Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Sep 1.
Published in final edited form as: Subst Use Misuse. 2020 May 19;55(11):1765–1773. doi: 10.1080/10826084.2020.1762652

Structured and unstructured activities and alcohol and marijuana use in middle-school: The role of availability and engagement

Nichea S Spillane a, Melissa R Schick a, Katelyn T Kirk-Provencher a, Danielle C Hill a, Janan Wyatt a, Kristina M Jackson b
PMCID: PMC8409240  NIHMSID: NIHMS1618347  PMID: 32423269

Abstract

BACKGROUND:

Behavioral Theories of Choice applied to substance use suggests that use depends on availability of substances and alternative activities. Applying this theory to younger adolescents offers the possibility of investigating potentially malleable prevention and intervention targets.

OBJECTIVES:

The current study examines the role of perceived availability and engagement in structured and unstructured activities on adolescent alcohol and marijuana use controlling for substance availability.

METHODS:

Data were collected as part of a longitudinal study of 6th-8th graders (N=1,023; 52% female; 76% White; Mage=12.23 years). Multiple logistic regressions analyzed the impact of availability and engagement in structured and unstructured activities at Wave 1on heavy drinking and marijuana use by study end in both unadjusted and adjusted models.

RESULTS:

Availability of unstructured activities was associated with increased likelihood of both marijuana use (b*=.06, p=.04) and heavy drinking (OR=1.20, p<.001), while availability of structured activities was not significantly associated with likelihood of either marijuana use (b*=−.10, p=.07) or heavy drinking (OR=0.86, p=.16). Engagement in unstructured activities was significantly associated with increased likelihood of both marijuana use (b*=.06, p=.03) and heavy drinking (OR=1.11, p=.003), while engagement in structured activities was significantly associated with decreased likelihood of both marijuana use (b*-=.25, p<.001) and heavy drinking (OR=0.85, p=.046).

CONCLUSIONS:

Perceived availability of and engagement in unstructured activities may present a risk, while perceived availability of and engagement in structured activities may serve as a protective factor for youth substance use.

Keywords: Activities, Substance Use, Youth, Adolescents, Behavioral Theories of Choice

Initiation of substance use earlier in life is associated with higher rates of use and dependence, as well as a variety of negative health, social, and behavioral issues later in life (Griffin & Botvin, 2010). Monitoring the Future (MTF) data highlights the steep increase in substance use over the middle and high school years. Specifically, MTF reports that rates of past two-week binge drinking were 4%, 9%, and 14%, and rates of current marijuana use were 6%, 17%, and 22% for 8th, 10th, and 12th graders respectively (Johnston et al., 2019). Rates of having ever been drunk were 9%, 26%, and 43%, and ever using marijuana were 14%, 33%, and 44%, for 8th, 10th, and 12th graders, respectively (Johnston et al., 2019). Adolescent substance use is associated with a host of short- and long-term negative outcomes including risky sexual behavior (Hall et al., 2016; Palen, Smith, Flisher, Caldwell, & Mpofu, 2006), school dropout (Green et al., 2016; Townsend, Flisher, & King, 2007), future substance use/problems (Grant et al., 2006; Green et al., 2016), and injury, violence, and suicide (Bauman & Phongsavan, 1999; Hall et al., 2016). Thus, it is imperative to develop effective prevention programs to target high-risk substance use behaviors in young adolescents.

A useful theoretical framework that offers potentially malleable alternatives to substance use is behavioral theories of choice (BTC; Audrain-McGovern et al., 2004; Bickel & Vuchinich, 2000). According to BTC, substance use varies based on two factors: availability of substances and availability of alternative reinforcers (Correia, Simons, Carey, & Borsari, 1998; Vuchinich & Tucker, 1988a). Factors influencing availability of alternative activities to substance use include income, cost, and temporal availability of the activity (Andrabi, Khoddam, & Leventhal, 2017; Correia, Murphy, Irons, & Vasi, 2010). For instance, among college students the choice to drink is linked to the relative availability and cost of alcohol in addition to the relative availability and cost of substance-free alternative activities (Murphy, Correia, & Barnett, 2007; Murphy & Dennhardt, 2016). For youth, availability of substances is one of the strongest correlates of initiation and use (Broman, 2016), however, there has been less attention paid to the relationship of availability of substances and alternative reinforcers in predicting substance use in the same model, as suggested by BTC, especially among young adolescents.

There is some research to suggest that higher availability and actual engagement in extracurricular activities are related to lower levels of substance use among adolescents (Moilanen, Markstrom, & Jones, 2014). A broader class of extracurricular activities are leisure activities (i.e., activities that youth participate in when not in school). High youth engagement in leisure activities has been shown to be associated with better life outcomes and academic performance (Bartko & Eccles, 2003), psychological well-being (Bartko & Eccles, 2003), and a clear sense of identity (Palen & Coatsworth, 2007). Additionally, adolescent engagement in a greater breadth of organized leisure activities has been linked with more positive psychological, behavioral (e.g., substance use), and educational outcomes by the 10th grade compared to adolescents with less engagement in such activities (Sharp, Tucker, Baril, Van Gundy, & Rebellon, 2015). These activities may play a buffering role against the effects of stress (Coleman, 1993; Sharp et al., 2015) and tend to be protective against the use of substances among youth (Eccles & Barber, 1999; Sharp et al., 2015).

Engagement in one specific type of activity, physical activity, has also been shown to be related to lower alcohol use, binge drinking, and marijuana use in 8th graders (Terry-McElrath, O’Malley, & Johnston, 2011); at the same time participation in athletic teams is associated with higher levels of alcohol use in high school (Eccles & Barber, 1999; Terry-McElrath et al., 2011). These results suggest that a complex relationship exists between participating in sports/exercise and participating in risk behavior and prosocial behavior, which may be partially explained by peer effects (e.g., peer pressure; Seek Moon & Rao, 2011). Perhaps another reason for the somewhat mixed findings with sports/exercise may be the type of activity that is under investigation, such as structured and unstructured activities. Unstructured activities are those that provide fewer opportunities for parental monitoring and increased likelihood of involvement with older peers, allowing for opportunities for initial experimentation with alcohol and other substances (Mahoney & Stattin, 2000). Engagement in unstructured activities is associated with more engagement in substance use and deviant behaviors, whereas engagement in structured activities is protective against these outcomes (Mahoney & Stattin, 2000).

Additionally, availability of alternative activities offers potentially important prevention implications that have not been made readily apparent in the existing literature, as youth cannot chose to engage in an activity that is unavailable to them. There may be numerous factors influencing availability of activities, including SES, access to transportation, and urbanicity versus rurality. Further, there is evidence to suggest that perception of opportunities to engage in behaviors influences actual engagement. For instance, some work has found that perceived availability of physical activity facilities in one’s community is significantly associated with engagement in organized sports and physical activity during leisure time (Prins, Oenema, van der Horst, & Brug, 2009). Thus, it is of great importance to examine the influence of activity availability – that is, the opportunity to participate in activities – on substance use to inform future prevention efforts.

Because youth participation in activities is a function of both deliberate choice (engagement) and opportunity (availability of activities), it is critical to understand each component, as is suggested by BTC (Vuchinich & Tucker, 1988b). The efficacy of a prevention strategy will be largely contingent on not only whether adolescents choose to engage in an alternative activity to substance use, but also whether or not the activity is available to them, as one cannot choose to engage in an activity that is not available. Further, there has been a paucity of research focusing on both substance availability and engagement and alternative activity availability and engagement in the same model, especially among young adolescents. As noted above, early adolescence is an ideal time for implementation of prevention strategies, given that early initiation of substance use is associated with negative short and long-term outcomes (Kahler, Hoeppner, & Jackson, 2009; WHO, 2014) and studies have found a rapid increase in initiation of alcohol use from ages 12 to 14 (Forman–Hoffman, Batts, Hedden, Spagnola, & Bose, 2018) which strongly suggest a need to evaluate the activities of youth in this age range. While these studies suggest that activities play an important role in youth substance use, nearly all of the studies have focused on older adolescents (i.e., high school students), and there is a gap in our understanding of younger adolescents.

Thus, the goal of the present paper is to study the role of both perceived availability and engagement in structured and unstructured activities and their relationship to alcohol and marijuana use in a sample of middle schoolers. In line with BTC, we hypothesized that perceived availability of structured activities will be negatively associated with endorsing marijuana use and heavy drinking, while perception of having more unstructured activities available would be positively associated with the likelihood of engaging in marijuana use or reporting heavy drinking. Finally, we hypothesized that engagement in structured activities would be protective and, therefore, associated with a lower likelihood of using marijuana or endorsing heavy drinking, while engaging in more unstructured activities would be associated with a greater likelihood of reporting marijuana use and heavy drinking. Alcohol and marijuana availability were included as covariates given that they have been robustly found to be associated with initiation of substance use among adolescents, and as they are a key component of BTC (Broman, 2016). Peer substance use and parental monitoring were chosen as covariates given that they have been identified as important contextual factors which influence students’ decisions to use substances (Mason et al., 2017; Monahan, Rhew, Hawkins, & Brown, 2014; Rodgers-Farmer, 2001). Finally, we controlled for alcohol and marijuana use at Wave 1 to prospectively examine the association of earlier availability of and engagement in activities on later heavy drinking and marijuana use.

Materials and Methods

Sample

The present study represents a prospective secondary analysis of data drawn from a prospective study on alcohol initiation and progression among early adolescents (Jackson, Barnett, Colby, & Rogers, 2015; Jackson, Colby, Barnett, & Abar, 2015). Students from six Rhode Island middle schools (one urban, two rural, three suburban) were invited to participate in a 3-year study resulting in a total sample of 1,023 participants. Using a cohort-sequential design (Nesselroade & Baltes, 1979), three grade cohorts were followed (6th, 7th, and 8th) with data collected in five school cohorts enrolled 6 months apart between October 2009 and October 2011. All procedures were approved by the institutional review board at the university. For sample demographic characteristics, see Table 1.

Table 1.

Sample Characteristics (N = 1,023).

% M (SD) Range
Grade:
 6th 33.0
 7th 32.1
 8th 34.9
Female 52.2
Race/Ethnicity
 White 72.1
 Black 4.0
 Hispanic 12.2
 Other 11.6
Lunch Subsidy 36.0
Peer Behavior – Substance Use 0.26 (0.84) 0 – 6
Parental Monitoring 4.12 (0.82) 1 – 5
Activity Availability
 Structured Activities (mean percent of assessed activities available) 85.6 85.6 (18.5)
 Unstructured Activities (mean percent of assessed activities available) 83.8 83.8 (16.8)
Activity Engagement
 Structured Activities (mean percent) 56.6
 Unstructured Activities (mean percent) 52.9
Substance Availability
 Alcohol Available (yes) 26.8
 Marijuana Available (yes) 12.6
Substance Use
 W1 Heavy Drinking (3+ drinks) 7.7
 W1 Marijuana Use 6.4
 Ever Heavy Drinking (3+ drinks) 16.2
 Ever Marijuana Use 24.9
Open in a new tab

Procedure

Informational materials were distributed through schools and by mail to all students on the registrar list; interested youth who provided written parental consent were invited to attend a 2-hour group orientation session (for a detailed description of the methods please see Jackson et al., 2015b). Following orientation, student informed assent was obtained and there was a 45-minute baseline (Wave [W] 1) web-based survey, completed at the schools on laptops provided by the study. Participants received a $25 gift card for completing this assessment. Following W1, participants completed a series of four semiannual (every 6 months) surveys (W2-W5) and a fifth follow-up survey one year later (W6). Follow-up assessments (i.e., W2-W6) were conducted using web-based surveys outside of school, and the students were compensated a $20 mall gift card. Retention rates were high across all six waves (ranging from 83% to 92%).

Measures

Perceived availability of and engagement in structured and unstructured activities.

Seventeen items were taken from the original 25-item Adolescent Reinforcement Survey Schedule (ARSS; Holmes, Heckel, Chestnut, Harris, & Cautela, 1987). To make the measure more relatable to a younger age cohort (the original ARSS was developed for college students), in the 21st century (original measure was developed in 1980s), items were changed and eight items were added to ensure appropriateness of the measure. The items were preceded by the prompt “Rate the frequency of each item over the past month,” and included such activities as participating in after school clubs and playing computer or video games. Response options ranged from “Less than once a week” (0) to “More than once a day” (5), plus “I don’t have the opportunity to do this” (6) and “I have the opportunity to do this, but I choose not to” (7). To assess for availability of activities, we dichotomized items into available, collapsing across categories (response options 0–5 and 7 =1) and not available (response option 6 = 0). To assess for engagement in activities, we dichotomized variables into yes (response options 0–5 = 1) and no (response options 6–7 = 0). Next, we grouped items based on whether they were structured (7 items) or unstructured (18 items) according to Mahoney and Stattin (2000), where structured activities require regular participation schedules, rule-guided engagement, direction and supervision by one or more adult activity leaders, emphasis on skill-building, requiring sustained attention, and clear feedback regarding progress or growth. Activities that did not fall within the structured activities because they did not meet the criteria described above were considered unstructured (Mahoney & Stattin, 2000). For examples of activities assessed, see Table 2. Structured and unstructured activities were measured at W1. We calculated two separate variables: percent of students endorsing structured and unstructured activities as available, and percent of students endorsing activities they have engaged in.

Table 2.

Summary of activities assessed (N = 1,023).

Activity % of adolescents reporting the activity as being available % of adolescents reporting engaging in the activity
Structured Activities
1. Participate in sports at school 83.3 44.6
2. Participate in organized sports outside of school 89.0 61.3
3. Participate in after-school activities (e.g., clubs, bands) 85.2 47.1
4. Studying/homework 97.6 89.3
5. Doing chores at home 92.3 78.8
6. Play a musical instrument, music lessons/practice 78.7 40.8
7. Participate in religious organizations (e.g., youth group, go to church) 71.6 32.4
Unstructured Activities
8. Play sports after school with friends or in your neighborhood (e.g., basketball) 86.8 57.9
9. Talk or text on the phone with friends 84.8 69.8
10. Go to parties with friends 72.0 17.3
11. Go hang out where friends meet (e.g., at the mall) 73.9 39.4
12. Write/receive email from friends 78.9 45.7
13. Read a magazine or book for pleasure 94.5 67.9
14. Go on a bike ride 87.0 49.3
15. Go to a movie 83.4 19.8
16. Listen to music (e.g., CDs, MP3 player, radio) 96.8 87.2
17. Buy clothes at stores or the mall 85.7 27.9
18. Watch TV/videos/DVDs 97.8 89.4
19. Play computer or video games 96.8 79.5
20. IM (instant messaging) 67.4 41.9
21. Do arts and crafts (e.g., drawing, scrapbooking, photography, model airplanes, etc.) 88.8 46.3
22. Connect with people online (e.g., Facebook, MySpace, message boards, chat rooms) 75.1 55.4
23. Go on the Internet (e.g., online shopping, watch things on YouTube, look up information on Wikipedia, etc.) 92.8 77.9
24. Hang out at a friend’s house 90.9 62.3
25. Go to a community center such as Boys and Girls club, after-school program 51.3 13.2
Open in a new tab

Substance Use Outcomes

Alcohol.

Students were asked at each wave if they ever had a heavy drinking episode, as heavy drinking in adolescence is associated with higher risk of future problems (Hill, White, Chung, Hawkins, & Catalano, 2000). A heavy drinking episode was defined in this age group as drinking 3+ drinks per occasion (Donovan, 2009).We created the cumulative heavy drinking variable by assigning a 1 (yes) to students who reported ever having a heavy drinking episode at any time point (W1 – W6) and a 0 (no) for those who did not.

Marijuana.

Students were asked at each wave if they had ever smoked marijuana. We created the marijuana variable by assigning a 1 (yes) to students who reported ever using marijuana at any time point (W1- W6) and a 0 (no) for those who did not.

Covariates

Grade, gender, race/ethnicity (recoded as a nominal variable to reflect the 4 categories listed in Table 1), and free or reduced lunch were measured as covariates. Parents were mailed home a questionnaire to report on whether their children receive free or reduced lunch at school, as this parent report has been shown in other research to be a valid proxy of SES (Nicholson, Slater, Chriqui, & Chaloupka, 2014). Other relevant environmental variables were included as covariates: alcohol availability, marijuana availability, peer substance use, and parental monitoring (Broman, 2016; Lee, 2012).

Alcohol and marijuana availability.

Students were asked at W1 if alcohol and marijuana were available to them. Response options were yes (1) or no (0) for each substance.

Peer substance use behavior.

A 6-item questionnaire to assess respondent’s close friends’ substance use was adapted from a measure on peer delinquent behavior from Arthur et al (2002). Students were asked whether or not (yes = 1; no = 0) their friends engaged in various substance use behaviors, such as drinking alcohol, smoking cigarettes, and using marijuana. A mean was taken across the items with higher scores indicating a greater number of substance-using friends.

Parental monitoring.

Parental monitoring was measured with nine items taken from Kerr and Stattin (2000) assessing knowing where their child was, who their friends were, etc. Items were on a five-point scale ranging from No, never (0%) to Yes, always (100%); a mean was taken across the nine items (α=.87).

Data Analyses

First, we examined the distributions and bivariate relationships (i.e., Pearson product-moment correlations for continuous variables and point-biserial correlations for dichotomous variables) among all variables to assess for assumptions of the general linear model (GLM). Then we conducted eight logistic regression models to evaluate (separately) the association between availability of and engagement in activities at W1 and substance use outcomes at all timepoints in both unadjusted models and adjusting for substance availability and other covariates. In the unadjusted models and adjusted models with heavy drinking as the outcome, all results are presented as odds ratios with 95% confidence intervals. In the adjusted models with marijuana use as the outcome, bootstrapping was used with 1,000 samples to estimate the properties of the sampling distribution from the sample because so few individuals had engaged in marijuana use at baseline that the models were not able to run. In these cases, results are presented as bootstrapped estimates (denoted with a *). For each model, listwise deletion was utilized for cases with missing data to allow for complete-case analysis.

Results1

Bivariate Correlations

Bivariate correlations were calculated for both predictor and outcome variables (see Table 3). Perceiving fewer structured activities as available was associated with ever using marijuana, but not with heavy drinking. Engaging in fewer structured activities was associated with both ever using marijuana and ever heavy drinking. Having more unstructured activities perceived as available and engaging in more unstructured activities were associated with being more likely to report ever having a heavy drinking episode or ever using marijuana.

Table 3.

Bivariate correlations among variables of interest (N = 1,023).

1 2 3 4 5 6 7 8 9 10 11 12 13
1. Grade - −.02 −.03 .28*** .15*** −.17*** .17*** .29*** .30*** .20*** .08* .15*** −.05
2. Male gender - −.01 −.09** −.06 −.02 −.01 .04 .02 −.14*** −.10** −.16*** −.02
3. Free/reduced lunch - .05 .05 −.09** .01 −.05 −.05 −.16** −.10** .01 −.14***
4. Ever heavy drinking - .39*** −.23*** .29*** .31*** .32*** .17*** .02 .16*** −.09**
5. Ever marijuana use - −.23*** .29*** .28*** .16*** .08* −.08* .11*** −.11**
6. Parental monitoring - −.31*** −.29*** −.25*** .04 .18*** −.004 .29***
7. Peer models of substance use - .47*** .29*** .12*** −.01 .15*** −.09**
8. Marijuana available - .47*** .14*** −.002 .11*** −.11***
9. Alcohol available - .16*** .06 .07* −.08*
10. Unstructured activities available - .54*** .62*** .24***
11. Structured activities available - .26*** .47***
12. Unstructured activities - .34***
13. Structured activities -
Open in a new tab

Note:

*

p<.05,

**

p<.01,

***

p<.001.

Logistic Regression Results for Perceived Availability and Substance Use Outcomes

Logistic regression results for perceived availability of structured and unstructured activities are presented in Table 4.

Table 4.

Association of availability of structured and unstructured activities and heavy drinking and marijuana use.

Heavy Drinking Marijuana Use
Parameter b SE p OR OR 95% CI b/b * SE p OR 95% CI
Unadjusted a
Constant −4.47 .69 <.001 0.01 −1.45 .43 .001 .23
Unstructured Activities .28 .05 <.001 1.32 1.20, 1.45 .13 .03 <.001 1.14 1.07, 1.22
Structured Activities −.25 .08 .003 0.78 0.66, 0.92 −.29 .07 <.001 .75 0.66, 0.86
Adjusted b, c
Constant −3.95 .96 <.001 0.02 −1.17 .59 .036 −2.35, −0.05
Unstructured Activities .18 .05 .001 1.20 1.08, 1.33 .06 .03 .038 .003, .12
Structured Activities −.15 .11 .156 0.86 0.70, 1.06 −.10 .07 .133 −.22, .02
Covariates
Male gender1 −.58 .23 .011 0.56 0.36, 0.88 −.16 .19 .392 −.48, .23
Grade2
7th 1.15 .38 .003 3.14 1.48, 6.66 .71 .27 .006 .20, 1.27
8th 1.68 .37 <.001 5.39 2.63, 11.02 .58 .26 .022 .09, 1.10
Race3
 Black, non-Hispanic −.25 .56 .616 0.78 0.26, 2.33 .40 .46 .352 −.55, 1.28
 Hispanic −.30 .37 .422 0.74 0.36, 1.54 −.21 .32 .515 −.83, .39
Other, non-Hispanic −.41 .36 .249 0.66 0.33, 1.34 .33 .28 .215 −.25, .84
Lunch Subsidy .25 .24 .292 1.28 0.81, 2.03 .19 .20 .315 −1.19, .60
Parental Monitoring −.31 .14 .021 0.73 0.56, 0.95 −.27 .12 .026 −.50, −.04
Peer Substance Use .10 .11 .378 1.10 0.87, 1.37 .18 .12 .099 −.05, .42
Alcohol/Marijuana available .98 .23 <.001 2.66 1.68, 4.20 .78 .27 .003 .27, 1.32
W1 Alcohol/Marijuana Use 2.45 .35 <.001 11.59 5.79, 23.19 22.07 .22 .001 21.67, 22.55
Open in a new tab

Note:

a

n = 995 included in analyses for unadjusted heavy drinking and unadjusted marijuana models,

b

n = 893 included in analyses for adjusted heavy drinking model,

c

n = 890 included in analyses for adjusted marijuana model,

*

denotes a bootstrapped estimate, reference group is denoted in italics; For the heavy drinking adjusted model, Nagelkerke R2 = .39, for the marijuana use adjusted model, Nagelkerke R2 = .33;

1

Reference group is female gender;

2

Reference group is 6th grade,

3

Reference group is non-Hispanic White.

Heavy drinking.

To estimate the independent association with perceived availability of both structured and perceived availability of unstructured activities, we conducted a binary logistic regression with the outcome as ever reporting a heavy drinking day. A second logistic regression included covariates and other predictors to examine the unique associations between perceived availability of structured and unstructured activities and ever heavy drinking. In the unadjusted model, perceived availability of unstructured activities was associated with a higher likelihood of ever endorsing heavy drinking (p<.001, OR=1.32; 95%CI [1.20, 1.45]) and perceived availability of structured activities was associated with a lower likelihood of ever endorsing heavy drinking (p=.003, OR=0.78; 95%CI [0.66, 0.92]). With covariates entered into the model, availability of structured activities became non-significant (p=.156, OR=0.86; 95% CI [0.70, 1.06]). However, availability of unstructured activities remained significant (p=.001, OR=1.20; 95%CI [1.08, 1.33]).

Marijuana use.

A similar approach was used to estimate the association between perceived availability of structured and perceived availability of unstructured activities and marijuana use both without and with adjusting for covariates. In the unadjusted model, perceived availability of unstructured activities predicted higher likelihood of ever using marijuana (p<.001, OR=1.14; 95%CI [1.07, 1.22]) and perceived availability of structured activities predicted lower likelihood of ever using marijuana (p<.001, OR=0.75; 95%CI [0.66, 0.86]). After controlling for covariates and marijuana availability, perceived availability of structured activities was not significantly associated with ever using marijuana (p=.133, b*=−.10; 95%CI [−0.22, 0.02]), while availability of unstructured activities remained significant (p=.038, b*=.06; 95%CI [.003, .12]).

Logistic Regression Results for Activity Engagement and Substance Use Outcomes

Logistic regression results for engagement in structured and unstructured activities are presented in Table 5.

Table 5.

Association of engagement in structured and unstructured activities and heavy drinking and marijuana use.

Heavy Drinking Marijuana Use
Parameter b SE p OR OR 95% CI b/b* SE p OR 95% CI
Unadjusted a
Constant −2.25 .30 <.001 0.11 −1.32 .25 <.001 0.27
Unstructured Activities .17 .03 <.001 1.19 1.13, 1.25 .12 .02 <.001 1.13 1.08, 1.18
Structured Activities −.30 .06 <.001 0.74 0.66, 0.83 −.25 .05 <.001 0.78 0.71, 0.86
Adjusted b, c
Constant −2.83 .75 <.001 0.06 −1.17 .61 .051
Unstructured Activities .11 .04 .003 1.11 1.04, 1.20 .06 .03 .032 .01, .12
Structured Activities −.16 .08 .046 0.85 0.73, 0.99 −.10 .07 .142 −.23, .04
Covariates
Male gender1 −.54 .23 .018 0.58 0.37, 0.91 −.16 .19 .418 −.54, .21
Grade2
7th 1.14 .38 .003 3.13 1.49, 6.58 .71 .25 .004 .27, 1.24
8th 1.70 .36 <.001 5.49 2.71, 11.14 .58 .25 .010 .13, 1.12
Race3
 Black, non-Hispanic −.37 .55 .497 0.69 0.24, 2.02 −.40 .44 .333 −.56, 1.20
 Hispanic −.32 .37 .382 0.72 0.35, 1.50 −.21 .32 .488 −.90, .39
Other, non-Hispanic −.42 .36 .242 0.66 0.33, 1.33 .33 .28 .215 −.25, .87
Lunch Subsidy .19 .23 .424 1.21 0.76, 1.90 .19 .20 .317 −.21, .58
Parental Monitoring −.25 .14 .071 0.78 0.60, 1.02 −.27 .12 .021 −.51, −.04
Peer Substance Use .09 .11 .431 1.09 0.88, 1.36 .18 .12 .105 −.05, .43
Alcohol/Marijuana available 1.07 .23 <.001 2.92 1.85, 4.60 .78 .27 .002 .26, 1.29
W1 Alcohol/Marijuana Use 2.36 .35 <.001 10.57 5.36, 20.84 22.07 .21 .001 21.70, 22.55
Open in a new tab

Note:

a

n = 995 included in analyses for unadjusted heavy drinking and unadjusted marijuana models,

b

n = 893 included in analyses for adjusted heavy drinking model,

c

n = 890 included in analyses for adjusted marijuana model, reference group is denoted in italics; For the heavy drinking adjusted model, Nagelkerke R2 = .33, for the marijuana use adjusted model, Nagelkerke R2 = .39;

1

Reference group is female gender;

2

Reference group is 6th grade,

3

Reference group is non-Hispanic White.

Heavy drinking.

We conducted a binary logistic regression with the outcome being ever reporting a heavy drinking day to estimate the independent association between engagement in structured and engagement in unstructured activities with alcohol use. Then, a second logistic regression model included covariates and other variables to examine the relationship between actual engagement in structured and unstructured activities and ever heavy drinking. The unadjusted model showed that engaging in more unstructured activities was associated with an increased likelihood of ever endorsing heavy drinking (p<.001, OR=1.19; 95%CI [1.13, 1.25]) and engagement in a greater number of structured activities was associated with a lower likelihood of ever endorsing heavy drinking (p<.001, OR=0.74; 95%CI [0.66, 0.83]). Next, covariates were entered into the model and results indicated that engaging in more structured activities was associated with a lower likelihood of ever endorsing heavy drinking (p=.046, OR=0.85; 95%CI [0.73, 0.99]). Meanwhile, engaging in more unstructured activities was associated with a higher likelihood of ever endorsing heavy drinking (p=.003, OR=1.11; 95%CI [1.04, 1.20]).

Marijuana use.

Next, we estimated the association between actual engagement in structured and unstructured activities and marijuana use both without and with adjusting for covariates. In the unadjusted model, engaging in unstructured activities was associated with higher likelihood of ever using marijuana (p<.001, OR=1.13; 95%CI [1.08, 1.18]) and engaging in more structured activities was associated with lower likelihood of ever using marijuana (p<.001, OR=0.78; 95%CI [0.71, 0.86]). After controlling marijuana availability and covariates, results suggested that engaging in structured activities was not significantly associated with ever using marijuana (p=.142, b*=−.10; 95%CI [−0.23, 0.04]). Finally, the relationship between engaging in a greater number of unstructured activities and ever using marijuana was found to be significant (p=.032, b*=.06; 95%CI [0.01, 0.12]).

Discussion

BTC posits that the decision to use substances is based on availability of the substance and availability of alternative activities (Correia et al., 1998; Vuchinich & Tucker, 1988a). Although availability of alternative activities is discussed in the theory (Murphy et al., 2007; Vuchinich & Tucker, 1988a), it is often not used as the variable of interest in relation to substance use outcomes (Palen & Coatsworth, 2007; Terry-McElrath et al., 2011), nor studied in middle school students. In the present study, we were interested in isolating the effects of perception of availability of structured and unstructured activities and engagement in such activities, motivated by theoretical considerations about actual engagement in activities (Mahoney & Stattin, 2000) versus availability of activities (Prins et al., 2009). Solely focusing on engagement in activities may gloss over the fact that lack of engagement could be the result of lack of availability (e.g., due to socioeconomic status, time, transportation-related constraints) rather than deliberate choice to not engage, and thus be less of an indicator of risk than engagement per se. Because middle school represents a time that many youth may start experimenting with substance use (Forman–Hoffman et al., 2018), it is important to evaluate factors that may prevent or contribute to risk.

Consistent with prior literature demonstrating that as adolescents age, their likelihood of having used substances increases (NIDA, 2020), we found that students in 7th and 8th grade were significantly more likely than 6th grade students to have used marijuana or to have had a heavy drinking episode. Further, we found that female participants were significantly less likely to report having used marijuana use or having at least one heavy drinking episode. This is consistent with national trends, in which adolescent females have been found to have rates of marijuana and alcohol use that are equal to or higher than their male counterparts (SAMHSA, 2016). Nonetheless, future research should continue to examine how such demographic characteristics influence these associations. After accounting for demographics and other relevant factors, we found that having alcohol available substantially increased the odds of endorsing a heavy drinking episode. Even after accounting for the powerful effect of alcohol availability and other covariates, the perception of having a greater number of unstructured activities available increased the odds of ever endorsing a heavy drinking episode for this young sample; notably this was a unique effect over and above number of structured activities available. However, the relationship between perceived availability of structured activities and ever endorsing a heavy drinking day did not reach significance after controlling for all covariates and number of unstructured activities available. Similarly, after controlling for demographics and other relevant factors, having marijuana available increased the likelihood of ever reporting marijuana use. In addition, after controlling for covariates and availability of structured activities, we found that perceiving a greater number of unstructured activities as available increased the likelihood of endorsing using marijuana. It is important to note that the observed effects accounted for those youth who had already initiated alcohol or marijuana use. Therefore, our results suggest that youth who have a greater number of unstructured activities available to them are at an increased risk of subsequently endorsing later heavy drinking or marijuana use. Similarly, youth who report actually engaging in a greater number of unstructured activities are at a greater risk of reporting heavy drinking or marijuana use, after accounting for engaging in structured activities. It appears that even after controlling for other variables, engaging in structured activities appears to play a protective role in that youth report a lower likelihood of endorsing heavy drinking.

Our findings that availability of and engagement in unstructured activities increase the likelihood of youth engaging in heavy drinking and marijuana use suggest the need to consider which factors are contributing to that risk. Based on prior literature, it seems possible that some of these unstructured activities expose adolescents to other substance-using peers (Hoeben, Meldrum, Walker, & Young, 2016) and that there is continued influence by these peers (including access to alcohol) even when the adolescent is not engaging in the activity per se. However, while peer substance use was associated at the bivariate level with both heavy drinking and marijuana use, this effect was diminished when entered into the model with availability of and engagement in unstructured activities. Further, given the robust effect of unstructured activities after controlling for availability of substances, availability of activities appears to have unique contributions and offers an additional intervention target. Further, our models also control for parental monitoring, indicating that unstructured activity availability presents a unique effect even for those whose parents have some knowledge of their activities and behaviors.

In contrast, increasing opportunities to engage in more structured activities may mitigate this risk by increasing access to non-substance using same-aged peers (Mahoney & Stattin, 2000). Additionally, unstructured activities are characterized by low adult supervision and monitoring (Mahoney & Stattin, 2000), which we found to be associated with greater likelihood of endorsing either substance use outcome. The finding that structured activities was no longer a significant predictor when substance-using peers and parental monitoring were added to models suggests that these constructs may serve as mechanisms underlying the effect of structured activities on substance, although testing formal mediation is beyond the scope of the current paper. Nevertheless, our results regarding the important role of parental monitoring suggests the importance of raising awareness among parents, caregivers, and other important adults regarding the increased risk posed by availability of substances to youth. Increasing the availability of structured activities for youth, thereby decreasing time available to engage in unstructured activities, may be one way to increase the presence of supportive adults who assist with progressing towards clear activity goals (Mahoney & Stattin, 2000).

Given that availability of and engagement in unstructured activities each appeared to present risk of later substance use, while structured activities appeared to confer some protective benefit, increasing the structure of activities in relatively unstructured environments such as community centers, the YMCA or the Boys and Girls Clubs may constitute an effective prevention component. Another alternative could be to focus on intervention efforts that are aligned with strategies similar to the Adolescent Community Reinforcement Approach (A-CRA; Meyers, Roozen, & Smith, 2011), promoting active participation in alternative, pleasant social activities to the extent to which it becomes rewarding and thereby competes with substance use (Meyers et al., 2011). This prevention strategy may have especially important implications for youth from low SES backgrounds (Andrabi, Khoddam, & Leventhal, 2017), as previous research has found that youth with low parental education may receive less reinforcement from engaging in pleasurable activities (Andrabi et al., 2017).

Although this study provides useful insight, the findings should be interpreted in the context of the study’s limitations. First, all data were collected in the Northeastern United States, a geographic area that is not exceptionally diverse, thus the results may not apply to geographic areas with higher racial/ethnic diversity. However, it may be worth noting that SAMHSA finds that this area of the country leads in reports of underage drinking and marijuana use (SAMHSA, 2014), highlighting the importance of this work among this population. Additionally, the general nature of self-report data, particularly regarding sensitive topics such as substance use, can be considered a limitation. However, while completing surveys, study participants were assured that their reports were private and confidential. Additionally, while the ARSS has been found to be psychometrically sound, the psychometric properties of the adapted version of the ARSS that was used in the present study have not been examined; future research should examine these properties. Finally, while we included parental monitoring as a relevant covariate, it is important to consider the role of parental involvement more broadly in these associations. For instance, it may be that some structured activities are available in an adolescent’s community, but that they require parents to have the availability to pay for the activity, to provide transportation to and from the activities, or to be involved in some other way. Parent’s ability to be engaged in this way is quite likely moderated by sociodemographic factors such as SES. As such, future studies should include these important factors in their design to more fully parse out these associations. Despite these limitations, the present study provides important knowledge regarding the role of availability of and engagement in structured and unstructured activities in young adolescent alcohol and marijuana use.

In conclusion, our findings suggest that both having substances available and that perceiving a lack of opportunities to be involved in structured alternative activities can have significant impacts on adolescents’ decisions to use substances within the subsequent three years. Moreover, engagement in unstructured activities also serves as a risk factor for subsequent substance use among this sample. Thus, it can be helpful to have more organized activities to be available for youth, or to change the composition of unstructured activities to include adult monitoring and enhance prosocial behaviors. Since parents play a protective role and peers present a risk for substance use, it is important that prevention efforts have components that include these factors. A deeper understanding of the role of social influence in adolescent substance use in the context of these alternate reinforcers is needed. Future research should also seek to better understand the underlying mechanisms behind the protective role of structured activities in particular for prevention among adolescents.

Acknowledgments

This work was supported by the National Institute on Drug Abuse (NIAAA) grant (R01AA 016838).

Footnotes

1

To examine the effect of activity availability and engagement overall (combining both structured and unstructured activities), we ran additional logistic regression analyses including a total availability and engagement score. These logistic regression models, adjusted for covariates, revealed that total engagement was not significantly associated with either heavy drinking (b = .04, SE = .03, p = .11, OR = 1.04, 95%CI [0.99, 1.10] or marijuana use (b = .02, SE = .02, p = .26, OR = 1.02, 95%CI [0.98, 1.07]). Total availability was not significantly associated with marijuana use (b = .02, SE = .03, p = .49, OR = 1.02, 95%CI [0.97, 1.07]). On the other hand, the total availability score was significantly associated with heavy drinking (b = .09, SE = .03, p = .008, OR = 1.09, 95%CI [1.02, .17]).

References

  1. Substance Abuse and Mental Health Administration. (2014). Center for behavioral health statistics and quality. In The NSDUH report: substance use and mental health estimates from the 2013 national survey on drug use and health: overview of findings. [PubMed] [Google Scholar]
  2. Andrabi N, Khoddam R, & Leventhal AM (2017). Socioeconomic disparities in adolescent substance use: Role of enjoyable alternative substance-free activities. Social Science & Medicine, 176, 175–182. doi: 10.1016/j.socscimed.2016.12.032 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Arthur MW, Hawkins JD, Pollard JA, Catalano RF, & Baglioni A Jr (2002). Measuring risk and protective factors for use, delinquency, and other adolescent problem behaviors: The Communities That Care Youth Survey. Evaluation Review, 26(6), 575–601. [DOI] [PubMed] [Google Scholar]
  4. Audrain-McGovern J, Rodriguez D, Tercyak KP, Epstein LH, Goldman P, & Wileyto EP (2004). Applying a behavioral economic framework to understanding adolescent smoking. Psychology of Addictive Behaviors, 18(1), 64. [DOI] [PubMed] [Google Scholar]
  5. Bartko WT, & Eccles JS (2003). Adolescent participation in structured and unstructured activities: A person-oriented analysis. Journal of Youth and Adolescence, 32(4), 233–241. [Google Scholar]
  6. Bauman A, & Phongsavan P (1999). Epidemiology of substance use in adolescence: prevalence, trends and policy implications. Drug & Alcohol Dependence, 55(3), 187–207. [DOI] [PubMed] [Google Scholar]
  7. Bickel WK, & Vuchinich RE (2000). Reframing health behavior change with behavioral economics: Psychology Press. [Google Scholar]
  8. Broman CL (2016). The availability of substances in adolescence: influences in emerging adulthood. Journal of Child & Adolescent Substance Abuse, 25(5), 487–495. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Coleman D (1993). Leisure based social support, leisure dispositions and health. Journal of Leisure Research, 25(4), 350. [Google Scholar]
  10. Correia CJ, Murphy JG, Irons JG, & Vasi AE (2010). The behavioral economics of substance use: Research on the relationship between substance use and alternative reinforcers. Journal of Behavioral Health and Medicine, 1(3), 216. [Google Scholar]
  11. Correia CJ, Simons J, Carey KB, & Borsari BE (1998). Predicting drug use: Application of behavioral theories of choice. Addictive Behaviors, 23(5), 705–709. [PubMed] [Google Scholar]
  12. Donovan JE (2009). Estimated blood alcohol concentrations for child and adolescent drinking and their implications for screening instruments. Pediatrics, 123(6), e975–e981. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Eccles JS, & Barber BL (1999). Student council, volunteering, basketball, or marching band: What kind of extracurricular involvement matters? Journal of Adolescent Research, 14(1), 10–43. [Google Scholar]
  14. Forman–Hoffman VL, Batts KR, Hedden SL, Spagnola K, & Bose J (2018). Comorbid mental disorders among adults in the mental health surveillance survey. Annals of Epidemiology. [DOI] [PubMed] [Google Scholar]
  15. Grant JD, Scherrer JF, Lynskey MT, Lyons MJ, Eisen SA, Tsuang MT, … Bucholz KK (2006). Adolescent alcohol use is a risk factor for adult alcohol and drug dependence: evidence from a twin design. Psychological Medicine, 36(1), 109–118. [DOI] [PubMed] [Google Scholar]
  16. Green KM, Musci RJ, Johnson RM, Matson PA, Reboussin BA, & Ialongo NS (2016). Outcomes associated with adolescent marijuana and alcohol use among urban young adults: A prospective study. Addictive Behaviors, 53, 155–160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Griffin KW, & Botvin GJ (2010). Evidence-Based Interventions for Preventing Substance Use Disorders in Adolescents. Child and Adolescent Psychiatric Clinics of North America, 19(3), 505–526. doi: 10.1016/j.chc.2010.03.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hall WD, Patton G, Stockings E, Weier M, Lynskey M, Morley KI, & Degenhardt L (2016). Why young people’s substance use matters for global health. The Lancet Psychiatry, 3(3), 265–279. [DOI] [PubMed] [Google Scholar]
  19. Harrison PA, Fulkerson JA, & Park E (2000). The relative importance of social versus commercial sources in youth access to tobacco, alcohol, and other drugs. Preventive Medicine, 31(1), 39–48. [DOI] [PubMed] [Google Scholar]
  20. Hill KG, White HR, Chung IJ, Hawkins JD, & Catalano RF (2000). Early adult outcomes of adolescent binge drinking: person-and variable-centered analyses of binge drinking trajectories. Alcoholism: Clinical and Experimental Research, 24(6), 892–901. [PMC free article] [PubMed] [Google Scholar]
  21. Hoeben EM, Meldrum RC, Walker D, & Young JTN (2016). The role of peer delinquency and unstructured socializing in explaining delinquency and substance use” A state-of-the-art review. Journal of Criminal Justice, 47, 108–122. [Google Scholar]
  22. Holmes GR, Heckel RV, Chestnut E, Harris N, & Cautela J (1987). Factor analysis of the Adolescent Reinforcement Survey Schedule (ARSS) with college freshmen. Journal of Clinical Psychology, 43(4), 386–390. [DOI] [PubMed] [Google Scholar]
  23. Jackson KM, Barnett NP, Colby SM, & Rogers ML (2015). The prospective association between sipping alcohol by the sixth grade and later substance use. Journal of Studies on Alcohol and Drugs, 76(2), 212–221. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Jackson KM, Colby SM, Barnett NP, & Abar CC (2015). Prevalence and correlates of sipping alcohol in a prospective middle school sample. Psychology of Addictive Behaviors, 29(3), 766. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Johnston LD, Miech RA, O’Malley PM, Bachman JG, Schulenberg JE, & Patrick ME (2019). Monitoring the Future National Survey Results on Drug Use, 1975–2018: Overview, Key Findings on Adolescent Drug Use. Institute for Social Research. [Google Scholar]
  26. Kahler CW, Hoeppner BB, & Jackson KM (2009). A Rasch model analysis of alcohol consumption and problems across adolescence and young adulthood. Alcoholism: Clinical and Experimental Research, 33(4), 663–673. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kerr M, & Stattin H (2000). What parents know, how they know it, and several forms of adolescent adjustment: further support for a reinterpretation of monitoring. Developmental Psychology, 36(3), 366. [PubMed] [Google Scholar]
  28. Kloos A, Weller RA, Chan R, & Weller EB (2009). Gender differences in adolescent substance abuse. Current Psychiatry Reports, 11(2), 120–126. [DOI] [PubMed] [Google Scholar]
  29. Lee J (2012). The influence of parent–adolescent relationship quality, parental monitoring, and peer substance use on substance use in South Korean high school students: Using latent growth curve modelling. Health Education Journal, 71(4), 505–515. [Google Scholar]
  30. Mahoney JL, & Stattin H (2000). Leisure activities and adolescent antisocial behavior: The role of structure and social context. Journal of Adolescence, 23(2), 113–127. [DOI] [PubMed] [Google Scholar]
  31. Mason MJ, Zaharakis NM, Rusby JC, Westling E, Light JM, Mennis J, & Flay BR (2017). A longitudinal study predicting adolescent tobacco, alcohol, and cannabis use by behavioral characteristics of close friends. Psychology of Addictive Behaviors, 31(6), 712. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Meyers RJ, Roozen HG, & Smith JE (2011). The community reinforcement approach: an update of the evidence. Alcohol Research & Health, 33(4), 380. [PMC free article] [PubMed] [Google Scholar]
  33. Moilanen KL, Markstrom CA, & Jones E (2014). Extracurricular activity availability and participation and substance use among American Indian adolescents. Journal of Youth and Adolescence, 43(3), 454–469. [DOI] [PubMed] [Google Scholar]
  34. Monahan KC, Rhew IC, Hawkins JD, & Brown EC (2014). Adolescent pathways to co-occurring problem behavior: The effects of peer delinquency and peer substance use. Journal of Research on Adolescence, 24(4), 630–645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Murphy JG, Correia CJ, & Barnett NP (2007). Behavioral economic approaches to reduce college student drinking. Addictive Behaviors, 32(11), 2573–2585. [DOI] [PubMed] [Google Scholar]
  36. National Institute on Drug Abuse. (2020). Monitoring the Future 2019 Survey Results: Overall Findings. Retrieved from https://www.drugabuse.gov/related-topics/trends-statistics/infographics/monitoring-future-2019-survey-results-overall-findings
  37. Nesselroade JR, & Baltes PB (1979). Longitudinal research in the study of behavior and development: Academic Press; New York. [Google Scholar]
  38. Nicholson LM, Slater SJ, Chriqui JF, & Chaloupka F (2014). Validating adolescent socioeconomic status: Comparing school free or reduced price lunch with community measures. Spatial Demography, 2(1), 55–65. [Google Scholar]
  39. Palen L-A, & Coatsworth JD (2007). Activity-based identity experiences and their relations to problem behavior and psychological well-being in adolescence. Journal of Adolescence, 30(5), 721–737. [DOI] [PubMed] [Google Scholar]
  40. Palen L-A, Smith EA, Flisher AJ, Caldwell LL, & Mpofu E (2006). Substance use and sexual risk behavior among South African eighth grade students. Journal of Adolescent Health, 39(5), 761–763. [DOI] [PubMed] [Google Scholar]
  41. Prins RG, Oenema A, van der Horst K, & Brug J (2009). Objective and perceived availability of physical activity opportunities: differences in associations with physical activity behavior among urban adolescents. International Journal of Behavioral Nutrition and Physical Activity, 6(1), 70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Rodgers-Farmer AY (2001). Parental monitoring and peer group association: Their influence on adolescent substance use. Journal of Social Service Research, 27(2), 1–18. [Google Scholar]
  43. Substance Abuse and Mental Health Services Administration. (2016). Results from the 2015 survey on drug use and health: Detailed tables. Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health. Retrieved from https://www.samhsa.gov/data/sites/default/files/cbhsq-reports/NSDUHDetailedTabs2018R2/NSDUHDetTabsSect1pe2018.htm [Google Scholar]
  44. Seek Moon S, & Rao U (2011). Social Activity, School-related Activity, and Anti–substance Use: Media Messages on Adolescent Tobacco and Alcohol Use. Journal of Human Behavior in the Social Environment, 21(5), 475–489. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Sharp EH, Tucker CJ, Baril ME, Van Gundy KT, & Rebellon CJ (2015). Breadth of participation in organized and unstructured leisure activities over time and rural adolescents’ functioning. Journal of Youth and Adolescence, 44(1), 62–76. [DOI] [PubMed] [Google Scholar]
  46. Terry-McElrath YM, O’Malley PM, & Johnston LD (2011). Exercise and substance use among American youth, 1991–2009. American Journal of Preventive Medicine, 40(5), 530–540. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Townsend L, Flisher AJ, & King G (2007). A systematic review of the relationship between high school dropout and substance use. Clinical Child and Family Psychology Review, 10(4), 295–317. [DOI] [PubMed] [Google Scholar]
  48. Van Ryzin MJ, Fosco GM, & Dishion TJ (2012). Family and peer predictors of substance use from early adolescence to early adulthood: An 11-year prospective analysis. Addictive Behaviors, 37(12), 1314–1324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Vuchinich RE, & Tucker JA (1988a). Contributions from behavioral theories of choice to an analysis of alcohol abuse. Journal of Abnormal Psychology, 97(2), 181. [DOI] [PubMed] [Google Scholar]
  50. Vuchinich RE, & Tucker JA (1988b). Contributions from behavioral theories of choice to an analysis of alcohol abuse. Journal of Abnormal Psychology, 97(2), 181–195. [DOI] [PubMed] [Google Scholar]
  51. World Health Organization. (2014). Global status report on alcohol and health 2014. Global Status Report on Alcohol. 1–392. [Google Scholar]

RESOURCES