Adolescent substance use and educational attainment: An integrative data analysis comparing cannabis and alcohol from three Australasian cohorts
Introduction
Successfully completing high school and attaining a university degree are critical developmental milestones linked to better health (Cutler and Lleras-Muney, 2010) and greater economic productivity (US Bureau of Labor Statistics, 2014). Alcohol and cannabis are commonly used by young people in the school-age years. Worldwide, 34% of 15–19 year olds are current drinkers (World Health Organization, 2014); European estimates suggest 12% of 15–16 year olds are past-month cannabis users although prevalence levels vary considerably between countries (European Monitoring Centre for Drugs and Drug Addiction, 2014). Adolescence may also be a vulnerable developmental period for the neurocognitive effects of substance use (Lisdahl et al., 2013). Heavy drinking and cannabis use have been linked to changes in central nervous system (CNS) structure and function in otherwise healthy adolescents (Lisdahl et al., 2013). Given the extent of exposure, the association of alcohol and cannabis use with subsequent levels of educational attainment is of increasing interest.
Research into the effects of cannabis has produced generally consistent findings to suggest that early use reduces the likelihood of progressing further in formal education (Macleod et al., 2004, Townsend et al., 2007). Typically, these associations attenuate but remain significant after controlling for potential confounding factors (e.g., socioeconomic status, family structure, peer behaviour; Esch et al., 2014, Macleod et al., 2004). The picture is less clear in relation to alcohol use. Cross-sectional studies provide some evidence of an association between alcohol and educational attainment (Townsend et al., 2007), but the evidence from longitudinal studies is equivocal. Some have found little linkage between adolescent alcohol use and low school commitment (Hemphill et al., 2014), years in education (Arria et al., 2013), and academic failure (Hemphill et al., 2014), after adjustment for potential confounders. Others have found early alcohol use is weakly associated with lower school grades (Crosnoe et al., 2012) and future educational achievement (Latvala et al., 2014), particularly for males (Balsa et al., 2011).
Two issues emerge in the literature. First, evidence of the effects of adolescent alcohol use on educational attainment is equivocal (McCambridge et al., 2011). Specifically, questions remain about the extent to which the apparent effects of adolescent drinking might be due to potential confounding factors not adequately controlled for in studies to date (McCambridge et al., 2011). Second, the relative contributions of cannabis or alcohol use to explaining non-progression with formal education have not been investigated.
We address these issues through the integration of data from three longitudinal studies from Australia and New Zealand: the Australian Temperament Study (ATP; Vassallo and Sanson, 2013), the Christchurch Health and Development Study (CHDS; Fergusson and Horwood, 2001), and the Victorian Adolescent Health Cohort Study (VAHCS; Patton et al., 2007). We integrated participant-level data rather than using the standard meta-analytic approach of combining study-level estimates. This approach increases sample size and statistical precision to investigate less common patterns of substance use (such as frequent use at a young age), provides the opportunity to include a wide range of potential confounding factors, and augments our ability to generalise findings to the region and internationally more realistically than is possible for any individual study (Curran and Hussong, 2009, Hofer and Piccinin, 2009). We build on earlier work which found that adolescent cannabis use was negatively associated with attaining secondary school and tertiary qualifications (Horwood et al., 2010, Silins et al., 2014) and extend the analysis to examine the relative contributions of cannabis and alcohol use to educational outcomes.
We examined the extent to which adolescent cannabis or alcohol use was associated with non-progression with formal education using data from three Australasian cohort studies. Specifically, we: (1) investigated the association between both frequency of cannabis use and frequency of alcohol use prior to age 17 and high school non-completion, university non-enrolment, and degree non-attainment by age 25 in each study and in combined data; (2) examined two other patterns of alcohol use (amount consumed and number of alcohol-related problems) prior to age 17; (3) adjusted the associations for potential confounders drawn from similar domains across studies; and (4) estimated the proportion of educational non-involvement attributable to adolescent cannabis or alcohol use if causality is assumed. The study has approval from the University of New South Wales Human Research Ethics Committee.
Section snippets
Design and participants
Integrative analyses were developed across the three studies (Hutchinson et al., 2015): ATP (Vassallo and Sanson, 2013), CHDS (Fergusson and Horwood, 2001), and VAHCS (Patton et al., 2007). Additional information about the longitudinal cohorts is provided in Supplementary Material 1. Analyses were based on data from these studies assessed between ages 13 and 25. The number of participants varied by analysis (from 2179 to 3678).
Measures
We assessed three educational outcomes in young people which reflect
Associations between cannabis use, alcohol use and educational attainment
Table 2 shows the associations of maximum frequency of cannabis use and alcohol use, before age 17, classified in three levels (never, <weekly, weekly+) with the three measures of educational attainment. The associations are reported without covariate adjustment, both separately for each study and for the combined data set:
For cannabis there were clear (p < 0.001) trends for increasing frequency of use to be associated with lower educational attainment both separately in each study and in the
Discussion
Our findings show clear and consistent associations between the frequency of adolescent cannabis or alcohol use and non-attainment of secondary school and tertiary qualifications. The associations had dose–response characteristics across all outcomes, with effects strongest for weekly users. After controlling for a wide range of potential confounding factors, the magnitude of associations for cannabis use reduced substantially but remained significant. In contrast, for frequency of alcohol use,
Role of the funding source
This study was supported by an Australian Government National Health and Medical Research Council Project Grant (numbers 1064893, 1009381). The National Drug and Alcohol Research Centre at UNSW Australia, Sydney, Australia is supported by funding from the Australian Government. G.C.P. is supported by a National Health and Medical Research Council (NHMRC) Senior Principal Research Fellowship (APP1019887). D.M.H. is supported by a Vice-chancellor's Postdoctoral Fellowship from UNSW Australia.
Contributors
This paper was conceptualised and designed by D.M.F., G.C.P., R.P.M. and L.J.H. Data were acquired by L.J.H., D.M.F., G.C.P., J.W.T., C.A.O. and C.C. Data analysis was conducted by E.S., L.J.H., C.A.O., D.M.H. and E. Spry. All authors contributed to data interpretation. E.S., L.J.H. and R.J.T. drafted sections of the paper. All authors revised the paper critically and approved the version published.
Conflict of interest
None declared.
Acknowledgements
We thank all individuals and families involved in the participating cohorts for their time and invaluable contribution to the study; all collaborators who have contributed to the Australian Temperament Project, especially Ann Sanson, Diana Smart, Margot Prior, and Frank Oberklaid; and Christina O’Loughlin, John Carlin, and Helen Romaniuk for their contributions to the Victorian Adolescent Health Cohort.
References (41)
- et al.
The effects of alcohol use on academic achievement in high school
Econ. Educ. Rev.
(2011) - et al.
Assessment of factors affecting the validity of self-reported health-risk behavior among adolescents: evidence from the scientific literature
J. Adolesc. Health
(2003) - et al.
Understanding differences in health behaviours by education
J. Health Econ.
(2010) - et al.
Drinking, smoking, and educational achievement: cross-lagged associations from adolescence to adulthood
Drug Alcohol Depend.
(2014) Covariance adjustment of rates based on the multiple logistic regression model
J. Chronic Dis.
(1981)- et al.
Psychological and social sequelae of cannabis and other illicit drug use by young people: a systematic review of longitudinal, general population studies
Lancet
(2004) - et al.
Predicting young adult degree attainment by late adolesecnt marijuana use
J. Adolesc. Health
(2015) - et al.
Global burden of disease and injury and economic cost attributable to alcohol use and alcohol-use disorders
Lancet
(2009) - et al.
Young adult sequelae of adolescent cannabis use: an integrative analysis across three Australasian Cohorts
Lancet Psychiatry
(2014) - et al.
Is the relationship between early-onset cannabis use and educational attainment causal or due to common liability?
Drug Alcohol Depend.
(2013)
Drug use patterns and continuous enrollment in college: results from a longitudinal study
J. Stud. Alcohol Drugs
The effects of adolescent health-related behavior on academic performance: a systematic review of the longitudinal evidence
Rev. Educ. Res.
Drinking, socioemotional functioning, and academic progress in secondary school
J. Health Soc. Behav.
Integrative data analysis: the simultaneous analysis of multiple data sets
Psychol. Methods
The Big Payoff: Educational Attainment and Synthetic Estimates of Work-Life Earnings
From emerging to established: longitudinal patterns in the timing of transition events among Australian emerging adults
Emerg. Adulthood
The downward spiral of mental disorders and educational attainment: a systematic review on early school leaving
BMC Psychiatry
European Drug Report 2014
The Christchurch Health and Development Study: review of findings on child and adolescent mental health
Aust. N. Z. J. Psychiatry
Cannabis and educational achievement
Addiction
Cited by (0)
- 1
Other members are listed in Appendix A.