The genetic and environmental architecture of substance
use development from early adolescence into young
adulthood: a longitudinal twin study of comorbidity of
alcohol, tobacco and illicit drug use
, Kees-Jan Kan
& Svenn Torgersen
Department of Psychology, University of Oslo, Oslo, Norway,
Department of Biological Psychology, Vrije Universiteit Amsterdam, Amsterdam, the Netherlands
Research Institute of Child Development and Education, Amsterdam, the Netherlands
To investigate how use of alcohol, illicit drugs and tobacco come from substance-speciﬁc pathways and from path-
ways general to all three substances through adolescent development.
Analysis of population-based survey. Ad-
olescent twins reported alcohol use (AU), tobacco use (TU) and illicit drug use (IDU) in three waves (2006, 2008, 2010).
Restructuring data by age allowed for variance decomposition into age- and substance-speciﬁc and common genetic and
environmental variance components.
Seven national twin birth cohorts from 1988 to
1994, totalling 1483 pairs (558 monozygotic; 925 dizygotic, same and opposite sex).
scores of AU, TU and IDU on items from the Monitoring the Future Study.
Substance use was found to be highly
= 0.73 [95% conﬁdence interval (CI) = 0.61–0.94] for AU, a
= 0.49 (95%
CI = 0.29–0.62) for IDU and a
= 0.46 (95% CI = 0.23–0.54); d
= 0.05 (95% CI = 0.00–0.07) for TU during the whole
adolescence period. General substance use (GSU) was also highly heritable at each age and averaged a
CI = 0.48–0.66). There was a high genetic carry-over from earlier age to later age. Genetic effects on GSU at ages 12–
14 years were still detectable 4 years later. New substance (general and speciﬁc)-genetic effects also appeared. IDU dem-
onstrated signiﬁcant non-additive genetic effects (ages 12–14 years). Shared environment had a small impact on AU only.
There was almost no non-shared environmental carry-over from age to age, the effect probably due partly to reliability de-
ﬁciency. Common genetic effects among substance and substance-speciﬁc genetic effects were observed at each age-period.
Among Norwegian adolescents, there appear to be strong genetic effects on both substance-speciﬁcand
comorbid use of alcohol, illicit drugs and tobacco; individual differences in alcohol use can be explained partially by family
Keywords Adolescence, alcohol use, comorbidity, heritability, illicit drug use, longitudinal, substance use, tobacco
use, twin study.
Correspondence to: Trine Waaktaar, Department of Psychology, University of Oslo, PO Box 1094 Blindern, N-0317 Oslo, Norway.
Submitted 9 December 2016; initial review completed 26 January 2017; ﬁnal version accepted 13 October 2017
Adolescence is the peak period for the onset of using sub-
stances. The median age of onset for alcohol and tobacco
use is 16–18 years world-wide, and somewhat later
for illicit drug use such as marijuana (18–19 years)
and cocaine (21–24 years) . Low use of alcohol or
tobacco are the most prevalent patterns . Using
multiple substances and including illicit drugs ) is
associated with different predictors and more problematic
The causal mechanisms behind substance use in
adolescence are still largely unknown, but genetic inﬂu-
ences are indicated. Molecular studies have shown several
genetic variants associated with substance use (individual
effect sizes typically low [5,6]), some also pointing to a
more non-speciﬁc liability for abuse and dependence
among substance types [7,8].
© 2017 Society for the Study of Addiction Addiction, 113,740–748