behavior (e.g., Brown, 1985, Christiansen and Goldman, 1983). Expectancies about the social behavior change effects produced by alcohol (e.g., “drinking alcohol will make me more outgoing” and “alcohol makes people act stupid at parties”) emerge during childhood and early adolescence, and are the best predictor of adolescent alcohol use (e.g. Aas et al., 1998, Greenbaum et al., 1995, Christiansen and Goldman, 1983), including initiation of alcohol use and early onset (elementary and middle school) drinking (Anderson et al., 2005, Killen et al., 1996, Reese et al., 1994). Studies also indicate that social facilitation expectancies predict drinking behavior in adults with social phobia (e.g., Ham, Hope, White, & Rivers, 2002). Additionally, social anxiety reduction expectancies were examined in late adolescents (Tran, Haaga, & Chambless, 1997), and among those who did not expect alcohol to reduce their social anxiety, high social anxiety participants drank significantly less than their low-anxiety counterparts. However, no studies have been conducted to explore the relationship between social anxiety, social behavior expectancies and drinking context in early adolescence.

The current study examined the effects of social anxiety, depressive symptoms, and alcohol expectancies of social behavior on alcohol involvement to determine whether the self-medication and/or social learning model better characterized alcohol use in early adolescence. We examined drinking behavior by selecting middle school students in the eighth grade who reported drinking alcohol in the past 30 days, and predicted the frequency, quantity and severity of their drinking as well as the contexts in which they reported alcohol use with separate multiple regression analyses. In order to determine whether or not drinking behaviors in our sample were consistent with either model, we made a priori predictions about specific patterns of relationships that should emerge based on each model.

The self-medication model predicts that negative affect in general will increase risk of alcohol use. Therefore, both social anxiety (SA) and depressive symptoms (DEP) should predict elevated drinking frequency and intensity and SA and DEP should interact to predict more severe drinking behavior than elevations in either symptom type alone. However, there should be evidence of symptom-specific influences on drinking context. Since youth high in SA are likely to self-medicate with alcohol in social situations, SA will predict higher rates of drinking at parties and with the opposite sex. Prior research suggests that depression is associated with solitary drinking among college drinkers (Christiansen, Vik, & Jarchow, 2002), and drinking to cope was found to predict solitary drinking (Gonzalez, Collins, & Bradizza, 2009). Therefore DEP symptoms will predict drinking alone. Finally, if youth with elevated SA symptoms are self-medicating by drinking alcohol in social settings, expectancies about the effect of alcohol use on social behavior (AEQ) will moderate the relationship between SA and alcohol use, such that high SA youth who expect alcohol to improve their ability to interact socially will use alcohol more than non-socially anxious youth in social contexts, whereas high SA youth who expect alcohol to have deleterious effects on their social behavior will be less likely to drink in social contexts than non-socially anxious youth.

The social learning model predicts that modeling of drinking behaviors and learned attitudes and beliefs about the effects of alcohol will predict drinking behavior. As in prior studies, positive AEQ will strongly predict alcohol use. However, SA will predict less frequent alcohol use overall and lower rates of drinking in social situations, presumably due to social avoidance behavior which will lead to less exposure to drinking models and access to alcohol. Furthermore, the relationship between AEQ and drinking will be diminished in SA youth due to less frequent opportunities for forming drinking-related social expectancies based on personal experience. Prior research suggests that girls low in SA are particularly at risk for heavier drinking in early adolescence (Tomlinson, Cummins, & Brown, in press), so we predict that the negative relationship between SA and drinking will be more pronounced in girls. Additionally, since prior research has found that the risk of early onset and progression of alcohol use differs by ethnicity (Luczak et al., 2006, Wallace et al., 2003), we conducted secondary analyses in which ethnicity was included as a covariate in order to test for ethnic differences in our models.

Section snippets

Participants

The current assessments were conducted as part of an anonymous survey on adolescent attitudes and behaviors related to alcohol and drug use. The sample included eighth grade students (N = 1058) in four schools which comprised an entire middle school district in the San Diego area. This district was selected as it reflects the socioeconomic and cultural diversity of the metropolitan area. Seventh grade students were not included in the present study because they were administered a state-funded

Association of social anxiety, depressive symptoms, and alcohol expectancies to frequency of drinking in a multiple regression analysis

We measured the frequency of drinking in the past 30 days with an ordinal variable that had four levels. An initial regression analysis revealed no differences between the two lower frequency groups (1–2 drinking episodes per month and 3–9 drinking episodes per month), and no differences between the higher two groups (10–19 drinking episodes per month and 20 or more drinking episodes per month). For parsimony, the groups were collapsed and a logistic regression was conducted comparing youth who

Discussion

The purpose of the current study was to compare the utility of two theories of alcohol use and misuse, the self-medication hypothesis and social learning theory, to further understanding of early adolescent alcohol involvement. Positive associations were found between depressive symptoms and alcohol use suggesting support for the self-medication hypothesis. However, alcohol use exhibited negative relations with social anxiety, and positive relations with social behavior change alcohol

Limitations

The present study is limited in a number of ways. The cross-sectional nature of the study limits our ability to draw causal inferences. For example, depressive symptoms could have been influenced by alcohol use rather than vice versa. In order to fully investigate the inter-relationships between social anxiety, depression, and alcohol expectancies, future studies are needed that assess these variables before the initiation of alcohol use and at multiple timepoints after the onset of drinking.

Conclusions

We found that that self-medication and social learning processes both play a role in predicting patterns of early adolescent alcohol use and the contexts in which youths drink. Depression symptoms already confer elevated risk for problematic drinking behavior, including binge drinking and drinking alone, and interventions that target depressed youth should include assessments of problem drinking. Youth with social anxiety who have not yet started drinking may benefit from prevention programs

Role of funding sources

This research was supported in part by a Ruth L. Kirschstein National Research Service Award (NRSA) pre-doctoral fellowship (F31 AA 13461) and a grant from the National Institute on Alcohol Abuse and Alcoholism (NIAAA; R01 AA 12171). The NIAAA and NRSA had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.

Contributors

K.L. Tomlinson and S.A. Brown designed the study and wrote the protocol. K.L. Tomlinson conducted the literature searches, statistical analysis, and wrote the first draft of the manuscript. S.A. Brown contributed to revisions of the manuscript, and both authors have approved the final manuscript.

Conflict of interest

All authors declare that they have no conflicts of interests.

References (64)

  • J.D. Killen et al.

    The role of expectancy and temperament in the prediction of drinking in a community sample of adolescents

    Addictive Behaviors

    (1996)
  • M.G. Kushner et al.

    The relationship between anxiety disorders and alcohol use disorders: A review of major perspectives and findings

    Clinical Psychology Review

    (2000)
  • B. Neighbors et al.

    Co-occurrence of substance abuse with conduct, anxiety, and depression disorders in juvenile delinquents

    Addictive Behaviors

    (1992)
  • R.H. Palmer et al.

    Developmental epidemiology of drug use and abuse in adolescence and young adulthood: Evidence of generalized risk

    Drug and Alcohol Dependence

    (2009)
  • H.N. Aas et al.

    Two-year longitudinal study of alcohol expectancies and drinking among Norwegian adolescents

    Addiction

    (1998)
  • L.S. Aiken et al.

    Multiple regression: Testing and interpreting interactions

    (1991)
  • A.M. Albano et al.

    Childhood anxiety disorders

  • K.G. Anderson et al.

    Elementary school drinking: The role of temperament and learning

    Psychology of Addictive Behaviors

    (2005)
  • K.E. Bauman et al.

    On the importance of peer influence for adolescent drug use: Commonly neglected considerations

    Addiction

    (1996)
  • N.M. Bekman et al.

    Affective and personality risk and cognitive mediators of initial adolescent alcohol use

    Journal of Studies on Alcohol and Drugs

    (2010)
  • J.S. Brook et al.

    Changes in drug involvement: A longitudinal study of childhood and adolescent determinants

    Psychological Reports

    (1989)
  • S.A. Brown

    Expectancies versus background in the prediction of college drinking patterns

    Journal of Consulting and Clinical Psychology

    (1985)
  • S.A. Brown et al.

    The Alcohol Expectancy Questionnaire: An instrument for the assessment of adolescent and adult alcohol expectancies

    Journal of Studies on Alcohol

    (1987)
  • J.D. Burke et al.

    Increased rates of drug abuse and dependence after onset of mood or anxiety disorders in adolescence

    Hospital & Community Psychiatry

    (1994)
  • B.A. Christiansen et al.

    Alcohol-related expectancies versus demographic/background variables in the prediction of adolescent drinking

    Journal of Consulting and Clinical Psychology

    (1983)
  • J.R. Ciesla et al.

    Reliability over time of self-reports given by adolescents and their parents in substance abuse outcome research

    Journal of Child & Adolescent Substance Abuse

    (1999)
  • D.B. Clark et al.

    Identifying anxiety disorders in adolescents hospitalized for alcohol abuse or dependence

    Psychiatric Services

    (1995)
  • D.B. Clark et al.

    Anxiety disorders and adolescent substance use disorders

  • C.R. Colder et al.

    Affectivity and impulsivity: Temperament risk for adolescent alcohol involvement

    Psychology of Addictive Behaviors

    (1997)
  • P. Curran et al.

    The relation between adolescent alcohol use and peer alcohol use: A longitudinal random coefficients model

    Journal of Consulting and Clinical Psychology

    (1997)
  • M.P. Dawkins

    Drug use and violent crime among adolescents

    Adolescence

    (1997)
  • D. Deas-Nesmith et al.

    Comorbid substance use and anxiety disorders in adolescents

    Journal of Psychopathology and Behavioral Assessment

    (1998)
  • Cited by (63)

    • Characteristics and reasons for use associated with solitary alcohol and marijuana use among U.S. 12th Grade Students, 2015–2021

      2022, Drug and Alcohol Dependence
      Citation Excerpt :

      The current study supported and extended prior research on associations between reasons for use and solitary use. Earlier research using smaller samples (Creswell et al., 2014; Skrzynski and Creswell, 2021; Tomlinson and Brown, 2012) identified associations between adolescent solitary alcohol use and depressive symptoms, negative affect, negative reinforcement, and social discomfort. One study using data from five U.S. regions identified associations between solitary alcohol and marijuana use and coping motives among American Indian adolescents (Guo et al., 2021).

    • Desire to drink as a function of laboratory-induced social stress among adolescents

      2021, Journal of Behavior Therapy and Experimental Psychiatry
      Citation Excerpt :

      For example, in a large, school-based study, Windle and Windle (2012) found that social anxiety during high school prospectively predicted coping (but not social or enhancement) motives for alcohol consumption in young adulthood. With a few exceptions (cf. Tomlinson & Brown, 2012), this body of work underscores escaping and/or avoiding social stress as central to understanding the link between social anxiety and alcohol use problems. However, the majority of this research relies on retrospective self-report, and has been conducted almost exclusively with adult samples.

    • Understanding the relationship between depression and alcohol among students

      2021, The Neuroscience of Depression: Features, Diagnosis, and Treatment
    View all citing articles on Scopus
    View full text