Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Jan 2.
Published in final edited form as: Subst Use Misuse. 2017 Jul 25;53(1):26–35. doi: 10.1080/10826084.2017.1322983

Post-high school changes in tobacco and cannabis use in the United States

FEARGHAL O’BRIEN a,b,*, BRUCE SIMONS-MORTON b, ASHOK CHAURASIA c, JEREMY LUK d, DENISE HAYNIE b, DANPING LIU b
PMCID: PMC6300995  NIHMSID: NIHMS1514756  PMID: 28742412

Abstract

Background:

The transition from high school into young adulthood is a critical developmental period with many young people going to college, moving residence, and entering the work force for the first time. The NEXT Generation Health Study (NEXT) is a nationally representative longitudinal study of adolescent health behaviors. Previous NEXT research has found that the post-high school environment is associated with changes in alcohol use.

Objectives:

The current study investigated the impact of school status, residential status, and work status on cannabis and cigarette use among post-high school participants.

Results:

Living in a dorm/fraternity/sorority was associated with an increased prevalence in cannabis use while attending a 4-year college was associated with a decreased prevalence in cigarette use.

Conclusions:

Some aspects of the post-high school environment are related to cannabis and cigarette use. Differences in the social circumstances of cigarette and cannabis use and recent campaigns in colleges to reduce smoking may explain some of these trends.

The transition from high school into young adulthood is marked by changes in interpersonal relationships and role responsibilities, and increased rates of substance use behaviors (Jackson, Sher, & Schulenberg, 2008; Newcomb & Bentler, 1987; White, Labouvie, & Papadaratsakis, 2005). While many young adults move toward the assumption of adult roles and responsibilities during this stage of life, considerable individual variation exists in decisions and pursuits regarding education, work, financial autonomy, and romantic and peer involvement (Cohen, Kasen, Chen, Hartmark, & Gordon, 2003; Schulenberg, Bryant, & O’Malley, 2004). Greater freedom, less social control, and increased exploratory behaviors within new contexts contribute to increases in substance use (Arnett, 2005; Schulenberg & Maggs, 2002). In this critical transition period, many adolescents go to college, take jobs, and live away from their parents for the first time.

Cigarette and cannabis use during this later stage of adolescence are important for multiple reasons. Research on the impact of early age at initiation of smoking has typically focused on those initiating before age 16 years (e.g. Khuder, Dayal, & Mutgi, 1999; Thompson, Tebes, & McKee, 2015). However, studies that incorporate a broader age range of initiation have found that individuals who begin smoking during this transition-age (17–19 years), compared with older initiators, have decreased interest in quitting later (Lando et al., 1999) and decreased probability of successfully doing so (Chen & Millar, 1998), smoke more cigarettes per day (Fernandez et al., 1999), and have more physical dependence symptoms (Lando et al., 1999). Among university students smoking has been associated with higher levels of depression and greater utilization of mental health services (Halperin, Smith, Heiligenstein, Brown, & Fleming, 2010). Cannabis use among undergraduates has been associated with major depressive, anxiety, and substance use disorders (Keith, Hart, McNeil, Silver, & Goodwin, 2015). Additionally, cannabis use has been found to reduce post-secondary school performance (Arria, Caldeira, Bugbee, Vincent, & O’Grady, 2015).

Cigarette Smoking and Cannabis Use by College, Residence, and Work Status

Community-based (e.g. White et al., 2005) and national cross-sectional studies of young adults (e.g. Johnston, O’Malley, Bachman, Schulenberg, & Miech, 2015) suggest that, in contrast to drinking, which is more prevalent among those college-bound, cigarette smoking and cannabis use are more common among those non-college-bound. Johnston et al. found, using data from Monitoring the Future, that non-college-bound young adults had higher 30-day prevalence rates of cigarette smoking (24.6%) and cannabis use (26%) than their college-bound peers (12.9% and 20.8% respectively). Multiple regional studies indicate that young adults attending 2-year colleges are more likely than those attending 4-year colleges to report smoking; though those who not attending college have the highest prevalence of smoking (Lenk et al., 2012; Sanem, Berg, An, Kirch, & Lust, 2009; VanKim, Laska, Ehlinger, Lust, & Story, 2010). Operationalization of college status differs across studies; some consider full- and part-time students as separate groups and some consider 2- and 4-year colleges as separate types of education (see Carter, Brandon, & Goldman, 2010). Accordingly, comparisons of cigarette smoking and cannabis use rates across these three groups (i.e., not in college, 2-year college, and 4-year college) are difficult.

About 28% of 18–24 year olds in the U.S. attend 4-year colleges and about 13% attend 2-year colleges (NCES, 2014). There is considerably more research on 4-year college students than other emerging adults. Notably, college students’ affiliation with on-campus organizations, e.g., fraternities or sororities has been linked to a cluster of risk behaviors including alcohol, tobacco, and cannabis use, as well as more sexual partners and sex under the influence of substances (Primack et al., 2012; Scott-Sheldon, Carey, & Carey, 2008; Sidani, Shensa, & Primack, 2013). A recent review showed that fraternity/sorority members are more likely to smoke cigarettes and use cannabis than nonmembers, and that members who lived in a fraternity/sorority house had greater rates of smoking than members who lived elsewhere (Cheney, Harris, Gowin, & Huber, 2014). Fraternity/sorority membership confers risk for college student substance use through both selection and socialization, although membership was associated with increases in drinking and cannabis use but not cigarette smoking over time (McCabe et al., 2005). In a community-based longitudinal study that directly examined living away from home, White et al. (2006) found that it was not associated with changes in cannabis use.

High school graduates not attending college are almost twice as likely to be working as those attending college (73% vs. 38%, respectively; Bureau of Labor Statistics, 2015). There is limited research on the relationship between employment and substance use in this age group. Among adults, working longer hours has been associated with increased cigarette use among regular smokers, greater likelihood of relapse among former smokers, and reduced likelihood of quitting smoking (Angrave, Charlwood, & Wooden, 2014). Angrave et al. speculate that this may be due to the increased stress experienced by those working longer hours. Among adolescents, number of hours worked per week has been associated with increased likelihood of cigarette and cannabis use among 8th- and 10th-graders (Bachman, Safron, Sy, & Schulenberg, 2003). Bachman et al. suggest that this may be due to some students adopting premature adult roles, though they do not offer evidence to support this. In a college sample, Sanem et al. (2009) found those who worked 20–39 hours had higher odds of current and daily cigarette use.

Peer and Parental Influences

Peer substance use strongly predicts increased substance use during the transition after high school (Kirst, Mecredy, Borland, & Chaiton, 2014). In a prospective study among a small non-national sample, Andrews, Tildesley, Hops, and Li (2002) found that socialization with substance-using peers predicted increased binge drinking and cigarette smoking but not cannabis use. A longitudinal investigation in New Jersey found that friends’ tobacco use was positively associated with the probability of smoking among adolescent females but not males (White, Pandina, & Chen, 2002). Positive parenting behaviors such as support and monitoring are protective against adolescent substance use and, to a lesser extent, substance use during young adulthood (Kirst et al., 2014; Stone, Becker, Huber, & Catalano, 2012). Parental influence on substance use may be indirect by influencing levels of deviant peer affiliation (Simons-Morton & Farhat, 2010; Van Ryzin, Fosco, & Dishion, 2012).

Current study and objectives

The NEXT Generation Health Study (NEXT) is a 7-year, nationally representative, longitudinal study beginning with 10th grade students in 2010. Previous research has generally examined community and local samples (e.g. White et al., 2005), with cross sectional studies (e.g. Johnston et al., 2015), or with relatively old data, such as the Monitoring the Future Project (Schulenberg et al., 2004) which began in 1975. The current study extends Simons-Morton et al.’s (2016) study of the relationship between school, residential, and work statuses and alcohol use in this age group. The purpose of this research is to examine how the exposure to different school, residential, and work environments shapes cigarette/cannabis use behaviors in the first year after high school.

Materials and methods

Sample

NEXT is a longitudinal study of a cohort of U.S. students ascertained through multistage sampling. Primary sampling units consisted of school districts or groups of school districts stratified across the nine U.S. census divisions. Eighty-one out of 137 (58.4%) randomly selected schools agreed to participate starting in the 2009–2010 school year (Conway et al., 2013). Within each school, 10th-grade classes were randomly selected; student assent and parental consent were obtained; participants provided consent at age 18. Students were surveyed annually, with a school-based assessment in the spring semester of 10th grade (Wave 1 [W1], 2010; n = 2,524), and web-based assessments in 11th grade (W2, 2011; n = 2,439), 12th grade (W3, 2013; n = 2,407), and the first year after high school (W4, 2013; n = 2,177). No information was collected on the characteristics of those schools that declined to participate, or on the students within those schools. The protocol was approved by Institutional Review Board of the Eunice Kennedy Shriver National Institute of Child Health and Human Development.

Measures (all waves unless otherwise indicated)

Cigarette and Cannabis use.

30-day cigarette use and 12-month cannabis use prevalence were assessed as the number of occasions used in the time-frame. For both questions, response ranged from never to 40 times or more. Sparsity in some categories necessitated collapsing each variable to three levels: never, 1–5 times, 6 or more times. The word “cannabis” is used in the present manuscript. However, all questions in the questionnaire that asked about this substance used the word “marijuana.”

Demographic covariates.

Baseline demographic variables included: sex; race/ethnicity (White, African American, Hispanic, other); family structure (two biological parents, biological parent and stepparent, single parent, other), parent reported parental education (high school or less, some college or technical school, bachelor’s degree or higher), and family affluence (low, moderate, high). The latter of these variables was calculated based on responses to 4 questions relating to the number of computers owned by family, number of cars/vans/trucks owned by family, number of vacations taken each year by family, and whether the participant had their own bedroom.

Parenting practices (Waves 1, 2, 3).

Perceived parental expectations for cigarette and cannabis use were measured. Participants were asked how important it is to their parents/guardians that they do not use cigarettes and, separately, not use cannabis (Hartos, Eitel, Haynie, & Simons-Morton, 2000; Hetherington et al., 1992). Response options ranged from 1 (not at all) to 7 (extremely). Parental monitoring-knowledge was assessed using questions adapted from a validated five-item scale (Brown, Mounts, Lamborn, & Steinberg, 1993). Adolescents reported their perceptions of their mother’s and, separately, father’s knowledge about who their friends were, how they spent their money, where they were after school, where they went at night, and what they did with their free time. Response options (1–4) ranged from don’t have/see father/mother, to he/she knows a lot. Higher scores reflect higher levels of monitoring-knowledge. Scores were averaged across items and across parents (α = .87, .88, and .89 for W1, W2, and W3, respectively).

Peer cigarette and cannabis use.

Participants indicated their perceptions of the frequency with which their five closest friends “smoke cigarettes” and, separately, “smoke/use marijuana”. Response options ranged from never, to almost always, collapsed to three levels (1=never, 2–3=almost never/sometimes, 4–5=often/almost always) because of sparsity in some categories.

Depressive symptoms (Waves 2, 3, and 4).

Respondents completed the Pediatric Depressive Symptoms Scale (Irwin et al., 2010), reporting how often they felt unhappy, sad, lonely, alone, that life was bad, that everything went wrong, that they could not do anything right, and that they could not stop feeling bad. Response options (1–5) ranged from never, to almost always. Responses were averaged, a higher score indicated more depressive symptoms (Dahlberg, Toal, Swahn, & Behrens, 2005) (α = .94, .95, and .96 for W2, W3, and W4, respectively).

Residential status (Wave 4).

Participants reported their current and previous residence in the past 12 months. Categories were combined to reflect living with parents, on their own (own place, rented room), on campus (residence hall/dorm or fraternity or sorority house), or other (dropped from analysis due to sparsity).

School status (Wave 4).

Participants indicated that they did not attend school, attended high school, technical/vocational school, community college, college/university, or graduate/ professional school, and provided the school’s name and location. Three categories were used in the analysis: not attending school, attending technical school/community college, and attending a college/university.

Work status (Wave 4).

Total hours per week in paid and unpaid employment was assessed using an item from Monitoring the Future (Bachman et al., 1983). Response options ranged from 0 to more than 30 hours. Working status was defined as not working, less than 20, 21–30, or more than 30 hours.

Analysis

One hundred twenty-six participants (those in high school and those who reported as their residence barracks, hospital, living with family members other than parents, and who were homeless) were excluded from this analysis. Multiple imputation by chained equations (MICE) based on the assumption of missing at random (Little & Rubin, 2002; van Buuren & Groothuis-Oudshoorn, 2011) was used to impute missing values due to subject or item nonresponse in both outcome and independent variables. Fifty multiply imputed data sets were generated using R package “mice.”

After descriptive statistics were calculated by wave, transition models (Diggle, Heagerty, Liang, & Zeger, 2002) were fitted to analyze the data (N = 2,659). The outcome variables at W2–W4, 30-day cigarette use and 12-month cannabis use, were analyzed separately. Ordinal logistic transition models were used to deal with the trichotomized outcomes. For model simplicity and ease of interpretation, we made a “Markov assumption” (i.e., use in the current wave depended on use in the immediate past wave, but not on prior waves). Therefore, we included cigarette /cannabis use in the previous wave as a covariate. Other previous wave covariates included peer cigarette /cannabis use, depressive symptoms, and parenting practice. Coefficients reflect the average effects of that correlate at each wave on cigarette and cannabis use in the subsequent wave (i.e., W1 on W2, W2 on W3, and W3 on W4).

Environmental variables (school, residential and working status) were assessed only at W4 and hence enter the model as a predictor of W4, but not W1–W3 outcomes. Three environmental variables (employment, residential status, and school status) were tested in separate transition models for both outcome measures for a total of 6 models. The environmental variables were not analyzed in a single model to avoid over-adjusting. For example, one of the categories of residence (dorm) is related to the school status variable and so including these two in the same may model could lead to an attenuated estimation of the residential effect. Coefficients are interpreted as the effect of the environment on W4 cigarette /cannabis use. Models adjusted for baseline demographic variables. Transition models were estimated using generalized estimating equations, implemented in SAS PROC GENMOD (SAS Institute Inc., Cary, NC). Features of complex survey design, including clustering and sampling weights, were taken into account. Analyses were repeated for 50 imputed data sets; results combined using Rubin’s combination rule (Little & Rubin, 2002) implemented in SAS PROC MIANALYZE.

Results

Of the 2,659 participants, 45% (weighted) were male, and 56% were White, 20 % African American, and 18% Hispanic. The mean age at W1 was 16.3 years (SE = 0.03). There were 23.6% of participant in the low family affluence category, 50.0% in the moderate affluence category, and 27.3% in the high affluence category. Prevalence of cigarette use and cannabis use and values of other time-varying covariates are shown in Table 1. Cigarette use changed very little between W1 and W2; 9%−10% of participants reported having smoked cigarettes six or more times in the last 30 days. There was an increase at W3 to roughly 14% and with little change at W4. Cannabis use fluctuated with the lowest prevalence at W2 (12.8% for 6 or more times) and the highest at W4 (20.2%), a similar fluctuation found for reports of using 1–5 times.

Table 1.

Summary statistics for longitudinal outcomes and covariates. Mean (SE) is reported for continuous variables and percentage (SE) is reported for categorical variables

Variable Range/labels W1, % W2, % W3, % W4, %
Cigarette Use Never 81.6 (2.4) 82.3 (2) 77.4 (2.5) 75 (1.97)
1–5 Times 8.7 (1.4) 8.5 (1) 8.5 (1.2) 10.6 (0.96)
6+ Times 9.7 (1.6) 9.2 (1.7) 14.1 (2) 14.4 (1.99)
Cannabis Use Never 74 (2.1) 75 (2) 72.7 (1.8) 65.8 (2.35)
1–5 Times 11.4 (1) 12.2 (1.2) 11.5 (0.9) 13.9 (1.59)
6+ Times 14.6 (1.7) 12.8 (1.5) 15.8 (1.7) 20.2 (1.99)
5 Friends Cigarette
Use		 Never 74.6 (2.3) 76.9 (2) 73.8 (1.9) 71.9 (2.22)
Almost
never/sometimes		 15.9 (1.4) 13.6 (1.4) 15.4 (1) 17.7 (1.42)
Often/always 9.5 (1.4) 9.5 (1.6) 10.9 (1.5) 10.4 (1.54)
5 Friends Cannabis
Use		 Never 72.5 (2.3) 74.7 (1.9) 72.3 (2.1) 68.8 (2.67)
Almost
never/sometimes		 17.8 (1.8) 17.2 (1.4) 16.3 (1.3) 19 (2.02)
Often/always 9.7 (1.1) 8.1 (1.1) 11.4 (1.3) 12.2 (1.5)
Parental Knowledge 1–4 2.35 (0.02) 2.32 (0.02) 2.3 (0.03) –
Parental Expectations
Cigarette Use		 1–7 6.19 (0.05) 5.96 (0.06) 5.91 (0.07) –
Parental Expectations
Cannabis Use		 1–7 6.16 (0.06) 5.99 (0.06) 5.9 (0.08) –
Depressive symptoms 1–5 2.34 (0.03) 2.08 (0.03) 2.04 (0.05) 2.01 (0.04)
Working status Not working – – – 48.9 (2.7)
20 hours or less – – – 20.4 (1.8)
21–30 hours – – – 14.2 (1.2)
>30 hours – – – 16.5 (1.7)
Residential status Parent’s home – – – 56.5 (2.6)
On campus – – – 16.1 (1.5)
Own place – – – 27.4 (2.3)
School status Not attending – – – 28.3 (2)
Technical school/
community
college		 – – – 28.9 (1.8)
College/university – – – 42.8 (2.5)
Open in a new tab

Notes: W = wave

Tables 2 and 3 show results of the transition models. Each transition model examined one environmental variable. Because the odds ratios for other variables (e.g., sex) were very similar across models for both cigarette and cannabis use, the non-environmental variables discussed below are taken from the model with school status (the first column in Tables 2 and 3).

Table 2.

Transition model for cigarette use in the last 30 days.

Cigarette use as the outcome
Variable Model 1a Model 2 Model 3
Sex
 Male (ref)
 Female 0.71 (0.57–0.89)** 0.7 (0.56–0.87)** 0.7 (0.56–0.86)***
Race/Ethnicity
 White (ref)
 African-American 0.34 (0.22–0.53)*** 0.34 (0.22–0.53)*** 0.34 (0.22–0.53)***
 Hispanic 0.57 (0.39–0.85)** 0.57 (0.39–0.83)** 0.57 (0.39–0.83)**
 Other 0.67 (0.33–1.38) 0.66 (0.32–1.37) 0.66 (0.32–1.38)
Parent Education
 High school or less (ref)
 Bachelor’s degree or higher 0.81 (0.59–1.13) 0.79 (0.57–1.1) 0.77 (0.56–1.06)
 Some college 1.12 (0.81–1.54) 1.07 (0.78–1.48) 1.07 (0.78–1.48)
Family structure
 Both biological parents (ref)
 Biological and stepparent 1.32 (1.03–1.7)* 1.35 (1.05–1.73)* 1.35 (1.06–1.73)*
 Other 1.38 (0.96–2) 1.41 (0.99–2.03) 1.43 (0.99–2.05)
 Single parent 1.2 (0.9–1.61) 1.21 (0.9–1.62) 1.21 (0.9–1.64)
Family Affluence
 Low (ref)
 Moderate 0.92 (0.73–1.17) 0.92 (0.72–1.18) 0.91 (0.72–1.16)
 High 1.17 (0.87–1.57) 1.16 (0.86–1.58) 1.14 (0.85–1.54)
Depression 1.25 (1.06–1.48)** 1.25 (1.06–1.48)** 1.25 (1.06–1.48)**
Parental Knowledge 0.85 (0.65–1.11) 0.86 (0.66–1.13) 0.86 (0.66–1.12)
Parental Expectations about cigarette
use		 0.9 (0.85–0.94)*** 0.89 (0.85–0.94)*** 0.89 (0.85–0.94)***
Cigarette use Previous Wave
 Never (ref)
 1 to 5 times 3.55 (2.57–4.9)*** 3.65 (2.65–5.02)*** 3.71 (2.73–5.05)***
 More than 6 times 18.08 (12.03–27.17)*** 18.67 (12.38–28.15)*** 19.17 (12.78–28.77)***
Five Closest Friends Smoke Cigarettes
 Never (ref)
 Seldom/Sometimes 1.85 (1.37–2.5)*** 1.83 (1.36–2.47)*** 1.81 (1.35–2.41)***
 Often/Always 2.72 (1.72–4.31)*** 2.72 (1.73–4.27)*** 2.68 (1.69–4.26)***
Wave
 2 (ref)
 3 1.81 (1.33–2.47)*** 1.82 (1.33–2.49)*** 1.82 (1.33–2.5)***
 4 1.23 (0.81–1.85) 1.71 (1.29–2.27)*** 1.76 (1.21–2.54)**
W4*School Status
 College/university (ref)
 Not attending 2.06 (1.33–3.19)**
 Technical school/community college 1.13 (0.64–1.99)
W4*Residential Status
 Parents’ Home (ref)
 Dorm/Sorority/Fraternity 0.72 (0.43–1.22)
 Own Place/Rented Room 1.14 (0.72–1.82)
W4*Work Status
 0 hours per week (ref)
 Less than 20 hours per week 0.71 (0.33–1.51)
 21–30 hours per week 1.08 (0.6–1.96)
 30+ hours per week 0.86 (0.48–1.55)
Open in a new tab
*

p < .05

**

p < .01

***

p < .001.

Notes. ref = referent.

a

Each model includes just one environmental variable without adjustment for the other two environmental variables.

Table 3.

Transition model for Cannabis use in the last year

Cannabis use as the outcome
Variable Model 1a Model 2 Model 3
Sex
 Male (ref)
 Female 0.66 (0.56–0.78)*** 0.66 (0.55–0.78)*** 0.66 (0.56–0.78)***
Race/Ethnicity
 White (ref)
 African-American 0.99 (0.72–1.36) 0.97 (0.72–1.32) 0.97 (0.71–1.32)
 Hispanic 0.8 (0.63–1)* 0.79 (0.64–0.99)* 0.77 (0.62–0.97)*
 Other 0.93 (0.44–1.96) 0.93 (0.44–1.94) 0.91 (0.43–1.92)
Parent Education
 High school or less (ref)
 Bachelor’s degree or higher 1.01 (0.75–1.37) 1 (0.74–1.34) 1.01 (0.75–1.37)
 Some college 0.94 (0.76–1.16) 0.93 (0.76–1.16) 0.93 (0.76–1.16)
Family structure
 Both biological parents (ref)
 Biological and stepparent 1.03 (0.8–1.33) 1.03 (0.8–1.33) 1.03 (0.8–1.33)
 Other 1.2 (0.87–1.65) 1.2 (0.87–1.67) 1.19 (0.86–1.65)
 Single parent 1.27 (0.98–1.64) 1.27 (0.99–1.64) 1.27 (0.99–1.64)
Family Affluence
 Low (ref)
 Moderate 1.2 (0.96–1.51) 1.19 (0.95–1.5) 1.21 (0.96–1.52)
 High 1.45 (1.04–2.01)* 1.42 (1.03–1.97)* 1.45 (1.04–2.01)*
Depression 1.11 (0.99–1.25) 1.11 (0.99–1.25) 1.11 (0.99–1.24)
Parental Knowledge 0.75 (0.6–0.94)* 0.75 (0.6–0.93)** 0.75 (0.6–0.93)**
Parental expectations about cannabis
use		 0.94 (0.89–0.99)* 0.94 (0.89–0.98)** 0.94 (0.89–0.99)*
Cannabis Use Previous Wave
 Never (ref)
 1 to 5 times 5.62 (4.11–7.69)*** 5.66 (4.14–7.74)*** 5.58 (4.09–7.62)***
 More than 6 times 17.15 (12.03–24.46)*** 17.31 (12.11–24.75)*** 17.09 (12.03–24.28)***
Five Closest Friends Use Cannabis
 Never (ref)
 Seldom/Sometimes 1.93 (1.42–2.62)*** 1.91 (1.41–2.59)*** 1.94 (1.44–2.61)***
 Often/Always 2.99 (1.99–4.51)*** 2.95 (1.97–4.41)*** 3 (2.01–4.49)***
Wave
 2 (ref)
 3 1.37 (0.97–1.95) 1.37 (0.97–1.94) 1.37 (0.96–1.95)
 4 2.13 (1.58–2.88)*** 1.87 (1.39–2.51)*** 2.13 (1.62–2.79)***
W4*School Status
 College/university (ref)
 Not attending 0.89 (0.58–1.38)
 Technical school/community college 0.83 (0.53–1.31)
W4*Residential Status
 Parents’ Home (ref)
 Dorm/Sorority/Fraternity 1.34 (1–1.81)
 Own Place/Rented Room 0.78 (0.45–1.33)
W4*Work Status
 0 hours per week (ref)
 Less than 20 hours per week 0.94 (0.56–1.57)
 21–30 hours per week 0.72 (0.43–1.2)
 30+ hours per week 0.86 (0.51–1.44)
Open in a new tab
*

p < .05

**

p < .01

***

p < .001.

Notes. ref = referent.

a

Each model includes just one environmental variable without adjustment for the other two environmental variables.

For cigarette use (Table 2), females had a lower prevalence than males (odds ratio [OR] = 0.71, 95% CI [0.57,0.89]). White participants had a higher prevalence than African American (OR = 0.34 [0.22,0.53]) and Hispanic participants (OR = 0.57 [0.39,0.85]). Living with a biological parent and stepparent was with an increased prevalence of cigarette use (OR = 1.32 ([1.03–1.7]). Higher levels of depression were associated with high prevalence of cigarette use (OR = 1.25 [1.06,1.48]), while higher parental expectations not to use were associated with lower levels (OR = 0.9 [0.85,0.94]). Having used cigarettes frequently was strongly associated with using cigarettes in the current wave (OR = 18.08 [12.03,28.15], OR = 3.55 [2.57,4.9], for more than 6 times and 1 to 5 times, respectively). Having friends who smoke was associated with increased cigarette use, (OR = 2.72 [1.72,4.31] and OR = 1.85 [1.37,2.5] often/always smoke and seldom/sometimes smoke, respectively). In the school status model, participants not attending school had an increased prevalence of cigarette use compared with participants in a four year college (OR = 2.06 [1.33,3.19]). There was no difference between non-attenders and those attending technical school/community college (OR = 1.13 [0.64,1.99]). Cigarette use was not associated with either residential status or work status.

Females had a lower prevalence of cannabis use than males (Table 3; OR = 0.66 [0.56,0.78]), Hispanic participants had a lower prevalence than White participants (OR = 0.8 [0.63,1.0]), increased parental knowledge was associated with less cannabis use (OR = 0.75 [0.6,0.94]), and increases in parental expectation to not use were associated with less use (OR = 0.94 [0.89,0.99]). Compared with those who had low family affluence, high family affluence was associated with cannabis use (1.45 [1.04–2.01]). Previous use was highly associated (OR = 17.15 [12.03–24.46] for more than 6 times) and friends’ use was associated (OR = 2.99 [1.99–4.51] for friends often/always use) with cannabis use. Regarding residential status, participants staying in a dorm/sorority/fraternity had marginally significant (p = 0.054) increased prevalence of cannabis use versus those living with their parents (OR = 1.34 [1.0,1.81]). There was no difference between participants living in their own place or a rented room and those living with their parents (OR = 0.78 [0.45,1.33]). No relationship was found between cannabis use and either school status or work status

Discussion

Similar to previous findings, we found females less likely to smoke cigarettes than males (Johnston et al., 2015) and African-American and Hispanic participants less likely to smoke cigarettes than white participants (Wallace et al., 2002). Higher levels of depressive symptoms were associated with increased cigarette use. Previous research suggests that cigarette smoking during adolescence may increase the likelihood of depression later in life (Brook, Schuster, & Zhang, 2010). We found that depressive symptoms, in turn, lead to continuation of cigarette use in the future. As expected, previous use and peer smoking strongly predicted smoking, (Kirst et al. (2014). Perceived parental expectations for not smoking was associated with less smoking, though the effect was small. Regarding the environmental variables, participants attending a 4-year college/university smoked less than their non-attending peers. This confirms in a longitudinal, nationally representative study findings from previous community (White et al., 2005) and cross sectional research (Johnston et al., 2015). Cigarette use was associated with neither residential nor work status. This latter finding contrasts with previous research suggesting that young people who work are more likely to smoke (Bachman et al., 2003; Sanem et al., 2009). However, past research has looked at younger adolescents or has focused solely on college students.

Females were less likely than males and Hispanic participants were less likely than white participants to use cannabis. While these sex differences are compatible with previous findings (Johnston et al., 2015), past research has found that African-Americans are less likely to use cannabis than white Americans (Wallace et al., 2002), which was not found in the current study. The current, nationally representative data are some of the most recently available, so this lack of differences could reflect a secular trend. Parental expectations not to use and higher monitoring/knowledge were associated with less cannabis use; both results consistent with a recent meta-analysis (Lac & Crano, 2009). As with cigarette use, cannabis use in a previous wave and peer cannabis use were the strongest predictors of current cannabis use. Regarding environmental variables, those living in a college dorm or a fraternity/sorority house were more likely to use cannabis than those living with their parents, those this different was only marginally significant (p = .054). This result somewhat contrasts with that of White et al. (2006) who found that living away from home was not associated with an increase in cannabis use. Given the almost 20 year gap between the current sample and that of White et al., this difference could reflect general societal changes.

Findings in the current study make an interesting comparison with those of Simons-Morton et al. (2016), using the same sample and analytic methods, that those living on campus had higher levels of alcohol use and heavy episodic drinking. In the current study, cannabis use matched this pattern; cigarette use did not. The dissimilarity may reflect the typically social nature of alcohol and cannabis use, though this is speculative. Living on campus may be conducive to socializing in ways that living at home or in other accommodations may not be. Simons-Morton et al. found almost the reverse of the current cigarette findings, that those in a 4-year university used alcohol more than nonstudents and those attending technical/community college. Current findings on cigarette use, but not cannabis use, support the efficacy of recent efforts by universities to reduce on-campus smoking (ANRF, 2016). Additionally, social circumstances in which cigarettes and cannabis are used may differ substantially. Collectively, these findings suggest that alcohol and cannabis use are associated with similar post-high school environmental risk factors which do not play the same role for cigarette use. Future research could examine factors that lead non-college-bound young adults to be at a greater risk for increased cigarette use and factors that lead those living in dorms/fraternities/sororities to be a greater risk for increased cannabis use after high school.

Strengths and limitations

Strengths of this research include data from a relatively large, nationally representative sample of adolescents surveyed over 4 years that include information on a breadth of related factors. Additionally, the study includes emerging adults who do not attend college immediately after high school, or who opt to attend community college or technical schools, groups that have received limited attention. Study limitations include the fact that all measures were self-reported. The different time frames used to measure use complicate direct comparisons of predictors of cigarette use with predictors of cannabis use. Lastly, there were too few respondents living in sorority/fraternity houses to consider them as separate analytic group. This is limiting because past research suggests increased cigarette use among this group (Cheney et al., 2014).

Conclusions

Cigarette and cannabis use among young adults are associated primarily with past use, but also with both parental support and peer use. Those who do not attend college/university are more likely to use cigarettes. This may reflect successes of recent college/university policies in reducing cigarette use, e.g., campus-wide bans. Residents of dorms or fraternity/sorority houses are most likely to use cannabis than their peers who live at home, perhaps reflecting social environments typifying cannabis use.

Footnotes

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

References

  1. Andrews JA, Tildesley E, Hops H, & Li F (2002). The influence of peers on young adult substance use. Health Psychology, 21(4), 349–357. doi: 10.1037/0278-6133.21.4.349 [DOI] [PubMed] [Google Scholar]
  2. Angrave D, Charlwood A, & Wooden M (2014). Working time and cigarette smoking: Evidence from Australia and the United Kingdom. Social Science & Medicine, 112, 72–79. doi: 10.1016/j.socscimed.2014.04.031 [DOI] [PubMed] [Google Scholar]
  3. ANRF. (2016). Smokefree and Tobacco-Free U.S. and Tribal Colleges and Universities [Google Scholar]
  4. Arnett JJ (2005). The Developmental Context of Substance use in Emerging Adulthood. Journal of Drug Issues, 35(2), 235–254. doi: 10.1177/002204260503500202 [DOI] [Google Scholar]
  5. Arria AM, Caldeira KM, Bugbee BA, Vincent KB, & O’Grady KE (2015). The academic consequences of marijuana use during college. Psychology of Addictive Behaviors, 29(3), 564–575. doi: 10.1037/adb0000108 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bachman JG, Safron DJ, Sy SR, & Schulenberg JE (2003). Wishing to work: New perspectives on how adolescents’ part-time work intensity is linked to educational disengagement, substance use, and other problem behaviours. International Journal of Behavioral Development, 27(4), 301–315. doi: 10.1080/01650250244000281 [DOI] [Google Scholar]
  7. Bachman JG, Johnston L, & O’Malley PM (1983). Monitoring the Future: A Continuing Study of the Lifestyles and Values of Youth, 1978 Ann Arbor, MI: Inter-university Consortium for Political and Social Research; Retrieved from http://www.icpsr.umich.edu/icpsrweb/RCMD/studies/7929 [Google Scholar]
  8. Bureau of Labor Statistics. (2014). College enrollment and work activity of 2013 high school graduates Retrieved from http://www.bls.gov/news.release/hsgec.nr0.htm
  9. Bureau of Labor Statistics. (2015). College Enrollment and Work Activity of 2014 High School Graduates Retrieved from http://www.bls.gov/news.release/hsgec.nr0.htm.
  10. Brook JS, Schuster E, & Zhang C (2010). Cigarette smoking and depressive symptoms: A longitudinal study of adolescents and young adults. Psychological Reports, 95(1), 159–166. doi: 10.2466/pr0.95.1.159-166 [DOI] [PubMed] [Google Scholar]
  11. Brown BB, Mounts N, Lamborn SD, & Steinberg L (1993). Parenting Practices and Peer Group Affiliation in Adolescence. Child Development, 64(2), 467–482. doi: 10.2307/1131263 [DOI] [PubMed] [Google Scholar]
  12. Carter AC, Brandon KO, & Goldman MS (2010). The College and Noncollege Experience: A Review of the Factors That Influence Drinking Behavior in Young Adulthood. Journal of Studies on Alcohol and Drugs, 71(5), 742–750. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Chen J, & Millar WJ (1998). Age of smoking inititation: implications for quitting. Health Reports, 9, 39–46. [PubMed] [Google Scholar]
  14. Cheney MK, Harris LW, Gowin MJ, & Huber J (2014). Smoking and membership in a fraternity or sorority: A systematic review of the literature. Journal of American College Health, 62(4), 264–276. doi: 10.1080/07448481.2014.891595 [DOI] [PubMed] [Google Scholar]
  15. Cohen P, Kasen S, Chen H, Hartmark C, & Gordon K (2003). Variations in patterns of developmental transmissions in the emerging adulthood period. Developmental Psychology, 39(4), 657–669. doi: 10.1037/0012-1649.39.4.657 [DOI] [PubMed] [Google Scholar]
  16. Conway KP, Vullo GC, Nichter B, Wang J, Compton WM, Iannotti RJ, & Simons-Morton B (2013). Prevalence and Patterns of Polysubstance Use in a Nationally Representative Sample of 10th Graders in the United States. Journal of Adolescent Health, 52(6), 716–723. doi: 10.1016/j.jadohealth.2012.12.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Dahlberg L, Toal S, Swahn M, & Behrens C (2005). Measuring violence-related attitudes, behaviors, and influences among youths: A compendium of assessment tools Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. [Google Scholar]
  18. Diggle P, Heagerty P, Liang KY, & Zeger S (2002). Analysis of longitudinal data Oxford, UK: Oxford University Press. [Google Scholar]
  19. Fernandez E, Schiaffino A, La Vecchia C, Borras JM, Nebot M, Salto E, … Segura A (1999). Age at starting smoking and number of cigarettes smoked in Catalonia, Spain. Preventive Medicine, 28(4), 361–366. doi: 10.1006/pmed.1998.0433 [DOI] [PubMed] [Google Scholar]
  20. Halperin AC, Smith SS, Heiligenstein E, Brown D, & Fleming MF (2010). Cigarette smoking and associated health risks among students at five universities. Nicotine & Tobacco Research, 12(2), 96–104. doi: 10.1093/ntr/ntp182 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hartos JL, Eitel P, Haynie DL, & Simons-Morton BG (2000). Can I Take the Car?: Relations Among Parenting Practices and Adolescent Problem-Driving Practices. Journal of Adolescent Research, 15(3), 352–367. doi: 10.1177/0743558400153003 [DOI] [Google Scholar]
  22. Hetherington EM, Clingempeel WG, Anderson ER, Deal JE, Hagan MS, Hollier EA, … Bennion LD. (1992). Coping with Marital Transitions: A Family Systems Perspective. Monographs of the Society for Research in Child Development, 57(2/3), i–238. doi: 10.2307/1166050 [DOI] [Google Scholar]
  23. Irwin DE, Stucky B, Langer MM, Thissen D, DeWitt EM, Lai J-S, … DeWalt DA (2010). An item response analysis of the pediatric PROMIS anxiety and depressive symptoms scales. Quality of life research : an international journal of quality of life aspects of treatment, care and rehabilitation, 19(4), 595–607. doi: 10.1007/s11136-010-9619-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Jackson KM, Sher KJ, & Schulenberg JE (2008). Conjoint Developmental Trajectories of Young Adult Substance Use. Alcoholism, Clinical and Experimental Research, 32(5), 723–737. doi: 10.1111/j.1530-0277.2008.00643.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE, & Miech RA (2015). Monitoring the Future: College Students & Adults Ages 19–55 (Vol. 2). Ann Arbor: University of Michigan. [Google Scholar]
  26. Keith DR, Hart CL, McNeil MP, Silver R, & Goodwin RD (2015). Frequent marijuana use, binge drinking and mental health problems among undergraduates. The American Journal on Addictions, 24(6), 499–506. doi: 10.1111/ajad.12201 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Khuder SA, Dayal HH, & Mutgi AB (1999). Age at smoking onset and its effect on smoking cessation. Addictive Behaviors, 24(5), 673–677. doi: 10.1016/s0306-4603(98)00113-0 [DOI] [PubMed] [Google Scholar]
  28. Kirst M, Mecredy G, Borland T, & Chaiton M (2014). Predictors of Substance Use Among Young Adults Transitioning Away from High School: A Narrative Review. Substance Use & Misuse, 49(13), 1795–1807. doi: 10.3109/10826084.2014.933240 [DOI] [PubMed] [Google Scholar]
  29. Lac A, & Crano WD (2009). Monitoring Matters: Meta-Analytic Review Reveals the Reliable Linkage of Parental Monitoring With Adolescent Marijuana Use. Perspectives on Psychological Science, 4(6), 578–586. doi: 10.1111/j.1745-6924.2009.01166.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Lando HA, Thai DT, Murray DM, Robinson LA, Jeffery RW, Sherwood NE, & Hennrikus DJ (1999). Age of initiation, smoking patterns, and risk in a population of working adults. Preventive Medicine, 29(6), 590–598. doi: 10.1006/pmed.1999.0590 [DOI] [PubMed] [Google Scholar]
  31. Lenk K, Rode P, Fabian L, Bernat D, Klein E, & Forster J (2012). Cigarette use among young adults: Comparisons between two-year college students, four-year college students, and those not in college. Journal of American college health : J of ACH, 60(4), 303–308. doi: 10.1080/07448481.2011.607481 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Little RJA, & Rubin DB (2002). Statistical analysis with missing data (2nd Ed.). Hoboken, NJ: Wiley. [Google Scholar]
  33. McCabe SE, Schulenberg JE, Johnston LD, O’Malley PM, Bachman JG, & Kloska DD (2005). Selection and socialization effects of fraternities and sororities on US college student substance use: a multi-cohort national longitudinal study. Addiction, 100(4), 512–524. doi: 10.1111/j.1360-0443.2005.01038.x [DOI] [PubMed] [Google Scholar]
  34. NCES. (2014). Percentage of 18- to 24-year-olds enrolled in degree-granting institutions, by level of institution and sex and race/ethnicity of student: 1967 through 2012 Retrieved February 22, 2016, from http://nces.ed.gov/programs/digest/d13/tables/dt13_302.60.asp
  35. Newcomb MD, & Bentler PM (1987). Changes in drug use from high school to young adulthood: Effects of living arrangement and current life pursuit. Journal of Applied Developmental Psychology, 8(3), 221–246. doi: 10.1016/0193-3973(87)90001-3 [DOI] [Google Scholar]
  36. Primack BA, Kim KH, Shensa A, Sidani JE, Barnett TE, & Switzer GE (2012). Tobacco, Marijuana, and Alcohol Use in University Students: A Cluster Analysis. Journal of American College Health, 60(5), 374–386. doi: 10.1080/07448481.2012.663840 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Sanem JR, Berg CJ, An LC, Kirch MA, & Lust KA (2009). Differences in Tobacco Use Among Two-Year and Four-Year College Students in Minnesota. Journal of American College Health, 58(2), 151–159. doi: 10.1080/07448480903221376 [DOI] [PubMed] [Google Scholar]
  38. Schulenberg JE, Bryant AL, & O’Malley PM (2004). Taking hold of some kind of life: How developmental tasks relate to trajectories of well-being during the transition to adulthood. Development and Psychopathology, 16(4), 1119–1140. doi: 10.1017/s0954579404040167 [DOI] [PubMed] [Google Scholar]
  39. Schulenberg JE, & Maggs JL (2002). A developmental perspective on alcohol use and heavy drinking during adolescence and the transition to young adulthood. Journal of Studies on Alcohol, Suppl14, 54–70. [DOI] [PubMed] [Google Scholar]
  40. Scott-Sheldon LAJ, Carey KB, & Carey MP (2008). Health Behavior and College Students: Does Greek Affiliation Matter? Journal of Behavioral Medicine, 31(1), 61–70. doi: 10.1007/s10865-007-9136-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Sidani JE, Shensa A, & Primack BA (2013). Substance and hookah use and living arrangement among fraternity and sorority members at US colleges and universities. Journal of community health, 38(2), 238–245. doi: 10.1007/s10900-012-9605-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Simons-Morton B, & Farhat T (2010). Recent Findings on Peer Group Influences on Adolescent Substance Use. The Journal of Primary Prevention, 31(4), 191–208. doi: 10.1007/s10935-010-0220-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Simons-Morton B, Haynie D, Liu D, Chaurasia A, Li K, & Hingson R (2016). 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, 77(1), 121–132. doi: doi:10.15288/jsad.2016.77.121 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Stone AL, Becker LG, Huber AM, & Catalano RF (2012). Review of risk and protective factors of substance use and problem use in emerging adulthood. Addictive Behaviors, 37(7), 747–775. doi: 10.1016/j.addbeh.2012.02.014 [DOI] [PubMed] [Google Scholar]
  45. Thompson AB, Tebes JK, & McKee SA (2015). Gender differences in age of smoking initiation and its association with health. Addiction Research & Theory, 23(5), 413–420. doi: 10.3109/16066359.2015.1022159 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. van Buuren S, & Groothuis-Oudshoorn K (2011). mice: Multivariate Imputation by Chained Equations in R. Journal of Statistical Software, 45(3), 67. doi: 10.18637/jss.v045.i03 [DOI] [Google Scholar]
  47. 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: 10.1016/j.addbeh.2012.06.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. VanKim NA, Laska MN, Ehlinger E, Lust K, & Story M (2010). Understanding young adult physical activity, alcohol and tobacco use in community colleges and 4-year post-secondary institutions: A cross-sectional analysis of epidemiological surveillance data. BMC Public Health, 10, 208-208. doi: 10.1186/1471-2458-10-208 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Wallace JM, Bachman JG, O’Malley PM, Johnston LD, Schulenberg JE, & Cooper SM (2002). Tobacco, alcohol, and illicit drug use: racial and ethnic differences among U.S. high school seniors, 1976–2000. Public Health Reports, 117(Suppl 1), S67–S75. [PMC free article] [PubMed] [Google Scholar]
  50. White HR, Labouvie EW, & Papadaratsakis V (2005). Changes in Substance use during the Transition to Adulthood: A Comparison of College Students and Their Noncollege Age Peers. Journal of Drug Issues, 35(2), 281–306. doi: 10.1177/002204260503500204 [DOI] [Google Scholar]
  51. White HR, McMorris BJ, Catalano RF, Fleming CB, Haggerty KP, & Abbott RD (2006). 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, 67(6), 810–822. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. White HR, Pandina RJ, & Chen P-H (2002). Developmental trajectories of cigarette use from early adolescence into young adulthood. Drug & Alcohol Dependence, 65(2), 167–178. doi: 10.1016/S0376-8716(01)00159-4 [DOI] [PubMed] [Google Scholar]

RESOURCES