This may be the author’s version of a work that was submitted/accepted for publication in the following source:
Plenty, Stephanie, Evans-Whipp, Tracy, Chan, Gary, Kelly, Adrian, Toum-bourou, John, Patton, George, Hemphill, Sheryl, & Smith, Rachel
(2019)
Predicting alcohol misuse among Australian 19-year-olds from adolescent drinking trajectories.
Substance Use and Misuse, 54(2), pp. 247-256.
This file was downloaded from: https://eprints.qut.edu.au/120956/
c
Consult author(s) regarding copyright matters
This work is covered by copyright. Unless the document is being made available under a Creative Commons Licence, you must assume that re-use is limited to personal use and that permission from the copyright owner must be obtained for all other uses. If the docu-ment is available under a Creative Commons License (or other specified license) then refer to the Licence for details of permitted re-use. It is a condition of access that users recog-nise and abide by the legal requirements associated with these rights. If you believe that this work infringes copyright please provide details by email to [email protected]
Notice: Please note that this document may not be the Version of Record (i.e. published version) of the work. Author manuscript versions (as Sub-mitted for peer review or as Accepted for publication after peer review) can be identified by an absence of publisher branding and/or typeset appear-ance. If there is any doubt, please refer to the published source.
Predicting alcohol misuse among Australian 19-year-olds from adolescent
drinking trajectories
Stephanie M. Plentyab*, Tracy J. Evans-Whipp cd Gary C. K. Chane, Adrian B. Kellyf, John W. Toumbouroucg, George C. Patton cd Sheryl A. Hemphillcdh, Rachel Smithcd
[IN PRESS – SUBSTANCE USE AND MISUSE]
aInstitute for Future Studies, 101 31 Stockholm, Sweden.
bSwedish Institute for Social Research, Stockholm University, Stockholm, 106 91, Sweden. cCentre for Adolescent Health, Murdoch Childrens Research Institute, Parkville, Victoria, 3052,
Australia.
dThe University of Melbourne Department of Paediatrics, Royal Children’s Hospital, Parkville,
Victoria, 3052, Australia.
eCentre for Youth Substance Abuse Research, The University of Queensland, Brisbane, Australia fSchool of Psychology and Counselling, Queensland University of Technology, Brisbane,
Australia
gCentre for Social and Early Emotional Development (SEED) and School of Psychology, Deakin
University, Geelong Waterfront Campus, Geelong Victoria, 3217, Australia.
hSchool of Psychology, Australian Catholic University, Melbourne, Victoria, Australia
THIS PAPER IS IN PRESS. PLEASE CITE THIS PAPER AS: Plenty, S. M., Evans-Whipp, T. J., Chan, G. C. K., Kelly, A. B., Toumbourou, J. W., Patton, G. C., Hemphill, S. A., & Smith, R. (2018). Predicting alcohol misuse at age 19 from adolescent drinking trajectories. Substance
*Corresponding author: Email: [email protected]; Phone: + 46 8 402 12 21; Address:
Holländargatan13, SE-101 31 Stockholm, Sweden.
Funding: The authors are grateful for the financial support of the National Institute on Drug
Abuse (R01-DA012140) for the International Youth Development Study. Data collection in Victoria, Australia was supported by three Australian Research Council Discovery Projects (DPO663371, DPO877359, and DP1095744) and an Australian National Health and Medical Research Council grant (NHMRC; project number 594793). The funding bodies had no role in the study design, procedures, analyses, interpretation of results or manuscript preparation.
Predicting alcohol misuse among Australian 19-year-olds from adolescent
drinking trajectories
Abstract
Background: Alcohol use in adolescence predicts future alcohol misuse.However, the extent to which different patterns of adolescent use present risk remains unclear.
Objectives: This study investigated how adolescent trajectories of alcohol consumption during the school years predict alcohol misuse at age 19 years.
Methods: Data were drawn from 707 students from Victoria, Australia, longitudinally followed for 7-years. Five alcohol use trajectories were identified based on frequency of alcohol use from Grade 6 (age 12 years) to Grade 11 (age 17 years). At age 19 years, participants completed measures indicating heavy episodic drinking (HED), dependency - Alcohol Use Disorders Identification Test (AUDIT) and social harms.
Results: At 19 years of age, 64% of participants reported HED, 42% high AUDIT scores (8+) and 23% social harms. Participants belonging to a steep escalator trajectory during adolescence had twice the odds at 19 years of age of high AUDIT scores and social harms, and three times greater odds of HED than participants whose alcohol use slowly increased. Stable moderate consumption was also associated with an increased risk of HED compared to slowly increasing use. Abstinence predicted a reduced likelihood of all forms of misuse at 19 years of age
compared to slowly increased alcohol use.
Conclusions: Trajectories of drinking frequency during adolescence predict alcohol misuse at age 19 years. Although rapid increasing use presents the greatest risk, even slowly increasing
drinking predicts increased risk compared to abstinence. The findings indicate that alcohol policies should recommend non-use and reduced frequency of use during adolescence.
Key words: alcohol misuse; trajectories; longitudinal; adolescence; young adulthood; heavy
1. Introduction
Alcohol misuse is a leading factor in the burden of disease and preventable deaths among adolescents and young adults worldwide (Mokdad et a., 2016; Rehm et al., 2009). The World Health Organisation estimates that approximately 6% of all deaths among 15 to 29 year olds are due to alcohol-related causes (2014). Although young people today are more likely to delay alcohol use and drink in smaller volumes compared to previous generations (Australian Institute of Health and Welfare, 2017; Livingston et al., 2016), adolescence is a time when many initiate alcohol use and frequency of use increases through the post-secondary school and young adult years (i.e. 18 to 25 years old) (Australian Institute of Health and Welfare, 2017; Substance Abuse and Mental Health Services Administration, 2016). This period is characterized by many legal and social transitions that may present risk for harmful alcohol consumption (Arnett, 2000, 2005). For Australian youth, the legal purchasing age for alcohol (18 years) coincides with the
completion of secondary school, increased independence and usually, the commencement of tertiary education or employment (Australian Bureau of Statistics, 2010). Given this pivotal life-period and the associated costs of alcohol-related harm, reducing alcohol misuse in young adults is a recognized priority for public health policy (Australian National Preventive Agency, 2013; Substance Abuse and Mental Health Services Administration, 2012; World Health Organisation, 2010; United Nations, 2012).
A key predictor of alcohol misuse in the post-secondary school years is prior alcohol consumption during the adolescent school years. Heavy and frequent drinking in adolescence are associated with an increased likelihood of heavier and problematic use, as well as alcohol-related social harms in young adulthood (Bonomo et al., 2004; Cable & Sacker, 2008; Jefferis et al.,
2005; Moore et al., 2009; Norström & Pape, 2012; Viner & Taylor, 2007). Alcohol misuse is also related to the age of initiation. Early initiation, particularly before the age of 14 years, increases the risk of future alcohol problems, alcohol-related injury, and dependence (Grant & Dawson, 1997; Hingson et al., 2006: Hingson & Zha, 2009; Pitkänen et al., 2008).
Much of the current research linking adolescent drinking with harmful alcohol use in young adulthood has used variable-centered single time-point measures to define adolescent consumption. However, such studies fail to identify sub-group differences in developmental patterns (trajectories) of adolescent alcohol consumption. Person-centered research utilizing latent growth mixture modeling (LGM) has identified subgroups with different growth patterns and age of initiation of alcohol use. These trajectories commonly include: abstainers, moderate users (with stable or slowly increasing use), early rapid and heavy users, as well as later onset rapid heavy users (reviewed in Chassin et al., 2013 and Sher et al., 2011). The different trajectories of alcohol consumption may arise because of different cognitive, experiential or biological reactions to early alcohol exposure (Agrawal et al., 2009; Wong et al., 2011; Yücel et al., 2007). They may also represent different developmental tendencies to engage in risky behaviors more generally (Wells et al., 2004). Accurately classifying early trajectories and identifying the associated risk for later misuse may have important benefits for early intervention programs aiming to minimize future alcohol use problems.
Prior trajectory studies have indicated that rapid increases or stable high alcohol use during adolescence signify a greater risk for future alcohol-related problems than trajectories defined by moderate use or abstinence (Colder et al., 2002; Chassin et al., 2002; Danielsson et al., 2010). However, there is an important need to replicate trajectory findings in contemporary data. Many trajectory studies are based on data from the 1980s and adolescent alcohol use has since
changed substantially (Livingston, 2008; 2014; Chan et al., 2016). General rates of youth drinking have declined internationally in recent decades, matched by increasing abstinence (Hallgren et al., 2012; Livingston, 2015). Despite this, a concerning proportion of youth remain heavy users and experience alcohol-related injuries (Australian Institute of Health and Welfare, 2017; Substance Abuse and Mental Health Services Administration, 2016). In addition, younger cohorts may increase drinking more rapidly after the age of 18 years than older cohorts (Jager et al., 2013). There have also been important generational changes in young adult social
development in recent decades. Many young people now delay establishing stable adult roles, such as economic independence and family formation (Arnett, 2007), which are associated with reductions in alcohol misuse (Temple et al., 1991). It is therefore important to understand how the developmental course of adolescent alcohol use predicts later alcohol misuse among more recent cohorts.
Research to date has also not resolved whether abstinence throughout adolescence holds fewer risks than other patterns of alcohol use for developing alcohol use problems. Previous trajectory studies addressing subsequent alcohol misuse tend to combine groups with low frequency use with those that abstain from alcohol (e.g., Danielsson et al., 2010). Therefore, the difference in likelihood of later alcohol problems between these two groups remains unclear. The outcomes of gradually increasing adolescent alcohol use are of particular interest in the context of harm minimization frameworks. Although abstinence through adolescence remains a primary strategy of harm minimization, the framework also acknowledges that many young people will experiment with alcohol and in such cases, advocates modest use (Lenton and Midford, 1996). Some harm minimization advocates and parents support the notion that gradual introduction of alcohol during adolescence, particularly in the home environment, will promote responsible use
and self-regulatory skills that protect against alcohol misuse at later ages (Jackson et al., 2012; Jones, 2016). Empirical evidence to test this viewpoint is needed.
There has also been little investigation of consequences for those who drink at high levels in early adolescence but then reduce their drinking. In addition, very few trajectory studies have controlled for a range of important covariates stemming from the individual, peer and family domains. Although growth in alcohol use during adolescence is a risk for alcohol misuse in itself, due to alcohol tolerance and alcohol-oriented behavioral norms, the potential role of individual, peer or family-related factors that shape the likelihood of future problematic use should be accounted for (French & Maclean, 2006; Lavikainen et al., 2011; McCambridge et al., 2011).
The current study
The current study aims to identify associations between alcohol use in early- and mid-adolescence and later alcohol misuse at 19 years of age in a contemporary Australian cohort. It extends previous research by examining if: (1) steep escalations in adolescent alcohol use are more predictive of later alcohol misuse; (2) drinking patterns characterized by an early age of initiation are at greater risk of later problems, and (3) slowly increasing alcohol use has similarly low risks for later alcohol use problems as abstinence. The opportunity to complete the current analysis arises from an earlier trajectory study addressing early determinants of adolescent alcohol use (Chan et al., 2013). Chan et al. examined the frequency of Australian adolescents’ (aged 12 -17 years) alcohol consumption and used Latent Class Growth Analysis (LCGA) to identify five trajectory groups of alcohol use: stable moderate drinkers, early high drinkers, slow increasers, steep escalators and non-users. The current study examines the associations between these adolescent alcohol use trajectories and different indicators of alcohol misuse at 19 years of age, adjusting for common risk factors that are associated with both early drinking patterns and
young adult alcohol misuse. Three indicators of adult alcohol misuse are examined: heavy episodic drinking, social harms and alcohol dependence symptoms. The following hypotheses are tested:
1. Consistent with previous research that highlights rapid growth in the frequency of alcohol use during adolescence as problematic (e.g., Colder, 2002; Danielsson et al., 2010), steep escalators are expected to show a greater likelihood of alcohol misuse at 19 years of age than slow increasers.
2. If early alcohol initiation is a key risk factor (e.g., Chassin et al., 2002; Danielsson et al., 2010), then high frequency of alcohol use prior to 14 years of age as represented by both the early high and stable moderate trajectories, will show greater likelihood of alcohol misuse at 19 years of age than slow increasers.
3. As suggested by some harm minimization frameworks, slow increasers are expected to show a similar or lower likelihood of alcohol misuse at 19 years of age when compared to
participants who abstain from alcohol throughout adolescence.
2. Method
2.1 Participants and procedure
The initial sample consisted of 927 Australian youth from the International Youth Development Study (IYDS), a longitudinal study on the development of substance use and associated problems (see McMorris et al., 2007 for details regarding recruitment). The sample was representative of the state of Victoria (McMorris et al., 2007). The participants were first assessed in 2003 at Grade 6 (mean age 12 years, range 11-13 years, from 100 state government, independent and catholic schools) and then reassessed annually until they were in Grade 11 (mean age 17 years, range 16-18 years) (with the exception of Grade 8). They were then
followed-up post-secondary school in 2010 (mean age 19 years, age range 17-20 years). For each wave of data collection, ethics approval was obtained from the relevant university and
educational authorities. For the adolescent waves, parents provided written consent for their child to participate. Students then provided assent and completed surveys at school. For the follow-up at 19 years of age, participants provided consent and completed an online survey.
Of the initial sample, 119 participants were excluded from the trajectory analysis due to missing data or questionable responses (refer to Chan et al., 2013 for detailed exclusion criteria for the trajectory analysis). Thirteen participants already aged 18 years at Grade 11 were
excluded from the current study. A further 86 participants who did not complete the follow-up at 19 years of age and two participants under 18 years of age were also excluded. This resulted in a final sample size of 707 (76% of the original sample). The final sample contained 45% males, mean age at follow-up was 19 years, range 18-21 years. Participants lost to follow-up were more likely to be male, have higher levels of Grade 11 antisocial behavior, and have lower socio-economic status. Thus, the results may under-represent disadvantaged youth, particularly males.
2.2 Measures
Unless otherwise stated, the measures of alcohol misuse and risk factors were adapted from the Communities That Care (CTC) youth survey (Arthur et al., 2002; Glaser et al., 2005).
2.2.1 Alcohol misuse at 19 years of age
Three aspects of adult alcohol misuse were examined: recent heavy episodic drinking, social harms and alcohol dependence symptoms.
Heavy Episodic Drinking (HED). Participants indicated how many times during the past
2 weeks they had drunk five or more alcoholic drinks in a row. Responses were coded as ‘None’ (0), ‘Once’ (1), ‘Twice’ (2) and ‘three or more times’ (3) (adapted from Johnston et al., 1998).
Social harms was based on a 9-item scale from the Victorian Adolescent Health Survey
(Hibbert et al., 1996) that addressed short-term social difficulties from alcohol consumption including: arguing with others, impaired work or study, feeling anxious or depressed, regretting sex, passing out or being asked to leave venues. Response options ranged from never, one or two
times to 40 or more times on an 8-point scale (Cronbach’s Alpha = .82). Due to skewed
responses, alcohol harms was scored ‘1’ if three or more harms were reported and ‘0’ if fewer harms were reported.
The Alcohol Use Disorders Identification Test (AUDIT) (Babor et al., 2001) screens for symptoms of alcohol dependence. It comprises 10 items addressing frequency and amount of drinking, as well as dependence symptoms and alcohol-related problems. Responses (scored from 0 to 4) were dichotomized and then summed using a cut-off of ≥ 8 to indicate hazardous alcohol use this. This cut-off indicates when individuals should be further evaluated by healthcare professionals for alcohol dependency (Babor et al., 2001).
2.2.2. Adolescent trajectories of alcohol use were based on the following item “In the past 30
days on how many occasions have you had more than just a few sips of an alcoholic beverage (like beer, wine or liquor/spirits)?” Response options ranged from never, one or two times to 40
or more times on an 8-point scale. The trajectories were identified using responses from Grade 6
to Grade 11 (excluding Grade 8) with latent class growth analysis. Details of model specification can be found in Chan et al., (2013). The five trajectories identified included: (1) Stable moderate drinkers who increased alcohol use from twice a month at Grade 6 to over 5 times a month from
Grade 7 and then remained stable (8%). (2) Early high drinkers who increased alcohol use rapidly from Grade 6 to Grade 7 and decreased gradually thereafter (3%). (3) Slow increasers comprised the normative group with a slow and steadily increasing pattern of use from less than once per month at Grade 6 to 3 times a month by Grade 11 (68%). (4) Steep escalators used alcohol less than once a month in Grade 6 but then rapidly increased consumption to over 8
times a month at Grade 11 (8%). (5) Non-users consistently reported no recent alcohol use (14%).
2.2.3. Control variables. To assess the unique associations between alcohol-use trajectories and
subsequent alcohol misuse, we controlled for several factors measured in Grade 11 known to predict alcohol misuse, as well as gender and family socioeconomic status (SES).
Peer drinking asked how many of the student’s friends had tried alcohol in the past year.
Response options ranged from none of my friends (0) to four of my friends (4) on a 5-point scale.
Antisocial behavior included 11 items asking participants how many times during the
past year (12 months) they had carried a weapon, stolen money or property, sold illegal drugs or stolen property, been suspended from school or arrested, physically attacked others, been drunk at school or appeared in court. Response options ranged on an 8-point scale from never, one or
two times to 40 or more times (Cronbach’s Alpha = 0.82). Scores were dichotomized to represent
one or more antisocial behaviors (72%) versus none (28%).
Parental attitudes favorable toward drug use was measured using four items that asked
participants about their parents’ approval regarding their use of alcohol, cigarettes or marijuana. Items were rated on a 4-point scale ranging from very wrong (1) to not wrong at all (4)
Poor family management asked participants if their parents tended to be aware of the
activities they engage in, know their whereabouts, inquire about where they have been and have clear family rules. Response options for the nine items ranged on a four-point scale from
definitely no (1) to definitely yes (4) (Cronbach’s Alpha = .82).
Depressive symptoms were measured with the Short Mood and Feelings Questionnaire
(Messer et al., 1995). Participants indicated how true 13 statements reflecting depressive symptoms during the last 30 days were for them. Response options were on a 3-point scale: not
true, sometimes true or true (Cronbach’s Alpha = .93).
Family attachment comprised of four items assessing participants’ attachment to their
mother and/or father (e.g. “Do you feel very close to your mother?”). Response options ranged on a 4-point scale from definitely no (1) to definitely yes (4) (Cronbach’s Alpha = .77).
Family Socioeconomic Status (SES) was an index combining parental educational
attainment and family income. This information was provided in a telephone interview with a parent/guardian in the first wave (Grade 5) of the study (described in Evans-Whipp et al., 2007). Parental educational attainment was measured as the average of maternal and paternal education according to: (1) non-completion of secondary school, (2) completion of secondary school, and (3) completion of post-secondary education. The mean score of parental education and family income (scaled from 1 to 3) formed a continuous indicator, with higher scores indicating higher family SES.
2.3 Statistical Analysis
Statistical analyses were performed with STATA 13 (Stata Corporation, 2013). Bivariate relationships between the alcohol use trajectories and control variables with alcohol misuse were examined using chi-square tests (for categorical predictors) and independent t-tests
(for continuous predictors). Three regression models then examined the associations between the alcohol use trajectories and each alcohol misuse outcome (at 19 years of age). Class
memberships were multiply imputed from the posterior probabilities obtained by the LCGA model (Goodman, 2007) and missing data in predictors were also imputed. One hundred datasets were imputed and models were estimated using Rubin’s (2009) multiple imputation technique. Binary logistic regression analyses were performed for AUDIT scores and social harms, and ordinal logistic regression analyses were performed for HED. Brant’s test (Brant, 1990) conducted on the first imputed dataset indicated that the parallel odd assumption required for ordinal logistic models was not violated. All models used the slow increasing trajectory as the reference (normative) group and adjusted for all control variables.
3. Results
3.1 Sample characteristics
Overall, 42% of participants reported high AUDIT scores, 23% reported social harms, and 64% reported HED. Table 1 shows the frequencies of alcohol misuse at 19 years of age within each drinking trajectory and according to each control variable. Trajectory membership was significantly related to all three types of alcohol misuse. The stable moderate and steep escalator trajectories both showed a substantially larger proportion of individuals reporting high AUDIT scores, social harms and HED. Surprisingly, the early high drinkers showed similar rates to the slow increasers. The non-users trajectory consistently showed the lowest rates of misuse across all three outcomes, even compared to the slow increasers.
3.2 Predicting Alcohol Misuse at 19 years of age from Adolescent Trajectories
Results for the three regression models are shown in Table 2. Participants in the non-users trajectory showed a reduced risk of high AUDIT scores, compared to those in the slow increaser trajectory. In contrast, participants in the steep escalator trajectory showed an increased likelihood of high AUDIT scores. Peer drinking and poor family management were also
associated with an elevated risk of high AUDIT scores.
Participants in the non-users trajectory were less likely to report social harms at 19 years of age compared to those in the slow increaser trajectory. The steep escalator trajectory showed odds of social harms that were approximately twice as high as the slow increaser trajectory. Stronger parental attitudes favorable toward drug use, depressive symptoms, poor family management and antisocial behavior were also significantly associated with an increased likelihood of social harms.
In relation to HED, participants in the stable moderate and the steep escalator trajectories had greater odds of recent excessive drinking than participants in the slow increaser trajectory. Those in the non-user trajectory showed reduced odds of HED relative to the slow increaser trajectory. Peer drinking, parental attitudes favorable toward drug use and antisocial behavior were each associated with a greater likelihood of HED at 19 years of age.
Sensitivity analyses were performed that excluded participants who reported no alcohol use at 19 years of age (n = 85), reducing the non-user trajectory by a third (from 99 to 62 participants). Of participants who consumed alcohol during the follow-up year, the non-user trajectory did not show a significantly reduced risk of alcohol misuse at 19 years of age
compared to the slow increaser trajectory (AUDIT OR = 0.56, 95% CI: .28-1.10; Social harms OR = 0.52, 95% CI: .19-1.39; HED OR = 0.81, 95% CI: .48-1.38).
[Table 2 here]
4. Discussion
This study of Australian youth showed that drinking trajectories in adolescence are associated with multiple indicators of alcohol misuse at 19 years of age. The first hypothesis that the steep escalator trajectory would predict higher levels of adult alcohol misuse was confirmed. This trajectory was characterized by a rapid increase in drinking frequency from twelve years of age. This is consistent with earlier research that has found escalations in heavy drinking
frequency across this age span to be associated with increased risk of alcohol abuse and
dependency in early adulthood (Chassin et al., 2002; Hill et al., 2000). The current findings also demonstrate that after taking into consideration the possibility that steep escalating drinking may promote involvement with drinking peers and other risk factors, a significant association
between steep escalation and later alcohol misuse was still observed. Although the steep
escalating trajectory represents a minority of adolescent drinkers (8 percent in this sample), they appear to be an important risk group with respect to developing harmful alcohol use across the transition to young adulthood.
Contrary to the second hypothesis, we did not find that trajectories with higher frequency of alcohol use in early adolescence (prior to age 14 years, i.e. early high and stable moderate) consistently predicted more adverse outcomes than the slow increaser trajectory. The only significant effect was for the stable moderate drinkers who had an almost 2-fold increased odds of HED at 19 years of age. Prior studies linking early onset of alcohol use with later problems might have missed important transitions in some groups if they examined mean population
trends (e.g. Hingson et al., 2006). Some of the early drinkers in our sample may have learned to regulate their earlier alcohol use to reduce risk of later alcohol misuse; the early high group by reducing alcohol use after a period of early experimentation and the stable moderate group by not increasing their drinking frequency beyond 5 times a month.
In contrast to the third hypothesis, participants who abstained from alcohol use during adolescence showed a reduced risk of alcohol misuse compared to the slow increasing trajectory group. Sensitivity analyses indicated that the protective effect of abstinence in adolescence was partly due to continued abstinence in early adulthood. In addition, there was no evidence that abstinent adolescents were less able to regulate their drinking upon initiation after 18 years of age than other groups. The finding that a slowly increasing frequency of alcohol use across adolescence represented a common behavioral pattern in our Australian sample, yet was predictive of increased alcohol problems at 19 years of age, raises questions as to the utility of some alcohol harm minimization approaches. The current results confirm that alcohol abstinence
throughout adolescence is a valuable public health target (National Health and Medical Research Council, 2009; Toumbourou et al., 2003; Windle & Windle, 2005). Abstinence provides youth with a clear message, opportunities to establish social routines that are independent of alcohol and can also minimize the normative status of alcohol use across the broader peer group. However, given that abstinence may not be embraced by all adolescents, regular, brief
(Substance Abuse and Mental Health Services Administration, 2012; World Health Organisation, 2010) and targeted interventions (Conrod et al., 2008) to reduce levels of consumption remain important.
Although longitudinal studies are important in determining temporal ordering, future research could test potential causal processes more clearly. For example, early or steep escalating alcohol use trajectories may cause changes in neuroadaptation, epigenetic effects (Wong et al., 2011) or social identity (Mason & Spoth, 2011). A second limitation is the possibility of confounding due to unmeasured factors. For example, temperament characteristics reflecting behavioral disinhibition such as impulsivity (Acton, 2003) or genetics (Sartor et al., 2009; Ystrom et al., 2014) may be important. A third limitation regards participation selectivity. Although the initial recruitment and retention levels were high (above 70%), extreme patterns of alcohol use may have been underrepresented. In addition, the small size of the stable moderate and steep escalating trajectories may have limited power in detecting effects on alcohol misuse. Finally, although repeated measures were used to chart adolescent drinking patterns, a single time point was used for the outcome measures. Since discontinuity in drinking behaviors has been observed through to mid-adulthood (Jefferis et al., 2005; Temple & Fillmore, 1985), further follow up in later stages of adulthood would be beneficial. This is particularly important given that declines in Australian young adult drinking rates (below 40 years of age) (Australian Institute of Health and Welfare, 2017; Livingston et al. 2015) have occurred since our outcomes were measured in 2010. Thus, the extent to which individuals in the steep escalator or stable moderate trajectories remain at greater risk of alcohol misuse throughout their 20s in the contemporary drinking cultural context is of key interest.
4.2. Conclusions
This study demonstrates in an Australian sample that distinct trajectories of alcohol use during adolescence are predictive of alcohol misuse at 19 years of age. The findings have potentially important implications for future research and practice. Trajectories characterized by
rapidly escalating adolescent alcohol use are confirmed as a potential early intervention target due to their heightened risk of dependence and other alcohol problems. Furthermore, the findings extend current knowledge by confirming the public health benefits of promoting alcohol
abstinence during adolescence, even in comparison to slowly increasing use, which may be supported by some harm minimization advocates.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
20
References
Acton, G. S. (2003). Measurement of impulsivity in a hierarchical model of personality traits: Implications for substance use. Substance Use and Misuse, 38, 67-83.
Agrawal, A., Sartor, C. E., Lynskey, M. T., Grant, J. D., Pergadia, M. L., Grucza, R., ... & Heath, A. C. (2009). Evidence for an interaction between age at first drink and genetic influences on DSM-IV alcohol dependence symptoms. Alcoholism: Clinical and
Experimental Research 33(12): 2047-2056.
Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teends through the twenties. American Psychologist, 55, 469-480.
Arnett, J. J. (2005). The developmental context of substance use in emerging adulthood.
Journal of Drug Issues, 35, 235-253.
Arnett, J. J. (2007). Emerging Adulthood: What Is It, and What Is It Good For? Child
Development Perspectives, 1(2), 68-73.
Arthur, M. W., Hawkins, J. D., Pollard, J. A., Catalano, R. F., & Baglioni, A. J. (2002). Measuring risk and protective factors for substance use, delinquency, and other adolescent problem behaviors: The Communities that Care Youth Survey. Evaluation
Review, 26, 575-601.
Australian Bureau of Statistics. (2010). Are Young People Learning of Earning? Canberra, Australia.
http://www.abs.gov.au/AUSSTATS/[email protected]/Lookup/4102.0Main+Features40Mar+ 2010
Australian Institute of Health and Welfare (2017). National Drug Strategy Household Survey 2016: detailed findings. Drug Statistics series no. 31. Cat. no. PHE 214. Canberra: AIHW.
21 Australian National Preventive Health Agency. (2013). State of Preventive Health 2013.
Report to the Australian Government Minister for Health. Canberra: ANPHA. Babor, T. F., Higgins-Biddle, J. C., Saunders, J. B., & Monteiro, M. G. (2001). A U D I T:
The Alcohol Use Disorders Identification Test Guidelines for Use in Primary Care: Department of Mental Health and Substance Dependence, World Health Organization. Bonomo, Y. A., Bowes, G., Coffey, C., Carlin, J. B., & Patton, G. C. (2004). Teenage
drinking and the onset of alcohol dependence: a cohort study over seven years.
Addiction, 99, 1520-1528. doi: ADD846 [pii]10.1111/j.1360-0443.2004.00846.x
Brant, R. (1990). Assessing proportionality in the proportional odds model for ordinal logistic regression. Biometrics, 46, 1171-1178.
Cable, N., & Sacker, A. (2008). Typologies of alcohol consumption in adolescence: Predictors and adult outcomes. Alcohol and Alcoholism, 43, 81-90. doi: 10.1093/alcalc/agm146
Chan, G. C. K., Kelly, A. B., Toumbourou, J. W., Hemphill, S. A., Young, R. M., Haynes, M. A., & Catalano, R. F. (2013). Predicting steep escalations in alcohol use over the teenage years: age-related variations in key social influences. Addiction, 108, 1924– 1932.
Chan, G. C., Leung, J. K., Quinn, C., Connor, J. P., Hides, L., Gullo, M. J., ... & Hall, W. D. (2016). Trend in alcohol use in Australia over 13 years: has there been a trend reversal?. BMC Public Health, 16(1), 1070.
Chassin, L., Pitts, S. C., & Prost, J. (2002). Binge drinking trajectories from adolescence to emerging adulthood in a high-risk sample: predictors and substance abuse outcomes.
Journal Of Consulting and Clinical Psychology, 70, 67-78.
Chassin, L., Shera, K. J., Hussonga, A., & Currana, P. (2013). The developmental
22 future directions. Development and Psychopathology, 25, 1567-1584. doi:
http://dx.doi.org.ezp.lib.unimelb.edu.au/10.1017/S0954579413000771
Colder, C. R., Campbell, R. T., Ruel, E., Richardson, J. L., & Flay, B. R. (2002). A finite mixture model of growth trajectories of adolescent alcohol use: predictors and consequences. Journal of Consulting and Clinical Psychology, 976.
Conrod, P. J., Castellanos, N., & Mackie, C. (2008). Personality-targeted interventions delay the growth of adolescent drinking and binge drinking. Journal of Child Psychology and Psychiatry, 49, 181-190. doi: 10.1111/j.1469-7610.2007.01826.
Danielsson, A.K., Wennberg, P., Tengstrom, A., & Romelsjo, A. (2010). Short
Communication: Adolescent alcohol use trajectories: Predictors and subsequent problems. Addictive Behaviors, 35, 848-852.
Evans-Whipp, T. J., Bond, L., Toumbourou, J. W., & Catalano, R. F. (2007). School, Parent, and Student Perspectives of School Drug Policies. Journal of School Health, 77, 138-146.
French, M. T., & Maclean, J. C. (2006). Underage alcohol use, delinquency, and criminal activity. Health Economics, 15, 1261-1281.
Glaser, R. R., Van Horn, M. L., Arthur, M. W., Hawkins, J. D., & Catalano, R. F. (2005). Measurement properties of the Communities That Care Youth Survey across demographic groups. Journal of Quantitative Criminology, 21, 73-102.
Goodman, L. A. (2007). On the assignment of individuals to latent classes. Sociological
Methodology, 37, 1-22.
Grant, B. F., & Dawson, D. A. (1997). Age at onset of alcohol use and its association with DSM-IV alcohol abuse and dependence: results from the national longitudinal alcohol epidemiologic survey. Journal of Substance Abuse, 9, 103-110.
23
Hallgren, M., Leifman, H., & Andréasson, S. (2012). Drinking less but greater harm: could polarized drinking habits explain the divergence between alcohol consumption and harms among youth? Alcohol and Alcoholism, 47(5), 581-590.
Hibbert, M., Caust, J., Patton, G. C., Rosier, M., & Bowes, G. (1996). The health of young people in Victoria: Adolescent Health Survey. Melbourne, Australia: Centre for Adolescent Health Monograph.
Hill, K. G., White, H. R., Chung, I.-J., Hawkins, J. D., & Catalano, R. F. (2000). Early Adult Outcomes of Adolescent Binge Drinking: Person- and Variable-Centered Analyses of Binge Drinking Trajectories. Alcoholism: Clinical and Experimental Research, 24, 892-901.
Hingson, R. W., Heeren, T., & Winter, M. R. (2006). Age at drinking onset and alcohol dependence: age at onset, duration, and severity. Archives of Pediatrics and
Adolescent Medicine, 160, 739-746.
Hingson, R. W., & Zha, W. (2009). Age of Drinking Onset, Alcohol Use Disorders, Frequent Heavy Drinking, and Unintentionally Injuring Oneself and Others After Drinking.
Pediatrics, 123, 1477-1484. doi: 10.1542/peds.2008-2176
Jackson, C., Ennett, S. T., Dickinson, D. M., & Bowling, J. M. (2012). Letting children sip: Understanding why parents allow alcohol use by elementary school-aged children.
Archives of Pediatrics & Adolescent Medicine, 166, 1053–1057.
Jager, J., Schulenberg, J. E., O'Malley, P. M., & Bachman, J. G. (2013). Historical variation in drug use trajectories across the transition to adulthood: the trend toward lower intercepts and steeper, ascending slopes. Development and Psychopathology, 25(02), 527-543.
Jefferis, B. J. M. H., Power, C., & Manor, O. (2005). Adolescent drinking level and adult binge drinking in a national birth cohort. Addiction, 100, 543-549.
24 Johnston, L. D., O'Malley, P. M., & Bachman, J. G. (1998). National survey results on drug
use from the Monitoring the Future study, 1975-1997. Vol. I. Secondary school students. Rockville, MD: National Institute on Drug Abuse.
Jones, S. C. (2016). Parental provision of alcohol: A TPB-framed review of the literature.
Health Promotion International, 31(3), 562-571.
Lavikainen, H., Salmi, V., Aaltonen, M., & Lintonen, T. (2011). Alcohol-related harms and risk behaviours among adolescents: Does drinking style matter. Journal of Substance
Use, 16, 243-255. doi: 10.3109/14659891.2010.499492
Lenton, S. & Midford, R. (1996). Clarifying 'harm reduction'? Drug and Alcohol Review
15(4), 411-413.
Livingston, M. (2008). Recent trends in risky alcohol consumption and related harm among young people in Victoria, Australia. Australian and New Zealand Journal of Public
Health, 32, 266-271. doi: doi:10.1111/j.1753-6405.2008.00227.x
Livingston, M. (2014). Trends in non-drinking among Australian adolescents. Addiction, 109, 922-929. doi: 10.1111/add.12524
Livingston, M. (2015). Understanding recent trends in Australian alcohol consumption. Canberra, Australia: Foundation for Alcohol Research and Education.
Livingston, M., Raninen, J., Slade, T., Swift, W., Lloyd, B., & Dietze, P. (2016).
Understanding trends in Australian alcohol consumption—an age–period–cohort model. Addiction, 111(9), 1590-1598.
Mason, W. A., & Spoth, R. L. (2011). Longitudinal Associations of Alcohol Involvement with Subjective Well-Being in Adolescence and Prediction to Alcohol Problems in Early Adulthood. Journal of Youth and Adolescence, 40, 1215-1224. doi:
25 McCambridge, J., McAlaney, J., & Rowe, R. (2011). Adult consequences of late adolescent
alcohol consumption: a systematic review of cohort studies. PLoS Med, 8(2), e1000413.
McMorris, B. J., Hemphill, S. A., Toumbourou, J. W., Catalano, R. F., & Patton, G. C. (2007). Prevalence of substance use and delinquent behavior in adolescents from Victoria, Australia and Washington State, United States. Health Education and
Behavior, 34, 634-650.
Messer, S. C., Angold, A., Costello, E. J., & Loeber, R. (1995). Development of a short questionnaire for use in epidemiological studies of depression in children and adolescents: Factor composition and structure across development. International
Journal of Methods in Psychiatric Research, 5, 251-262.
Mokdad, A. H., Forouzanfar, M. H., Daoud, F., Mokdad, A. A., El Bcheraoui, C., Moradi-Lakeh, M., . . . Murray, C. J. L. (2016). Global burden of diseases, injuries, and risk factors for young people's health during 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. The Lancet, 387(10036), 2383-2401. Moore, E., Coffey, C., Carlin, J. B., Alati R, & Patton, G. C. (2009). Assessing alcohol
guidelines in teenagers: results from a 10-year prospective study. Australian And New
Zealand Journal Of Public Health, 33, 154-159.
National Health and Medical Research Council. (2009). Australian Guidelines to Reduce Health Risks from Drinking Alcohol. Canberra: Commonwealth of Australia. Norström, T., & Pape, H. (2012). Associations Between Adolescent Heavy Drinking and
Problem Drinking in Early Adulthood: Implications for Prevention. Journal of Studies
26 Pitkänen, T., Kokko, K., Lyyra, A.L., & Pulkkinen, L. (2008). A developmental approach to
alcohol drinking behaviour in adulthood: a follow-up study from age 8 to age 42.
Addiction, 103, 48-68.
Rehm, J., Mathers, C., Popova, S., Thavorncharoensap, M., Teerawattananon, Y., & Patra, J. (2009). Series: Global burden of disease and injury and economic cost attributable to alcohol use and alcohol-use disorders. The Lancet, 373, 2223-2233. doi:
10.1016/s0140-6736(09)60746-7
Rubin, D. B. (2009). Multiple imputation for nonresponse in surveys. New York, John Wiley & Sons.
Sartor, C. E., Lynskey, M. T., Bucholz, K. K., Madden, P. A. F., Martin, N. G., & Heath, A. C. (2009). Timing of first alcohol use and alcohol dependence: evidence of common genetic influences. Addiction, 104, 1512-1518. doi:
10.1111/j.1360-0443.2009.02648.x
Sher, K. J., Jackson, K. M., & Steinley, D. (2011). Alcohol use trajectories and the ubiquitous cat's cradle: Cause for concern? Journal of Abnormal Psychology, 120, 322-335. doi: 10.1037/a0021813
Stata Corporation. (2013). Stata Statistical Software: Release 11: College Station, TX: StataCorp LP.
Substance Abuse and Mental Health Services Administration. (2012). Report to Congress on the Prevention and Reduction of Underage Drinking. Washington, DC: U.S.
Department of Health and Human Services.
Substance Abuse and Mental Health Services Administration. (2016). Key substance use and mental health indicators in the United States: Results from the 2015 National Survey on Drug Use and Health:Rockville, MD: Substance Abuse and Mental Health Services Administration
27 Temple, M. T., & Fillmore, K. M. (1985). The Variability of Drinking Patterns and Problems
among Young Men, Age 16-31: A Longitudinal Study. International Journal of the
Addictions, 20, 1595.
Temple, M. T., Fillmore, K. M., Hartka, E., Johnstone, B., Leino, E. V., & Motoyoshi, M. (1991). The Collaborative Alcohol-Related Longitudinal Project. A meta-analysis of change in marital and employment status as predictors of alcohol consumption on a typical occasion. British Journal of Addiction, 86, 1269-1281.
Toumbourou, J. W., Williams, I. R., Snow, P. C., & White, V. M. (2003). Adolescent
alcohol-use trajectories in the transition from high school. Drug & Alcohol Review, 22, 111-116.
United Nations (2012). Political Declaration of the High-level Meeting of the General
Assembly on the Prevention and Control of Non-communicable Diseases, New York, USA September 2011.
http://www.who.int/nmh/events/un_ncd_summit2011/political_declaration_en.pdf?ua =1
Viner, R. M., & Taylor, B. (2007). Adult outcomes of binge drinking in adolescence: findings from a UK national birth cohort. Journal Of Epidemiology and Community Health, 61, 902-907.
Wells, J. E., Horwood, L. J., & Fergusson, D. M. (2004). Drinking patterns in mid-adolescence and psychosocial outcomes in late mid-adolescence and early adulthood.
Addiction, 99, 1529-1541.
Windle, M., & Windle, R. C. (2005). Alcohol consumption and its consequences among adolescents and young adults. Recent Developments in Alcoholism, 17, 67-83. Wong, C. C. Y., Mill, J., & Fernandes, C. (2011). Drugs and addiction: an introduction to
28 World Health Organisation. (2010). Global strategy to reduce the harmful use of alcohol.
Geneva: WHO.
World Health Organisation. (2014). Global Status Report on Alcohol and Health 2014. Geneva: WHO
Ystrom, E., Kendler, K. S., & Reichborn-Kjennerud, T. (2014). Early age of alcohol initiation is not the cause of alcohol use disorders in adulthood, but is a major indicator of genetic risk. A population-based twin study. Addiction, 109, 1824-1832. doi: 10.1111/add.12620.
Yücel, M., Lubman, D. I., Solowij, N., & Brewer, W. J. (2007). Understanding drug
addiction: A neuropsychological perspective. Australian and New Zealand Journal of
29 Table 1. Age 19 alcohol misuse outcomes by key analysis variables (n = 707).
AUDIT Social harm HEDa
Low High Low High None Any
n (%) n (%) χ2 n (%) n (%) χ2 n (%) n (%) χ2
Whole sample 398 (58.4) 284 (41.6) 528 (77.2) 156 (22.8) 257 (36.4) 449 (63.6)
Alcohol use trajectory 56.0*** 43.8*** 57.7***
Slow increasers (68%) 274 (59.1) 190 (41.0) 365 (78.5) 100 (21.5) 170 (35.5) 309 (64.5) Non-users (14%) 79 (85.0) 14 (15.0) 88 (94.6) 5 (5.4) 64 (65.3) 34 (34.7) Stable moderate (8%) 19 (33.3) 38 (66.7) 32 (56.1) 25 (43.9) 11 (20.4) 43 (79.6) Early high (3%) 9 (56.2) 7 (43.8) 13 (81.2) 3 (18.8) 6 (31.6) 13 (68.4) Steep escalators (8%) 19 (33.3) 38 (66.7) 30 (56.6) 23 (43.4) 6 (10.7) 50 (89.3) Gender 7.4** 3.0 1.5 Male 160 (52.6) 144 (47.4) 226 (74.1) 79 (25.9) 107 (34.0) 208 (66.0) Female 238 (63.0) 140 (37.0) 302 (79.7) 77 (20.3) 150 (38.4) 241 (61.6) Antisocial behavior 20.9*** 32.2*** 18.8*** No 297 (63.7) 169 (36.3) 385 (82.4) 82 (17.6) 201 (41.5) 283 (58.5) Yes 73 (43.5) 95 (56.6) 101 (60.1) 67 (39.9) 40 (23.3) 132 (76.7) M (SD) M (SD) t M (SD) M (SD) t M (SD) M (SD) t Peer drinking 3.73 (1.58) 4.48 (1.09) 6.6*** 3.92 (1.51) 4.45 (1.15) 4.0*** 3.58 (1.62) 4.29 (1.27) 6.2*** Family attachment 2.89 (0.68) 2.87 (0.64) 0.4 2.92 (0.68) 2.79 (0.61) 2.0* 2.90 (0.70) 2.89 (0.64) 0.2 Poor family management 1.92 (0.51) 2.13 (0.52) 5.1*** 1.94 (0.52) 2.21 (0.50) 5.6*** 1.90 (0.54) 2.06 (0.51) 3.7*** Parental attitudes favorable toward drug
use
1.79 (0.64) 2.07 (0.67) 5.4*** 1.83 (0.64) 2.18 (0.68) 5.8*** 1.70 (0.65) 2.02 (0.64) 6.0*** Depressive symptoms 1.62 (0.52) 1.65 (0.54) 0.5 1.59 (0.51) 1.77 (0.56) 3.6*** 1.62 (0.54) 1.64 (0.52) 0.4 Family socio-economic status 1.96 (0.49) 1.95 (0.49) 0.2 1.96 (0.50) 1.94 (0.49) 0.4 1.95 (0.51) 1.96 (0.49) 0.1
30 Table 2. Adjusted logistic regression analyses predicting age 19 alcohol misuse from adolescent alcohol use trajectories
AUDITa Social harmsa HEDb
Alcohol use trajectory OR 95% CI OR 95% CI OR 95% CI
Slow increasers Ref. - Ref. - Ref. -
Non-users 0.40** (0.21, 0.76) 0.37* (0.14, 0.96) 0.51** (0.32, 0.83) Stable moderate 1.82+ (0.95, 3.48) 1.62 (0.85, 3.09) 2.01* (1.13, 3.57) Early high .89 (0.31, 2.53) 0.71 (0.19, 2.69) 1.47 (0.59, 3.69) Steep escalators 2.08* (1.09, 3.94) 2.12* (1.12, 4.01) 2.93*** (1.68, 5.09) Control variables Peer drinking 1.32*** (1.16, 1.51) 1.13 (0.96, 1.33) 1.22*** (1.10, 1.37) Gender – Female 0.76 (0.53, 1.07) 0.79 (0.52, 1.19) 0.81 (0.60, 1.09) Family attachment 1.15 (0.84, 1.54) 1.01 (0.72, 1.42) 1.14 (0.88, 1.48)
Parental attitudes favorable toward drug use 1.29 (0.96, 1.72) 1.51* (1.08, 2.11) 1.62*** (1.26, 2.09)
Depressive symptoms 0.95 (0.67, 1.34) 1.58* (1.08, 2.32) 0.88 (0.65, 1.19)
Poor family management 1.62* (1.09, 2.41) 1.62* (1.03, 2.55) 1.24 (0.89, 1.72)
Any antisocial behavior 1.37 (0.91, 2.06) 1.72* (1.10, 2.67) 1.62** (1.14, 2.33)
Family socio-economic status 1.04 (0.74, 1.47) 1.04 (0.69, 1.56) 1.18 (0.88, 1.57)
Note. Reference group = slow increaser trajectory; *p < .05; **p < .01; ***p < .001, + p < .08; OR = adjusted odds ratio; a logistic regression; b ordinal logistic regression.
