Skip to main content
NCBI home page
Journal of Studies on Alcohol and Drugs logoLink to Journal of Studies on Alcohol and Drugs
. 2019 Dec 3;80(6):669–678. doi: 10.15288/jsad.2019.80.669

Close Friends’ Drinking and Personal Income as Mediators of Extreme Drinking: A Prospective Investigation

Jeremy W Luk a,b,*, Denise L Haynie a, Federico E Vaca c, Kaigang Li d, Ralph Hingson e, Bruce G Simons-Morton a
PMCID: PMC6900994  PMID: 31790357

Abstract

Objective:

This study examined longitudinal associations between college attendance, residence on- or off-campus, and work status during the first 2 years after high school with extreme binge drinking at 4 years after high school and tested peer drinking and personal income at 3 years after high school as mediators.

Method:

Data were drawn from Waves 4–7 of the NEXT Generation Health Study (n = 2,081). Multinomial logistic regressions and mediation analyses were conducted. Extreme binge drinking was measured using the largest number of drinks on a single day in the past year.

Results:

Univariate analyses indicated that attending university, living on campus, and working more than 30 hours at any point during the first 2 years after high school were associated with increased risk of drinking two to three times above the binge drinking threshold (relative risk ratios [RRR] ranged from 1.79 to 5.70). In multivariate analyses, dropping out of university was associated with drinking two times above the binge drinking threshold (RRR = 4.88), whereas living on campus (RRR = 4.54) and working more than 30 hours (RRR = 5.26) were associated with increased risk of drinking three times above the binge drinking threshold. Close friends’ drinking and personal income were significant mediators.

Conclusions:

Living on campus and working more than 30 hours per week during the first 2 years after high school increased risk for drinking three times above the binge drinking threshold at 4 years after high school.

Transitioning out of high school is often marked by major changes in school, residence, and work status (e.g., attending college, moving out of parent’s home, first time employment). These transitions may expose young adults to more opportunities to drink (Arnett, 2005; Carter et al., 2010; Schulenberg & Maggs, 2002). Although studies have documented associations between school, residence, and work status with alcohol use, binge drinking, or alcohol-related consequences (Lee et al., 2018; Simons-Morton et al., 2016; White et al., 2006), research focusing on extreme binge drinking as the primary outcome is relatively limited, leaving a significant knowledge gap. This is of particular importance as recent research shows that drinking beyond the binge drinking threshold is common among young adults in the United States (Patrick & Azar, 2017; Patrick et al., 2016).

Data from the Monitoring the Future survey indicate that 6.0% of 12th graders reported 10 or more drinks in a row in the last 2 weeks, and 3.1% reported 15 or more drinks in the last 2 weeks (Johnston et al., 2018). These statistics are concerning as extreme binge drinking is associated with poor school grades, increased substance use, driving while impaired, risky sexual behaviors, physical fights, injuries, emergency department visits, and arrests/detentions (Hingson & Zha, 2018; Hingson et al., 2017). Given these negative consequences, the current study examined which aspects of the post–high school environment would longitudinally predict extreme binge drinking (measured as the largest number of drinks on a single day in the past year) and tested close friends’ drinking and personal income as mediators.

School, residence, and work status and alcohol-related outcomes

In 2017, 66.7% of recent high school graduates were enrolled in colleges or universities (Bureau of Labor Statistics, 2018). Early studies show that college students are at increased risk for alcohol use (Crowley, 1991; Dawson et al., 2004). However, these findings are difficult to interpret because of different conceptualizations and definitions of college and noncollege status across studies (Carter et al., 2010). Recent research suggests that community college students may be an at-risk group for drinking that is distinguished from students at 4-year universities (Sheffield et al., 2005; Velazquez et al., 2011; Wall et al., 2012). For instance, Linden-Carmichael and Lanza (2018) found that college students were more likely than noncollege peers to be members of the latent classes of frequent drinkers with occasional bingeing and high-intensity drinkers.

Similarly, existing research on residence and alcohol-related outcomes is limited as prior studies typically used college samples and varied definitions of residence status (Benz et al., 2017; DiBello et al., 2018; Harford et al., 2002). Although living away from parent’s home is a risk factor for increased drinking behaviors (Gfroerer et al., 1997; Simons-Morton et al., 2016; White et al., 2006), it is unclear if living on campus or off campus would pose the greater risk for problematic drinking (DiBello et al., 2018; Harford et al., 2002) and whether these associations would extend to predicting high-intensity drinking. Using data from the Monitoring the Future study, Patrick and Terry-McElrath (2017) found that prevalence of high-intensity drinking was higher among youth who attended 4-year colleges and did not reside with their parents. Replication and extension of research on residence status and extreme binge drinking using other nationally representative and gender-specific criteria for extreme binge drinking would be valuable.

In terms of work status, part-time work has been associated with increased alcohol use, binge drinking, and marijuana use in adolescent samples (Breslin & Adlaf, 2005; Leeman et al., 2014). Longitudinal research indicated that adolescents with higher work intensity were more likely to engage in heavy drinking a year later (Paschall et al., 2004) and were more likely to experience alcohol-related negative consequences (Osilla et al., 2013). Although less frequently studied among young adults, Lee and colleagues (2018) found that young adults who were working full-time consumed more alcohol than peers who were unemployed. Moreover, working full-time was also associated with increased risk of heavy episodic drinking and alcohol-related negative consequences. Despite these findings, it is unclear if the effects of work status on various drinking outcomes would extend to predict extreme binge drinking.

Close friends’ drinking and personal income as mediators

Understanding whether and how school, residence, and work status during the 2 years after high school might longitudinally predict extreme binge drinking at 4 years after high school can aid the identification of at-risk groups and guide targeted prevention. The college environment provides a unique context for emerging adults to make new friends, and prior research has identified peer influence as a robust risk factor for college drinking (Borsari & Carey, 2001). Especially during the first 2 years of college, students may seek to establish new friendships that are maintained during the college years. Affiliation with close friends who are drinkers may therefore be an avenue for students to engage in risker drinking behaviors. Accordingly, it is important to examine the extent to which the longitudinal associations between college and residence statuses during the first 2 years after high school with extreme binge drinking at 4 years after high school are mediated by close friends’ drinking.

The experiences of emerging adults who primarily work may differ from those who attend community college or university. Specifically, their peers at work are less likely to be same-age peers and the mechanism through which they are exposed to increased risk for drinking may vary. Prior research has linked higher socioeconomic status to increased drinking behaviors (Casswell et al., 2003; Huckle et al., 2010). Among emerging adults, having increased personal income has been associated with increased risk of heavy drinking (Kar et al., 2018). Although increased personal income itself may not necessarily lead to increased drinking, it might increase access to alcohol and provide potential opportunities for drinking-related social activities that could heighten the likelihood of extreme binge drinking.

Current study

Using Wave 1 (10th grade) to Wave 4 (first year after high school) of the NEXT Generation Health Study, Simons-Morton and colleagues (2016) found that living on campus was associated with increased odds of alcohol use and heavy episodic drinking during the first year after high school. The current study examined the subsequent 4-year time frame from the first year after high school (Wave 4) to 4 years after high school (Wave 7), focusing on modeling extreme binge drinking as the primary outcome and testing mediation effects. First, we examined whether school, residence, and work status during the first 2 years after high school were longitudinally associated with extreme binge drinking at 4 years after high school. Second, we tested the specific aspects of post–high school environments that are independently associated with extreme binge drinking. Last, we evaluated close friends’ drinking and personal income as potential mediators of the associations between post–high school environments and extreme binge drinking.

Method

Sample

Data were drawn from Wave 4 to Wave 7 of the NEXT Generation Health Study, a 7-year longitudinal study of adolescents in the United States (Conway et al., 2013). Using a three-stage stratified design, a nationally representative sample of 2,785 adolescents enrolled in U.S. high schools from 22 states was obtained. The retention rates were 78.2% (n = 2,177) for Wave 4, 79.1% (n = 2,202) for Wave 5, 82.8% (n = 2,306) for Wave 6, and 83.4% (n = 2,323) for Wave 7. The analytic sample included 2,081 participants (74.7% of the full sample) who did not have any missing data on the covariates, predictors, and outcomes of interest. Parents provided written consent, and participants provided assent to participate in this study. Participants provided consent when they turned 18 years old. The annual follow-up assessments were completed online. This study was approved by the institutional review board of the Eunice Kennedy Shriver National Institute of Child Health and Human Development.

Measures

Post–high school environment (Waves 4 and 5).

Participants were asked whether they were currently attending school, where they currently lived, and how many total hours (on average) they spent in paid and/or unpaid jobs. To minimize missing data and capture exposure to post–high school environments that may enhance the risk of extreme drinking, we used data from the first 2 years after high school (Waves 4 and 5) and coded for the highest risk environment.

For school status, participants who attended college/university at Waves 4 and/or 5 were coded as “university.” For participants who did not attend college/university, those who attended technical/community colleges at Waves 4 and/or 5 were coded as “technical/community colleges.” All remaining participants were coded as “not attending school.”

For residence status, participants who lived on campus at Waves 4 and/or 5 were coded as “living on campus.” For participants who did not live on campus, those who lived in their own place at Waves 4 and/or 5 were coded as “living in their own place.” All remaining participants were coded as “living in parent’s home.” Among those who lived on campus, 90.5% had a roommate, whereas 49.3% of those who lived in their own place and 4.9% of those who lived at home with parents had a roommate.

For work status, we used an adopted item from Monitoring the Future with response options coded as 0 = not working; 1 = 1 to 20 hours per week; 2 = 21 to 30 hours per week; and 3 = more than 30 hours per week (Bachman et al., 1993). For participants who responded to both Waves, we took the higher number of work hours.

To capture school transitions, we created two variables using data from Waves 4–7. First, dropout of university was defined as having attended university at either one or both years after high school but did not attend any school (including both university and community college) during third and fourth years after high school. Second, late entry to university was defined as not being in university during the first 2 years after high school but being in university at 4 years after high school.

Close friends’ drinking (Wave 6).

Participants were asked to think of their five closest friends whom they spend time with and report their perception of how often these friends drink alcohol. Response options were on a 5-point scale, ranging from “never” to “almost always.”

Personal income (Wave 6).

Participants were asked: “What is your best guess of your personal earnings before taxes, for the past year?” Twelve response options ranged from “I had no personal earnings last year” and “less than $2,500” to “$75,000 to $99,999.” Because of less frequent endorsement of personal income response options beyond $15,000 (Supplemental Table A), this item was recoded for analysis. (Supplemental material appears as an online-only addendum to the article on the journal’s website.) Following a recent study (Kar et al., 2018), five ordered categories with approximately $5,000 increments were created: 0 = no personal income; 1 = up to $5,000; 2 = $5,000–$9,999; 3 = $10,000–$14,999; and 4 = ≥$15,000.

Extreme binge drinking (Wave 7).

Participants were asked the question: “During the past 12 months, what was the largest number of drinks that you drank on a single day?” Referencing a recent study (Hingson et al., 2017), we used the gender-specific binge drinking criteria and created five groups: 0 = abstain; 1 = non–binge drinking (1–4 drinks for males and 1–3 drinks for females); 2 = binge drinking (5–9 drinks for males and 4–7 drinks for females); 3 = two times the binge drinking threshold (10–14 drinks for males and 8–11 drinks for females); and 4 = three times the binge drinking threshold (15 or more drinks for males and 12 or more drinks for females). We dichotomized this variable into drinking two to three times the binge drinking threshold versus all other responses for the primary mediation analyses.

Covariates.

Gender, race/ethnicity, family affluence, and binge drinking (yes/no) during the first 2 years after high school were included as covariates. Family affluence was measured by the Health Behaviour in School-Aged Children (HBSC) Family Affluence Scale, with items inquiring about participants’ family car and computer ownership, whether they had their own bedroom, and frequency of family holidays (Currie et al., 2008). Scores on the Family Affluence Scale were coded into 0–4 = low; 5–6 = medium; and 7 = high. Binge drinking during the first 2 years after high school was a binary variable reflecting whether participants engaged in binge drinking (5 or more drinks for males and 4 or more drinks for females) within 2 hours in the past 30 days at any point during the first 2 years after high school.

Analysis

Statistical analyses followed three steps. First, unadjusted relative risk ratios (RRRs) were obtained using three univariate multinomial logistic regressions to examine the longitudinal associations between (a) school, (b) residence, and (c) work status during the first 2 years after high school with extreme binge drinking at 4 years after high school. The “abstain” group was set as the base outcome in planned analyses, and the “non–binge drinking” group was set as the base outcome in supplemental analyses. Second, a multivariate multinomial logistic regression was conducted to capture the independent influences of all post–high school environment variables (school status, dropping out, late entry into college, residence status, and work status) on extreme drinking while covariates were controlled for (gender, race/ethnicity, family affluence, and baseline binge drinking). The multinomial logistic regressions were conducted in STATA 14 (StataCorp LP, College Station, TX) and used the “svy” procedure to account for the complex survey design of the NEXT study. Third, mediation analyses were conducted within a path analysis framework, with close friends’ drinking and personal income at 3 years after high school treated as potential mediators. Mediation analyses were conducted in Mplus 8 (Muthén & Muthén, 1998–2017) using the product of coefficient approach, modeling extreme binge drinking as a binary variable (drinking two to three times the binge threshold versus all other responses) in the planned analyses and as an ordered categorical variable with five categories in supplemental analyses. Categorical outcomes were modeled using the robust weighted least squares estimator. Bias-corrected indirect effects and their 95% confidence intervals were obtained via bootstrapping (with 20,000 resamples). All analyses accounted for the clustering and stratification in the NEXT study design, with weights applied to adjust for attrition bias and retain the representativeness of the national sample.

Results

Descriptive statistics of the study sample and variables are presented in Table 1. Of all participants, 45.7% were in university at any point during the first 2 years after high school, whereas 30.4% were in technical school/community college and 23.9% were not attending school during the same 2-year period. Less than half (45.1%) of participants lived in parent’s home during the first 2 years after high school, whereas 25.8% and 29.1% of participants, respectively, lived in their own place or on campus during the same 2-year period. More than 70% of participants worked at any point during the first 2 years after high school, with 29.2% reported working more than 30 hours per week. Cross tabulations of school status with residence and work statuses during the 2 years after high school are presented in Supplemental Table B. Binge drinking beyond the typical threshold at 4 years after high school was common, with 15.7% drinking two times above the binge drinking threshold and 12.3% drinking three times above the binge drinking threshold.

Table 1.

Descriptive statistics of the study sample and variables (n = 2,081)

graphic file with name jsad.2019.80.669tbl1.jpg

Variable Label (range) Frequency (M) Weighted percentage (SE)
Gender Female 1,247 60.9%
Race/ethnicity Non-Hispanic White 809 56.7%
Non-Hispanic African American 551 20.2%
Hispanic 619 19.1%
Other 102 4.1%
Family affluencea Low 672 24.4%
Medium 964 47.3%
High 445 28.3%
School status (Waves 4–5) Not attending school 459 23.9%
Technical school/community college 621 30.4%
University 1,001 45.7%
Residence status (Waves 4–5) Parent’s home 1,155 45.1%
Own place 391 25.8%
On campus 535 29.1%
Work Status (Waves 4–5) Not working 705 28.7%
20 hours or less per week 466 23.8%
21–30 hours per week 402 18.3%
>30 hours per week 508 29.2%
Binge drinking (Waves 4–5) Yes 762 41.5%
Close friends’ drinking (Wave 6)b (0–4) (1.32) (0.07)
Personal income (Wave 6) No earning 320 10.0%
$1–$4,999 726 38.5%
$5,000–$9,999 317 15.6%
$10,000–$14,999 271 15.5%
≥$15,000 352 20.5%
Extreme binge drinking (Wave 7) Abstain 439 17.7%
Non–binge drinking 517 20.0%
Binge drinking 651 34.3%
Two times binge threshold 290 15.7%
Three times binge threshold 184 12.3%
Open in a new tab

Notes: Weighted percentages reflect nationally representative estimates accounting for oversampling of African Americans.

a

Scores on the Family Affluence Scale were coded into three categories: 0–4 = low; 5–6 = medium; 7 = high.

b

Close friends’ drinking reflects participants’ perception of how often their five closest friends drink alcohol, with response options ranging from 0 = never to 4 = almost always.

Table 2 presents percentages of levels of extreme drinking at 4 years after high school by school, residence, work status, and two school transition patterns (dropping out of university and late entry into university). Table 3 presents unadjusted RRRs from three univariate multinomial logistic regressions. University attendance at any point during the first 2 years after high school was associated with increased risk of all levels of drinking 4 years after high school, including drinking two times (RRR = 1.79) and three times (RRR = 3.02) the binge drinking threshold. Living on campus at any point during the first 2 years after high school was associated with increased risk of all levels of drinking 4 years after high school, including drinking two times (RRR = 3.38) and three times (RRR = 5.70) the binge drinking threshold. Working 1–20 hours per week at any point during the first 2 years after high school was associated with increased risk of non–binge drinking, binge drinking, and drinking two times above the binge threshold (RRRs ranged from 1.92 to 2.54). Working 21–30 hours per week at any point during the first 2 years after high school was associated with increased risk of drinking two times above the binge threshold (RRR = 2.22), whereas working more than 30 hours per week at any point during the first 2 years after high school was associated with increased risk of drinking two to three times above the binge threshold (RRRs ranged from 3.04 to 3.39). Results with non–binge drinking set as the base outcome are presented in Supplemental Table C.

Table 2.

Percentages of levels of extreme drinking 4 years after high school by post–high school environment during the 2 years after high school, dropout from university, and late-entry to university

graphic file with name jsad.2019.80.669tbl2.jpg

Post–high school environment Weighted column percentages
Abstain Non–binge drinking Binge drinking Two times binge threshold Three times binge threshold
School
 Not attending school 36.5% 23.2% 17.9% 27.6% 18.8%
 Technical school/community college 29.5% 30.2% 33.7% 26.3% 28.1%
 University 34.1% 46.6% 48.4% 46.1% 53.1%
Residence
 Parent’s home 56.5% 54.0% 42.4% 38.8% 30.0%
 Own place 30.2% 18.5% 24.3% 30.3% 29.7%
 On campus 13.3% 27.6% 33.3% 30.9% 40.3%
Work status
 Not working 40.0% 27.8% 30.3% 20.0% 20.8%
 1–20 hours per week 17.9% 28.0% 26.1% 22.8% 20.1%
 21–30 hours per week 17.1% 22.3% 17.4% 19.0% 14.9%
 >30 hours per week 25.0% 21.9% 26.3% 38.2% 44.2%
School transitions
 Dropout of university 3.5% 4.5% 4.6% 9.4% 3.3%
 Late entry to university 5.3% 9.2% 6.3% 6.5% 5.2%
Open in a new tab

Notes: Post–high school environment was measured during the first 2 years after high school. Dropout of university was defined as having attended university at either 1 or both years after high school but did not attend any school during 3–4 years after high school. Late entry to university was defined as not being in university during the first 2 years after high school but being in university at 4 years after high school.

Table 3.

Unadjusted relative risk ratios from three separate multinomial logistic regression models predicting drinking levels 4 years after high school

graphic file with name jsad.2019.80.669tbl3.jpg

Post–high school environment Unadjusted relative risk ratios [95% CI]
Abstain Non–binge drinking Binge drinking Two times binge threshold Three times binge threshold
Model 1: School status
 Not attending school Base outcome Referent Referent Referent Referent
 Technical school
community college Base outcome 1.61 [0.78, 3.31] 2.33 [0.90, 6.02] 1.18 [0.60, 2.32] 1.85 [0.82, 4.16]
 University Base outcome 2.15 [1.24, 3.72] 2.90 [1.54, 5.45] 1.79 [1.01, 3.17] 3.02 [1.43, 6.37]
Model 2: Residence status
 Parent’s home Base outcome Referent Referent Referent Referent
 Own place Base outcome 0.64 [0.32, 1.29] 1.07 [0.56, 2.06] 1.46 [0.75, 2.86] 1.85 [0.78, 4.40]
 On campus Base outcome 2.17 [1.13, 4.17] 3.33 [1.76, 6.28] 3.38 [1.86, 6.15] 5.70 [3.37, 9.66]
Model 3: Work status
 Not working Base outcome Referent Referent Referent Referent
 1–20 hours per week Base outcome 2.25 [1.20, 4.21] 1.92 [1.03, 3.60] 2.54 [1.50, 4.30] 2.15 [0.93, 4.95]
 21–30 hours per week Base outcome 1.88 [0.96, 3.66] 1.34 [0.71, 2.53] 2.22 [1.05, 4.67] 1.67 [0.84, 3.32]
 >30 hours per week Base outcome 1.26 [0.82, 1.94] 1.39 [0.64, 2.94] 3.04 [1.62, 5.72] 3.39 [1.90, 6.06]
Open in a new tab

Note: Significant associations are highlighted in bold.

Results from the multivariable multinomial logistic regressions adjusting for covariates are presented in Table 4. Dropping out of university was associated with drinking two times above the binge drinking threshold (RRR = 4.88), whereas living on campus and working 21–30 hours or more than 30 hours were associated with increased risk of drinking two and/or three times above the binge drinking threshold (RRRs ranged from 2.39 to 5.26). When other post–high school environments were controlled for, attending technical school/community college at any point during the first 2 years after high school was associated with increased risk of drinking three times above the binge drinking threshold (RRR = 2.77). Binge drinking during the first 2 years after high school was strongly associated with drinking two to three times above the binge drinking threshold at 4 years after high school (RRRs were 7.48 and 14.61, respectively).

Table 4.

Multivariable multinomial logistic regression models predicting drinking levels 4 years after high school

graphic file with name jsad.2019.80.669tbl4.jpg

Variables Adjusted relative risk ratios [95% CI]
Non–binge drinking Binge drinking Two times binge threshold Three times binge threshold
School status
 Not attending school Referent Referent Referent Referent
 Technical school/ community college 1.34 [0.61, 2.93] 2.40 [0.91, 6.31] 1.43 [0.69, 2.97] 2.77 [1.01, 7.63]
 University 1.36 [0.73, 2.51] 1.65 [0.76, 3.60] 0.98 [0.52, 1.85] 2.39 [0.66, 8.68]
School transitions
 Not dropout Referent Referent Referent Referent
 Dropout of university 1.33 [0.45, 3.87] 1.73 [0.76, 3.94] 4.88 [1.73, 13.74] 1.23 [0.43, 3.55]
 Not late entry Referent Referent Referent Referent
 Late entry to university 1.77 [0.71, 4.42] 1.40 [0.61, 3.20] 1.69 [0.64, 4.45] 1.57 [0.35, 6.99]
Residence
 Parent’s home Referent Referent Referent Referent
 Own place 0.74 [0.39, 1.42] 1.03 [0.51, 2.09] 1.12 [0.50, 2.54] 1.30 [0.54, 3.16]
 On campus 2.14 [0.90, 5.12] 2.94 [1.57, 5.50] 3.29 [1.80, 5.99] 4.54 [1.72, 11.98]
Work status
 Not working Referent Referent Referent Referent
 1–20 hours per week 1.79 [1.02, 3.16] 1.27 [0.61, 2.64] 1.75 [0.96, 3.18] 1.42 [0.58, 3.46]
 21–30 hours per week 2.05 [0.93, 4.54] 1.36 [0.64, 2.88] 2.39 [1.10, 5.19] 2.36 [0.88, 6.36]
 >30 hours per week 1.65 [1.07, 2.54] 1.71 [0.80, 3.62] 3.07 [1.41, 6.69] 5.26 [2.70, 10.23]
Gender
 Male Referent Referent Referent Referent
 Female 1.10 [0.65, 1.87] 1.20 [0.76, 1.87] 0.59 [0.32, 1.06] 0.43 [0.22, 0.83]
Race/ethnicity
 White Referent Referent Referent Referent
 African American 1.30 [0.87, 1.95] 0.41 [0.24, 0.69] 0.19 [0.09, 0.38] 0.09 [0.03, 0.30]
 Hispanic 1.32 [0.76, 2.31] 1.19 [0.45, 3.15] 0.79 [0.27, 2.26] 0.62 [0.24, 1.56]
 Other 2.35 [0.57, 9.71] 0.70 [0.22, 2.29] 1.77 [0.44, 7.18] 1.19 [0.15, 9.48]
Family affluence
 Low Referent Referent Referent Referent
 Medium 0.86 [0.51, 1.44] 1.11 [0.61, 2.01] 1.15 [0.52, 2.52] 1.35 [0.53, 3.47]
 High 1.81 [0.98, 3.32] 1.66 [0.84, 3.28] 2.66 [1.25, 5.65] 1.82 [0.80, 4.15]
Baseline drinking
 Did not binge drink Referent Referent Referent Referent
 Binge drinking 1.16 [0.57, 2.36] 3.33 [1.74, 6.36] 7.48 [3.96, 14.13] 14.61 [6.58, 32.44]
Open in a new tab

Notes: The “abstain” group was set as the base outcome. Significant associations are highlighted in bold.

Results from the mediation model are presented in Figure 1. Close friends’ drinking mediated the longitudinal associations between school attendance, living in own place, and living on campus during the first 2 years after high school with extreme binge drinking at 4 years after high school. Personal income mediated the longitudinal associations between work statuses during the first 2 years after high school with extreme binge drinking at 4 years after high school, as well as the association between dropping out of university and extreme binge drinking. The bootstrapped 95% confidence intervals of the direct and indirect paths are summarized in Table 5. After we controlled for the mediators and covariates, only two direct paths from working more than 30 hours and dropping out of university to extreme binge drinking remained significant. Mediation analyses modeling extreme binge drinking as an ordered categorical variable yielded similar findings (Supplemental Table D and Supplemental Figure A).

Figure 1.

Figure 1.

Open in a new tab

Results from path analysis linking post–high school environment 2 years after high school to extreme binge drinking 4 years after high school through close friends’ drinking and personal income 3 years after high school. Extreme binge drinking was dichotomized into drinking two to three times the binge drinking threshold versus all other responses. Standardized regression coefficients are presented for significant associations. Associations between work status and close friends’ drinking were evaluated and were nonsignificant. Associations between school and residence status with personal income were also evaluated and were nonsignificant.

Table 5.

Direct and indirect associations from post–high school environment to extreme binge drinking 4 years after high school

graphic file with name jsad.2019.80.669tbl5.jpg

Pathways Estimate of direct effect [95% CI] Estimate of indirect effect [95% CI]
Community college → close friends’ drinking → extreme binge drinking -0.022 [-0.086, 0.045] 0.049 [0.024, 0.081]
University → close friends’ drinking → extreme binge drinking -0.060 [-0.197, 0.077] 0.067 [0.033, 0.109]
Lived in own place → close friends’ drinking → extreme binge drinking 0.026 [-0.110, 0.133] 0.014 [0.004, 0.031]
Lived on campus → close friends’ drinking → extreme binge drinking 0.081 [-0.021, 0.200] 0.021 [0.002, 0.042]
Worked 1–20 hours → personal income → extreme binge drinking 0.023 [-0.078, 0.126] 0.018 [0.006, 0.036]
Worked 21–30 hours → personal income → extreme binge drinking 0.061 [-0.001, 0.137] 0.025 [0.009, 0.049]
Worked >30 hours → personal income → extreme binge drinking 0.150 [0.042, 0.263] 0.048 [0.017, 0.088]
Dropout of university → personal income → extreme binge drinking 0.069 [0.004, 0.131] 0.012 [0.004, 0.025]
Open in a new tab

Notes: Bias-corrected bootstrapped 95% confidence intervals are presented. Significant mediation effects are indicated by 95% confidence interval bounds that do not include zero and are highlighted in bold.

Discussion

This study documents the influence of multiple post–high school environments on extreme drinking beyond the binge threshold in a national longitudinal sample. Extending emerging research on school or residence status on high-intensity drinking (Linden-Carmichael & Lanza, 2018; Patrick et al., 2017), we used a definition of extreme binge drinking based on the largest number of drinks consumed in the past year and took into account gender-specific differences. Univariate analyses showed that drinking two to three times above the binge drinking threshold were predicted by college attendance, living on campus, and working 20–30 hours or more than 30 hours at any point during the first 2 years after high school. In multivariate analyses, dropping out of university, living on campus, and working 20–30 hours or more than 30 hours were found to be independent predictors of extreme binge drinking. Importantly, these longitudinal associations were significant even after we controlled for baseline binge drinking. This suggests that targeted prevention solely based on drinking status in the post–high school years may not be sufficient. Given the serious consequences extreme binge drinking behaviors commonly pose (Hingson & Zha, 2018; Hingson et al., 2017), interventions tailored for emerging adults who live on campus or work 20 or more hours per week are needed.

Extending research on drinking behaviors among community college students (Sheffield et al., 2005; Wall et al., 2012), multivariate analyses revealed that students who attended technical school/community colleges at any point during the first 2 years after high school were more likely to drink three times over the binge drinking threshold at 4 years after high school. Similar to the adolescent literature, the multivariate analyses also showed that dropping out of university was associated with increased risk of drinking two times the binge threshold at 4 years after high school, potentially reflecting school disengagement or proneness to deviancy (Henry et al., 2012; Mensch & Kandel, 1988; Townsend et al., 2007). Data from the National Longitudinal Survey of Youth showed that individuals who dropped out of high school experienced higher levels of alcohol-related problems in their mid-thirties (Muthén & Muthén, 2000). The current study provided initial evidence that dropping out of university may have a shorter term but similar effect on extreme binge drinking.

The mediation analyses revealed close friends’ drinking as a mechanism linking college/residence status and extreme binge drinking in this longitudinal sample. Although a robust literature exists on peer influences on college drinking (Borsari & Carey, 2001), the focus on close friends’ drinking may capture more proximal peer context than perceived drinking by a “typical” peer or college student. Presumably, the college and campus environments provided a platform for students to establish close friendships that are maintained over time, such that peer drinking behaviors reinforce their own drinking behaviors, leading to effects lasting through 4 years after high school. These findings underscore the importance of prevention and intervention efforts focusing on the peer context among college and university students.

Past research on work status and drinking was primarily found in adolescent samples. The current study extends recent findings by Lee and colleagues (2018) by including past-year extreme binge drinking on the day of highest consumption as the primary outcome. It is notable that in the multivariate analyses, working 21–30 hours and more than 30 hours were associated with increased risk of drinking two and/or three times the binge drinking threshold, suggesting a need to develop or adapt interventions targeting working emerging young adults. Furthermore, we identified personal income as a mediator of the associations between all levels of work and extreme binge drinking. Although personal income itself may not directly lead to increase drinking, it could provide easier access to alcohol, possibly enabling drinking-related social activities and encounters. Accordingly, the use of behavioral economic approaches to reduce drinking among working young adults may be warranted (Murphy et al., 2007). Brief interventions that use a motivational approach to encourage alternative use of income other than purchasing alcohol may be considered (Martens et al., 2004, 2007). Policy implications such as taxation on alcohol and minimum pricing should also be explored as additional approaches to discourage at-risk emerging adults from buying alcohol.

This study has several limitations. First, post–high school environments were measured using collapsed data from the first 2 years after high school. Future research using more intensive assessments may overcome the missing data issues encountered in this study and could examine the impact of microtransitions in school, residence, and work statuses on extreme binge drinking. Second, the extreme drinking outcome measure in this study was only assessed at Wave 7. Thus, examination of changes in extreme drinking frequency by time-varying post–high school environments was not feasible. Third, personal income was determined based on personal earning; accordingly, other sources of income (e.g., money from parents) were not examined. Fourth, although participants were weighted as part of the complex survey design procedures to restore the national representativeness of the sample, potential selection bias may still exist. Last, we did not collect information about the nature of work (e.g., being a waiter/waitress at a restaurant that serves alcohol) and did not assess for potential drinking-related social activities (e.g., frequency of work happy hours involving alcohol). Future research should consider these variables in relation to extreme binge drinking.

Despite these limitations, the current study shows that on-campus housing and the work environment during the first 2 years after high school may be important places to promote alcohol-related harm-reduction strategies. The identification of close friends’ drinking and personal income as mediators provides further insights into prevention efforts, highlighting the relevance of early screening and education on campus housing and work-related settings. Additional research is needed to examine how alcohol-related social experiences in the workplace may elevate risk for extreme drinking and determine the optimal intervention strategies to minimize risks of extreme drinking among working emerging adults.

Footnotes

This project (contract HHSN275201200001I) was supported in part by the Intramural Research Program of the Eunice Kennedy Shriver National Institute of Child Health and Human Development; the National Heart, Lung, and Blood Institute; the National Institute on Alcohol Abuse and Alcoholism; the National Institute on Drug Abuse; and the Maternal and Child Health Bureau of the Health Resources and Services Administration. The opinions and assertions expressed herein are those of the authors and do not necessarily reflect the official policy or position of the Uniformed Services University or the Department of Defense.

References

  1. Arnett J. J. The developmental context of substance use in emerging adulthood. Journal of Drug Issues. 2005;35:235–254. doi:10.1177/002204260503500202. [Google Scholar]
  2. Bachman J. G., Johnston L., O’Malley P. M. Monitoring the Future: A continuing study of the lifestyles and values of youth, 1992. Ann Arbor, MI: Inter-university Consortium for Political and Social Research; 1993. [Google Scholar]
  3. Benz M. B., DiBello A. M., Balestrieri S. G., Miller M. B., Merrill J. E., Lowery A. D., Carey K. B. Off-campus residence as a risk factor for heavy drinking among college students. Substance Use & Misuse. 2017;52:1236–1241. doi: 10.1080/10826084.2017.1298620. doi:10.1080/10826084.2017.1298620. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Borsari B., Carey K. B. Peer influences on college drinking: A review of the research. Journal of Substance Abuse. 2001;13:391–424. doi: 10.1016/s0899-3289(01)00098-0. doi:10.1016/S0899-3289(01)00098-0. [DOI] [PubMed] [Google Scholar]
  5. Breslin F. C., Adlaf E. M. Part-time work and adolescent heavy episodic drinking: The influence of family and community context. Journal of Studies on Alcohol. 2005;66:784–794. doi: 10.15288/jsa.2005.66.784. doi:10.15288/jsa.2005.66.784. [DOI] [PubMed] [Google Scholar]
  6. Bureau of Labor Statistics. College enrollment and work activity of high school graduates. 2018. Retrieved from https://www.bls.gov/news.release/hsgec.toc.htm. [Google Scholar]
  7. Carter A. C., Brandon K. O., Goldman M. S. The college and noncollege experience: A review of the factors that influence drinking behavior in young adulthood. Journal of Studies on Alcohol and Drugs. 2010;71:742–750. doi: 10.15288/jsad.2010.71.742. doi:10.15288/jsad.2010.71.742. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Casswell S., Pledger M., Hooper R. Socioeconomic status and drinking patterns in young adults. Addiction. 2003;98:601–610. doi: 10.1046/j.1360-0443.2003.00331.x. doi:10.1046/j.1360-0443.2003.00331.x. [DOI] [PubMed] [Google Scholar]
  9. Conway K. P., Vullo G. C., Nichter B., Wang J., Compton W. M., Iannotti R. J., Simons-Morton B. Prevalence and patterns of polysubstance use in a nationally representative sample of 10th graders in the United States. Journal of Adolescent Health. 2013;52:716–723. doi: 10.1016/j.jadohealth.2012.12.006. doi:10.1016/j.jadohealth.2012.12.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Crowley J. E. Educational status and drinking patterns: How representative are college students? Journal of Studies on Alcohol. 1991;52:10–16. doi: 10.15288/jsa.1991.52.10. doi:10.15288/jsa.1991.52.10. [DOI] [PubMed] [Google Scholar]
  11. Currie C., Molcho M., Boyce W., Holstein B., Torsheim T., Richter M. Researching health inequalities in adolescents: The development of the Health Behaviour in School-Aged Children (HBSC) family affluence scale. Social Science & Medicine. 2008;66:1429–1436. doi: 10.1016/j.socscimed.2007.11.024. doi:10.1016/j.socscimed.2007.11.024. [DOI] [PubMed] [Google Scholar]
  12. Dawson D. A., Grant B. F., Stinson F. S., Chou P. S. Another look at heavy episodic drinking and alcohol use disorders among college and noncollege youth. Journal of Studies on Alcohol. 2004;65:477–488. doi: 10.15288/jsa.2004.65.477. doi:10.15288/jsa.2004.65.477. [DOI] [PubMed] [Google Scholar]
  13. DiBello A. M., Benz M. B., Miller M. B., Merrill J. E., Carey K. B. Examining residence status as a risk factor for health risk behaviors among college students. Journal of American College Health. 2018;66:187–193. doi: 10.1080/07448481.2017.1406945. doi:10.1080/07448481.2017.1406945. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gfroerer J. C., Greenblatt J. C., Wright D. A. Substance use in the US college-age population: Differences according to educational status and living arrangement. American Journal of Public Health. 1997;87:62–65. doi: 10.2105/ajph.87.1.62. doi:10.2105/AJPH.87.1.62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Harford T. C., Wechsler H., Muthén B. O. The impact of current residence and high school drinking on alcohol problems among college students. Journal of Studies on Alcohol. 2002;63:271–279. doi: 10.15288/jsa.2002.63.271. doi:10.15288/jsa.2002.63.271. [DOI] [PubMed] [Google Scholar]
  16. Henry K. L., Knight K. E., Thornberry T. P. School disengagement as a predictor of dropout, delinquency, and problem substance use during adolescence and early adulthood. Journal of Youth and Adolescence. 2012;41:156–166. doi: 10.1007/s10964-011-9665-3. doi:10.1007/s10964-011-9665-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hingson R. W., Zha W. Binge drinking above and below twice the adolescent thresholds and health-risk behaviors. Alcoholism: Clinical and Experimental Research. 2018;42:904–913. doi: 10.1111/acer.13627. doi:10.1111/acer.13627. [DOI] [PubMed] [Google Scholar]
  18. Hingson R. W., Zha W., White A. M. Drinking beyond the binge threshold: Predictors, consequences, and changes in the US. American Journal of Preventive Medicine. 2017;52:717–727. doi: 10.1016/j.amepre.2017.02.014. doi:10.1016/j.amepre.2017.02.014. [DOI] [PubMed] [Google Scholar]
  19. Huckle T., You R. Q., Casswell S. Socio-economic status predicts drinking patterns but not alcohol-related consequences independently. Addiction. 2010;105:1192–1202. doi: 10.1111/j.1360-0443.2010.02931.x. doi:10.1111/j.1360-0443.2010.02931.x. [DOI] [PubMed] [Google Scholar]
  20. Johnston L. D., Miech R. A., O’Malley P. M., Bachman J. G., Schulenberg J. E., Patrick M. E. Monitoring the Future national survey results on drug use, 1975-2017: Overview, key findings on adolescent drug use. Ann Arbor, MI: University of Michigan, Institute for Social Research; 2018. [Google Scholar]
  21. Kar I. N., Haynie D. L., Luk J. W., Simons-Morton B. G. Personal income and substance use among emerging adults in the United States. Substance Use & Misuse. 2018;53:1984–1996. doi: 10.1080/10826084.2018.1449863. doi:10.1080/10826084.2018.1449863. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lee C. M., Cadigan J. M., Fairlie A. M., Lewis M. A. Transitions into young adulthood: Extent to which alcohol use, perceived drinking norms, and consequences vary by education and work statuses among 18–20 year olds. Addictive Behaviors. 2018;79:107–112. doi: 10.1016/j.addbeh.2017.12.004. doi:10.1016/j.addbeh.2017.12.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Leeman R. F., Hoff R. A., Krishnan-Sarin S., Patock-Peckham J. A., Potenza M. N. Impulsivity, sensation-seeking, and part-time job status in relation to substance use and gambling in adolescents. Journal of Adolescent Health. 2014;54:460–466. doi: 10.1016/j.jadohealth.2013.09.014. doi:10.1016/j.jadohealth.2013.09.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Linden-Carmichael A. N., Lanza S. T. Drinking patterns of college- and non-college-attending young adults: Is high-intensity drinking only a college phenomenon? Substance Use & Misuse. 2018;53:2157–2164. doi: 10.1080/10826084.2018.1461224. doi:10.1080/10826084.2018.1461224. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Martens M. P., Pederson E. R., LaBrie J. W., Ferrier A. G., Cimini M. D. Measuring alcohol-related protective behavioral strategies among college students: Further examination of the Protective Behavioral Strategies Scale. Psychology of Addictive Behaviors. 2007;21:307–315. doi: 10.1037/0893-164X.21.3.307. doi:10.1037/0893-164X.21.3.307. [DOI] [PubMed] [Google Scholar]
  26. Martens M. P., Taylor K. K., Damann K. M., Page J. C., Mowry E. S., Cimini M. D. Protective behavioral strategies when drinking alcohol and their relationship to negative alcohol-related consequences in college students. Psychology of Addictive Behaviors. 2004;18:390–393. doi: 10.1037/0893-164X.18.4.390. doi:10.1037/0893-164X.18.4.390. [DOI] [PubMed] [Google Scholar]
  27. Mensch B. S., Kandel D. B. Dropping out of high school and drug involvement. Sociology of Education. 1988;61:95–113. doi:10.2307/2112267. [Google Scholar]
  28. Murphy J. G., Correia C. J., Barnett N. P. Behavioral economic approaches to reduce college student drinking. Addictive Behaviors. 2007;32:2573–2585. doi: 10.1016/j.addbeh.2007.05.015. doi:10.1016/j.addbeh.2007.05.015. [DOI] [PubMed] [Google Scholar]
  29. Muthén B. O., Muthén L. K. The development of heavy drinking and alcohol-related problems from ages 18 to 37 in a U.S. national sample. Journal of Studies on Alcohol. 2000;61:290–300. doi: 10.15288/jsa.2000.61.290. doi:10.15288/jsa.2000.61.290. [DOI] [PubMed] [Google Scholar]
  30. Muthén L. K., Muthén B. O. Mplus user’s guide. Eighth edition. Los Angeles, CA: Author; 1998–2017. [Google Scholar]
  31. Osilla K. C., Hunter S. B., Ewing B. A., Ramchand R., Miles J. N., D’Amico E. J. The effects of employment among adolescents at-risk for future substance use. Addictive Behaviors. 2013;38:1616–1619. doi: 10.1016/j.addbeh.2012.09.012. doi:10.1016/j.addbeh.2012.09.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Paschall M. J., Flewelling R. L., Russell T. Why is work intensity associated with heavy alcohol use among adolescents? Journal of Adolescent Health. 2004;34:79–87. doi: 10.1016/s1054-139x(03)00206-4. doi:10.1016/S1054-139X(03)00206-4. [DOI] [PubMed] [Google Scholar]
  33. Patrick M. E., Azar B. High-intensity drinking. Alcohol Research: Current Reviews. 2018:49–55. doi: 10.35946/arcr.v39.1.08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Patrick M. E., Terry-McElrath Y. M. High-intensity drinking by underage young adults in the United States. Addiction. 2017;112:82–93. doi: 10.1111/add.13556. doi:10.1111/add.13556. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Patrick M. E., Terry-McElrath Y. M., Kloska D. D., Schulenberg J. E. High-intensity drinking among young adults in the United States: Prevalence, frequency, and developmental change. Alcoholism: Clinical and Experimental Research. 2016;40:1905–1912. doi: 10.1111/acer.13164. doi:10.1111/acer.13164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Patrick M. E., Terry-McElrath Y. M., Miech R. A., O’Malley P. M., Schulenberg J. E., Johnston L. D. Current high-intensity drinking among eighth and tenth grade students in the US. American Journal of Preventive Medicine. 2017;53:904–908. doi: 10.1016/j.amepre.2017.06.027. doi:10.1016/j.amepre.2017.06.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Schulenberg J. E., Maggs J. L. A developmental perspective on alcohol use and heavy drinking during adolescence and the transition to young adulthood. Journal of Studies on Alcohol, Supplement. 2002;14:54–70. doi: 10.15288/jsas.2002.s14.54. doi:10.15288/jsas.2002.s14.54. [DOI] [PubMed] [Google Scholar]
  38. Sheffield F. D., Darkes J., Del Boca F. K., Goldman M. S. Binge drinking and alcohol-related problems among community college students: Implications for prevention policy. Journal of American College Health. 2005;54:137–141. doi: 10.3200/JACH.54.3.137-142. doi:10.3200/JACH.54.3.137-142. [DOI] [PubMed] [Google Scholar]
  39. Simons-Morton B., Haynie D., Liu D., Chaurasia A., Li K., Hingson R. The effect of residence, school status, work status, and social influence on the prevalence of alcohol use among emerging adults. Journal of Studies on Alcohol and Drugs. 2016;77:121–132. doi: 10.15288/jsad.2016.77.121. doi:10.15288/jsad.2016.77.121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Townsend L., Flisher A. J., King G. A systematic review of the relationship between high school dropout and substance use. Clinical Child and Family Psychology Review. 2007;10:295–317. doi: 10.1007/s10567-007-0023-7. doi:10.1007/s10567-007-0023-7. [DOI] [PubMed] [Google Scholar]
  41. Velazquez C. E., Pasch K. E., Laska M. N., Lust K., Story M., Ehlinger E. P. Differential prevalence of alcohol use among 2-year and 4-year college students. Addictive Behaviors. 2011;36:1353–1356. doi: 10.1016/j.addbeh.2011.07.037. doi:10.1016/j.addbeh.2011.07.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Wall A. F., BaileyShea C., McIntosh S. Community college student alcohol use: Developing context-specific evidence and prevention approaches. Community College Review. 2012;40:25–45. doi:10.1177/0091552112437757. [Google Scholar]
  43. White H. R., McMorris B. J., Catalano R. F., Fleming C. B., Haggerty K. P., Abbott R. D. Increases in alcohol and marijuana use during the transition out of high school into emerging adulthood: The effects of leaving home, going to college, and high school protective factors. Journal of Studies on Alcohol. 2006;67:810–822. doi: 10.15288/jsa.2006.67.810. doi:10.15288/jsa.2006.67.810. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Journal of Studies on Alcohol and Drugs are provided here courtesy of Rutgers University. Center of Alcohol Studies

RESOURCES