This is a preprint of an article published in Drug and Alcohol Dependence
© 2008 Elsevier Ireland Ltd. All rights reserved.
Fergusson DM, Boden JM, Horwood LJ. The developmental antecedents of illicit drug use:
Evidence from a 25-year longitudinal study.
Drug and Alcohol Dependence 2008; 96:165-177
The developmental antecedents of illicit drug use: Evidence from a 25-year longitudinal study
David M. Fergusson
Joseph M. Boden
L. John Horwood
Department of Psychological Medicine, University of Otago, Christchurch School of Medicine and
Health Sciences, PO Box 4345, Christchurch Mail Centre, Christchurch 8140, New Zealand
Corresponding author: Prof. David M. Fergusson, Christchurch Health and Development Study,
University of Otago, Christchurch School of Medicine and Health Sciences, PO Box 4345,
Christchurch, New Zealand
Phone: +64 3 372 0406 Fax: +64 3 372 0407 Email: email@example.com
Acknowledgements: This research was funded by grants from the Health Research Council of New
Zealand, the National Child Health Research Foundation, the Canterbury Medical Research
Foundation and the New Zealand Lottery Grants Board.
The developmental antecedents of illicit drug use: Evidence from a 25-year longitudinal study
David M. Fergusson, Joseph M. Boden, and L. John Horwood
Background: The present study examined the developmental antecedents of illicit drug use and
Methods: A 25-year prospective longitudinal study of the health, development, and adjustment of a
birth cohort of 1,265 New Zealand children. Measures included assessments of adolescent and
young adult illicit drug use and abuse/dependence; cannabis use to age 25; measures of parental
adjustment; measures of exposure to childhood sexual abuse, physical abuse, and interparental
violence; novelty-seeking; childhood and early adolescent adjustment and substance use; and
affiliation with substance-using peers.
Results: Illicit drug use and abuse/dependence from ages 16-25 were significantly associated (all p
values < .05) with a range of parental adjustment measures; exposure to abuse in childhood;
individual factors; and measures of childhood and early adolescent adjustment. Analyses using
repeated measures logistic regression models suggested that parental illicit drug use, gender,
novelty seeking, and childhood conduct disorder predicted later illicit drug use and
abuse/dependence. Further analyses revealed that these pathways to illicit drug use and
abuse/dependence were mediated via cannabis use, affiliation with substance using peers, and
alcohol use during ages 16-25.
Conclusions: The current study suggested that the illicit drug use and abuse/dependence were
associated with a range of early life circumstances and processes that put individuals at greater risk
of illicit drug use and abuse/dependence. However, the use of cannabis in late adolescence and
early adulthood emerged as the strongest risk factor for later involvement in other illicit drugs.
Keywords: illicit drug use, cannabis use, peer substance use, family background, longitudinal study
In recent decades, many developed societies have grappled with issues concerning the use and
abuse of a range of illicit drugs, including hallucinogens (LSD, Ecstasy, PCP), opiates (heroin,
morphine), stimulants (methamphetamine), cocaine, barbiturates, solvents, prescription
medications, and plant extracts such as mushrooms (Substance Use and Mental Health Services
Administration, 2007; World Health Organization, 2000). The use of such drugs has raised issues
about the risk factors and life processes that lead young people to experiment with, use, and abuse
these drugs. This topic has been the subject of a growing literature (for reviews see: Anderson,
2006; Bloor, 2006; Compton et al., 2005; Galea et al., 2004; Hawkins et al., 1992; Mayes and
Suchman, 2006; Rehm et al., 2006; Ripple and Luther, 1996) that has identified a range of factors
associated with the increased usage of illicit drugs, including: family background factors, family
support, and parental supervision; exposure to abuse during childhood and adolescence; individual
factors such as temperament and gender; adjustment problems in childhood and early adolescence;
and affiliation with substance using peers. The purpose of the present investigation is to examine
the extent to which both: (a) childhood factors; and (b) adolescent adjustment and substance use;
may be related to later illicit drug use. A review of representative findings in this area is given
1.1 Parental adjustment factors
Research suggests that parental adjustment problems, including substance use and criminality, may
be related to adjustment problems in children, via genetic influences, social learning, or exposure to
an unstable home environment (e.g. Bahr et al., 2005; Keyes et al., 2007; Pears et al., 2007;
Stallings et al., 1997; e.g. van der Zwaluw et al., 2008). It could therefore be argued that parental
adjustment factors may play a key role in the development of substance use in offspring.
A wide range of studies have examined family background factors and their relationship to
risk of illicit drug use and abuse/dependence in adolescence and early adulthood. A number of
studies have found that parental substance use may be related to substance use and abuse in
adolescents (Johnson and Leff, 1999). For example, Kilpatrick and colleagues (Kilpatrick et al.,
2000), using data from a national household survey, found that exposure to illicit drug abuse by
parents increased the risk of substance abuse and dependence amongst adolescents. Similarly,
Merikangas et al (1998) found that individuals who had first-degree relatives who were diagnosed
with substance use disorders were at an eight-fold risk of also being diagnosed with a substance use
disorder. Pears and colleagues (2007), using prospective data, found evidence to suggest that
patterns of substance use and abuse were transmitted across three generations.
Other factors such as lower levels of parental interest and monitoring, and higher levels of
parental criminality may be related to higher levels of substance use and abuse/dependence. For
example, Chilcoat and Anthony (1996) found that rates of illicit drug use were higher amongst
adolescents who perceived lower levels of parental concern and involvement. Similarly, Duncan et
al (1998) reported that inept parental monitoring was related to increased levels of substance use in
adolescents. Hayatbakhsh and colleagues (2007), using data from a prospective birth cohort, found
that maternal partner criminality predicted cannabis use by age 21. Similarly, Felitti and colleagues
(1998) found that parental criminality was associated with an increased risk of drug abuse amongst
a large (n > 10,000) cross-sectional sample. In general, the evidence suggests that family
dysfunction and parental adjustment problems may be associated with increased risk of illicit drug
1.2 Exposure to abuse in childhood
A second set of factors that has been linked to the risk of involvement with illicit drugs is exposure
to abuse during childhood. It could be argued that exposure to abuse in childhood may lead to
increased risk of substance use in adulthood, at least in part due to concomitant increases in risk of
mental illness (e.g. Mulder, 2002). However, it should be noted that longitudinal data are critical to
demonstrate links between childhood abuse and later outcomes (Putnam and Trickett, 1993).
A number of longitudinal studies have found that exposure to sexual abuse in childhood
(CSA), exposure to physical abuse, and witnessing interparental violence may be linked to
increased risk of substance use and abuse/dependence in adolescence and adulthood (Miller et al.,
1997; Simpson and Miller, 2002). For example, Widom and colleagues (2006), using data from a
prospective cohort study, found that those exposed to sexual and/or physical abuse in childhood
were 1.5 times more likely to report using illicit drugs at age 40. Similarly, Kilpatrick et al (2000)
found that adolescents who reported having been sexually or physically abused, or who witnessed
interparental violence, were at increased risk for substance abuse/dependence. Also, Wilsnack and
colleagues (1997) found that women who reported exposure to CSA were more likely to report
using a range of illicit drugs than women who reported no CSA exposure. Felitti et al. (1998) found
that one of the key risk factors for later substance abuse/dependence was exposure to interparental
violence. Christofferson and Soothill (2003), using data from over 80,000 children born in
Denmark in 1966, found that violence between parents, stemming from alcohol abuse, increased
risks of drug addiction in children aged 15-27. In addition, Dube and colleagues (2002), using a
case-control design, found that exposure to interparental violence during childhood was associated
with increased likelihood of illicit drug use, including intravenous drug use. The findings of these
studies suggest that exposure to abuse during childhood increases the risk of illicit drug use and
abuse/dependence in adulthood.
1.3 Individual factors
Research suggest that there are a range of individual factors that may predict illicit drug use and
dependence. However, two of the strongest predictors of involvement in illicit drugs are gender and
novelty-seeking. For example, a range of studies has shown that males are more likely than females
to report the illicit drug use and abuse/dependence (Bloor, 2006), and that those reporting higher
levels of novelty-seeking (Acton, 2003; Staiger et al., 2007) demonstrate higher levels of illicit drug
use and abuse/dependence. For example, Duncan et al. (1998), using growth curve modelling
techniques, found that gender predicted adolescent substance use, with males more likely than
females to report substance use. Also, Rodham and colleagues (2005), in a sample of English 15
and 16 year old adolescents, found that males were more likely to report illicit drug use. Similar
findings were reported for an Australian secondary school sample (n > 20,000) by Lynskey et al.
(1999). In terms of novelty-seeking, Evren et al. (2007) found that novelty-seeking predicted illicit
drug dependence in a sample of Turkish adolescents. Also, Adams et al. (2003), using a case-
control sample of adolescent substance users, found that those who were high in novelty-seeking
reported using a wider range of substances than those low in novelty-seeking. Khan and colleagues
(2005) reported that novelty-seeking had a modest positive association with externalizing disorders,
including substance dependence. These findings suggest that males and those higher in novelty-
seeking are at a greater risk of illicit drug use and abuse/dependence.
1.4 Childhood and early adolescent conduct and attention problems
An additional set of factors that may be linked to risks of illicit drug involvement are conduct and
attention problems in childhood. A number of studies have suggested that individuals who display
conduct and attention problems are at greater risk of illicit drug use and abuse/dependence (Deas
and Brown, 2006). For example, Khan and colleagues (2005), using twin-registry data, found
associations between conduct problems and substance abuse problems. Also, Molina and Pelham
(2003) found that conduct and attention disorder in early adolescence predicted later use and abuse
of a range of substances in later adolescence in a prospective study. Flory et al (2003) found an
interactive relationship between conduct and attention problems and later substance use and
dependence, such that individuals with high levels of both conduct and attention problems were at
greatest risk of later illicit drug use and dependence. In addition, Fergusson and colleagues (2007)
found that the associations between attention problems in childhood and later illicit drug use were
mediated via conduct problems. The findings of these studies suggest that conduct and attention
problems in childhood may be related to increased risks of later illicit drug involvement.
1.5 Adolescent substance use and social processes
A further set of factors that may be linked to increasing risks of illicit drug involvement are
substance use and social processes in adolescence. A large number of studies have found evidence
for “gateway” effects for cannabis use in illicit drug use and abuse/dependence, in which the use of
cannabis may increase the risk of use of other illicit drugs (Kandel, 2003; MacCoun, 1998), while
others have suggested that other drugs such as alcohol and tobacco may also be involved in the
gateway to illicit drug involvement (Botvin et al., 2002; Center On Addiction And Substance
Abuse, 1994). The nature of these “gateway” effects is a matter of some debate (Fergusson et al.,
2006; Kandel et al., 2006; MacCoun, 1998, , 2006; Morral et al., 2002), and there may be evidence
of “reverse gateway” effects (Viveros et al., 2006) in which the use of cannabis may increase the
risk of tobacco use. However, a range of studies have clearly shown that the use of cannabis is
associated with increasing risks of other illicit drug use and abuse/dependence (Kandel, 2003;
MacCoun, 1998). Furthermore, peer influence, and in particular affiliation with substance-using
peers, may be associated with increased risks of illicit drug use and abuse/dependence (Bloor,
The gateway effects of cannabis and other substances have been examined in a number of
studies. For example, Lynskey and colleagues (2003) using a discordant twin design, found that
those who had used cannabis were 2.1 to 5.2 times more likely to use other illicit drugs. Also,
Kandel and colleagues (1992) found that 80% to 90% of individuals who reported using both
cannabis and other illicit drugs used cannabis prior to using other illicit drugs. Further, Fergusson
et al. (2006), using a prospective cohort design, found that frequency of cannabis use predicted risks
of other illicit drug use, such that those who used cannabis more frequently were more likely to use
other illicit drugs. Kandel and Yamaguchi (2002) found that both tobacco use and alcohol use
tended to precede the use of cannabis and other illicit drugs in a large national data set.
Peer influence has also been examined in a number of studies. For example, Duncan and
colleagues (1998) found that peer deviance predicted substance use trajectories during adolescence.
Jenkins (1996), using a large cross-sectional data set, found that peer drug use was the strongest
predictor of illicit drug use during adolescence. Also, Wills and colleagues (1998) found that peer
influence mediated the associations between early risk factors and later substance use.
Collectively, the findings of these studies suggest that involvement with substance use in
early adolescence, and affiliation with substance-using peers, increase the risk of illicit drug
involvement in adolescence and early adulthood.
1.6 Background to the present study
While there has been growing evidence on the risk and protective factors for illicit drug use, there
have been relatively few studies that have reported longitudinal data on the role of a wide range of
risk factors assessed over the period from childhood to adulthood. In this paper we report on the
results of a longitudinal study of use and abuse of illicit drugs in a birth cohort of New Zealand
young people studied from birth to the age of 25 years. The analyses reported center on examining
a range of issues in the development of illicit drug use and abuse. These issues include:
1. The role of childhood and parental factors in predisposing young people to use illicit drugs;
2. The role of peer affiliation and substance (tobacco, alcohol, and cannabis) use in the
development of illicit drug use and abuse.
The hypothesized relationships between childhood and parental factors, adolescent peer affiliation
and substance use factors, and illicit drug use and abuse/dependence are presented in Figure 1.
INSERT FIGURE 1 HERE
More generally, the aims of the paper are to develop a multivariate account of the ways in
which a wide range of social, family, and individual factors combine over the life course to
influence the use and abuse of illicit drugs.
The data were gathered as part of the Christchurch Health and Development Study (CHDS). The
CHDS is a longitudinal study of a birth cohort of 1,265 children (635 males, 630 females) born in
the Christchurch (New Zealand) urban region in mid-1977. The cohort has been studied at birth, 4
months, 1 year and at annual intervals to age 16 years, and again at ages 18, 21, and 25. The study
has collected information from a variety of sources including: parental interviews, teacher reports,
self-reports, psychometric assessments, medical, and other record data. All aspects of the study’s
design have been approved by the Canterbury Ethics Committee. An overview of the study design,
methodology, and major findings can be found in Fergusson, Horwood, Shannon, and Lawton
(1989) and Fergusson and Horwood (2001).
All analyses were based on all cohort members assessed at each point of observation.
Sample sizes were as follows: 18 years (n = 1025); 21 years (n = 1011); and 25 years (n = 1003).
These samples represented between 79% and 81% of the original cohort of 1265 participants. In
addition, as a result of missing data on some covariates the sample number included in the covariate
adjustment analyses was reduced to approximately 900. The present analysis used the following
2.1 Illicit (non-cannabis) drug use and illicit drug abuse/dependence. As part of the interviews at
ages 18, 21, and 25, cohort members were questioned about their use of illicit drugs other than
cannabis since the previous assessment. Each assessment included questions about other illicit drug
use for each year of the assessment period. The cohort members were questioned about their use of
a range of illicit drugs, including solvents (glue, petrol, paint); stimulants (including
methamphetamine); barbiturates; other prescription medications that were illicitly obtained; opiates,
including both heroin and morphine; cocaine (in any form); hallucinogens including ecstasy, LSD,
and PCP; and any other substances (primarily plant extracts) including mushrooms and datura. In
this way the data collection provided an account of the individual’s reported frequency of use of a
range of other illicit drugs for each year from age 16-17 to age 24-25. For the purposes of this
analysis, participants were classified as having used other illicit drugs in a given year if they
reported using any of the above classes of substances on at least one occasion, creating a
dichotomous outcome measure. In addition, at each assessment, participants were questioned about
problems associated with their use of drugs since the previous assessment with items from the
Composite International Diagnostic Interview (CIDI: World Health Organization, 2000). This
information was compared with DSM-IV criteria (American Psychiatric Association, 1994) to
construct a diagnosis of illicit drug abuse/ dependence during each year from ages 16-17 to age 24-
25, which was also a dichotomous outcome measure. The measure of illicit drug use and
abuse/dependence has been reported to have adequate reliability and validity (Andrews and Peters,
1998). For the purposes of the present analyses, all participants classified as having a diagnosis of
other illicit drug abuse/dependence were also classified as having used other illicit drugs.
2.2 Predictors of illicit drug use and illicit drug abuse/dependence
To examine the antecedents of illicit drug use and illicit drug abuse/dependence, a series of
measures was chosen from the data base of the study for inclusion in the analysis. These measures
were selected on the basis of: (a) a review of the literature identifying factors which previously have
been found to be associated with increased risks of illicit drug involvement (see above); and (b)
previous analyses based on the Christchurch Health and Development Study cohort which have
identified factors associated with illicit drug involvement (Boden et al., 2006; Fergusson et al.,
2006). The factors chosen for inclusion in the analyses were as follows:
2.2.1 Measures of parental adjustment and substance use
Parental illicit drug use. When the young person was aged 11, resident parents were
questioned as to their history of illicit drug use. The young person was classified as having a parent
history of illicit drug use if one of his/her parents was reported to have a history of illicit drug use.
Parental criminality. When the young person was aged 15, resident parents were
questioned as to their history of criminal offending, including property and violent offending, and
any official convictions. The young person was classified as having a parent history of criminality if
one of his/her parents was reported to have a history of offending.
Parental alcohol problems. When the young person was aged 11, resident parents were
questioned as to whether they had ever experienced either problems with alcohol, or alcoholism.
These reports were used to form a dichotomous measure of whether or not the young person's
parents reported experiencing alcoholism or problems with alcohol.
2.2.2 Measures of exposure to abuse in childhood
Childhood sexual abuse. At ages 18 and 21 years sample members were questioned
retrospectively about their experience of sexual abuse during childhood (<16 years) (Fergusson et
al., 1996). Questioning spanned an array of abusive experiences from episodes involving non-
contact abuse (e.g. indecent exposure) to episodes involving attempted or completed intercourse.
Sample members who reported an abusive episode were then questioned further about the nature
and context of the abuse. Using this information a 4-level scale was devised reflecting the most
extreme form of sexual abuse reported by the young person at either age. This classification was:
no sexual abuse; non-contact abuse only; contact sexual abuse not involving attempted or
completed intercourse; attempted/completed oral, anal, or vaginal intercourse.
Parental use of physical punishment (childhood physical abuse). At ages 18 and 21 sample
members were asked (retrospectively) to describe the extent to which their parents used physical
punishment during childhood (Fergusson and Lynskey, 1997). Separate questioning was conducted
for mothers and fathers. This information was used to create a 4-level scale reflecting the most
severe form of physical punishment reported for either parent: parents never used physical
punishment; parents rarely used physical punishment; at least one parent used physical punishment
on a regular basis; at least one parent used physical punishment too often or too severely, or treated
the respondent in a harsh or abusive manner. The interpretation of scale items was left to the
discretion of respondents.
Interparental violence (0-16 years). At the age of 18, sample members were questioned
concerning their (retrospective) experience of interparental violence during their childhood (prior to
age 16 years). The questioning was based on a series of eight items derived from the Conflict
Tactics Scale (CTS: Straus, 1979). Separate questioning was conducted for both father-initiated
and mother-initiated interparental violence. An overall measure was created by summing the
responses for both father- and mother-initiated violence.
2.2.3 Individual factors
Gender. Recorded at birth.
Novelty-seeking. Novelty-seeking was assessed via self-report at age 16 using the novelty-
seeking items from the Tridimensional Personality Questionnaire (Cloninger, 1987), α = .76. For
the purposes of the present analyses, novelty-seeking scores were split into quartiles to create a
four-level measure of novelty-seeking at age 16.
2.2.4 Childhood and adolescent adjustment
Child conduct and attention problems (ages 7-13). At each assessment from age 7 - 13
years, information on child behavior problems related to conduct and attention issues was obtained
from parental and teacher report using a behavior questionnaire that combined items from the
Rutter, Tizard, and Whitmore (1970), Conners (1970), and Conners (1969) parental and teacher
questionnaires. For the purposes of the present analysis, the parent and teacher reports were
summed and the resulting scores averaged over the seven year period to produce scale score
measures reflecting the extent of the child’s tendencies to conduct and attention problems at ages 7-
13 (Fergusson and Horwood, 1993; Fergusson et al., 1991). The alpha reliabilities of these scales
were .97 and .93, respectively.
2.2.5 Time-dynamic measures of substance use and peer influence (ages 16-25).
In all cases, the time-dynamic measures used in the analyses were re-scaled to categorical measures
with four levels. The use of the four-level measures allowed direct effect-size comparisons with the
other time-dynamic measures (below).
Annual frequency of cannabis use (ages 16-25). At ages 18, 21, and 25, cohort members
were questioned about their use of cannabis since the previous assessment. Each assessment
included questions about cannabis use for each year of the assessment period; for example, the 18-
year assessment included questions on cannabis use over the periods 16-17 and 17-18 years. In this
way the data collection provided an account of the individual’s reported frequency of cannabis use
for each year from age 16-17 to age 24-25. For the purposes of this analysis, the annual frequency
data were classified into a series of class intervals as follows: did not use cannabis; used less than
monthly on average (1-11 times); used at least monthly on average (12 – 50 times); used at least
weekly (> 50 times); resulting in a four-level measure representing frequency of cannabis use
during each year from ages 16 to 25.
Affiliation with substance-using peers (ages 16-25). This was assessed at ages 16, 18, 21,
and 25 on the basis of participant reports of the extent to which their friends used tobacco, alcohol,
or illicit drugs or had problems resulting from alcohol or illicit drugs during the previous 12
months, α = .69-.77. For the purposes of the present analysis, these scales were split into quartiles
in order to create a four-level measure representing the extent of the individual’s involvement with
substance-using peers at each assessment (Fergusson and Horwood, 1999; Fergusson et al., 1999).
Frequency of cigarette smoking (ages16-25). At ages 16, 18, 21, and 25, participants were
questioned about their current frequency of cigarette smoking. For the purposes of the present
analyses, these data were classified into a series of four class intervals ranging from non-smoker to
20+ cigarettes per day.
Frequency of alcohol use (ages 16-25). This was assessed at ages 16, 18, 21, and 25 for the
previous 12 months. For the purposes of the present analyses, these data were classified into a
series of four class intervals representing the self-reported frequency of drinking alcohol ranging
from “never” to “almost every day”.
2.3 Missing Data
As noted previously, the analysis was based on approximately 900 sample members, which
represented between 79% and 81% of the original cohort of 1265 participants. To assess the
possible effects of sample selection bias, tests were conducted to examine the extent to which the
obtained sample was representative of the original cohort of 1,265 participants enrolled in the study.
This analysis showed that there were slight but statistically significant (p < .05) tendencies for the
obtained sample to under-represent individuals from more socially disadvantaged backgrounds (low
parental education, low socioeconomic status, single-parent family). To take these biases into
account, the sample was poststratified into a series of groups on the basis of these characteristics,
and the probability of study participation estimated for each group using the methods described by
Carlin, Wolfe, Coffey, and Patton (1999). All analyses were then repeated with the data for the
analysis sample weighted by the inverse of the probability of study participation. In addition, there
were small amounts of missing data for some covariate factors. To examine the implications of
missing values, regression imputation of missing data was conducted and the analyses repeated with
the missing values on each covariate replaced by the imputed values. In all cases, these reanalyses
produced essentially the same pattern of results to those reported here, suggesting that the
conclusions of this study were unlikely to have been influenced by missing data and selection bias.
3.1 Rates of illicit drug use and illicit drug abuse/dependence
Over the period from ages 16-25 years, 42.9% (n = 458) of the sample reported using illicit drugs
other than cannabis on at least one occasion, and 10.8% (n = 115) met DSM-IV criteria for
abuse/dependence (see Methods). Amongst the cohort, 35% reported having used hallucinogens
(ecstasy, LSD); 26% reported having used stimulants (including methamphetamine), barbiturates,
or other (illicitly obtained) prescription medicines; 17% reported having used substances such as
mushrooms and datura; 5% reported having used solvents; 9% reported having used cocaine; and
4% reported having used opiates including heroin and morphine.
3.2 Factors associated with illicit drug use and illicit drug abuse/dependence, ages 16-25
Table 1 shows the results of regression analyses linking a series of factors to annual rates of illicit
drug use and illicit drug abuse/dependence over the period from 16-25 years. Assessments of illicit
drug use and abuse/dependence (other than cannabis) were made at annual intervals from age 16 to
age 25. To analyze these repeated measures data, logistic generalized estimating equation (GEE)
models were fitted to each predictor variable (Gibbons et al., 1988) in which the log odds of either:
(a) illicit drug use; or (b) illicit drug abuse/dependence in a given year were modelled as a linear
function of the predictor and age. Random effects models provide a single estimate of the
regression coefficients pooled over the repeated measures of the outcome variable. The models
fitted were of the form:
Logit (Yit) = B0i + B1Xi + B2At + Uit (EQ1)
where Yit was the measure of use or abuse/dependence for participant I at time t, B0i was an
individual-specific random-effects intercept term, Xi was the predictor of interest (or Xit, if the
predictor was time-dynamic), At was the measure of the respondent’s age (centered at age 21), and
Uit was the disturbance term for the models. The results of these analyses are presented in Table 1,
which shows parameter estimates, standard errors, and significance levels for the statistically
significant risk factors for illicit drug use and abuse/dependence (see Methods). The Table shows:
1. For both use and abuse/dependence there were significant associations with parental illicit drug
use (p < .0001), and parental criminal offending (p < .0001). In addition, illicit drug use (but
not abuse/dependence) was associated with parental alcoholism (p <.001).
2. Illicit drug use and abuse/dependence were significantly associated with: exposure to
interparental violence (p < .001); exposure to sexual abuse in childhood (p < .001); and
exposure to childhood physical abuse (p < .001).
3. Both illicit drug use and abuse/dependence were also significantly associated with individual
factors, including male gender (p < .01) and novelty-seeking (p < .0001).
4. Use and abuse/dependence were also significantly associated with conduct problems during
ages 7-13 (p < .0001). Abuse/dependence, but not illicit drug use, was associated with attention
problems during ages 7-13 (p < .0001).
5. Illicit drug use and abuse/dependence were also significantly associated with a range of time-
dynamic measures of substance use and peer influence during ages 16-25, including annual
frequency of cannabis use (p < .0001), affiliation with substance-using peers (p < .0001),
frequency of cigarette smoking (p < .0001), and frequency of alcohol use (p < .0001).
INSERT TABLE 1 HERE
3.3 Multivariate analyses of risk factors for illicit drug use and illicit drug abuse/dependence, ages
The results presented in Table 1 were extended to fit two multivariate logistic GEE models of the
linkages between the factors listed in Table 1 and the risks of illicit drug use and illicit drug
abuse/dependence over the period 16-25 years. These models were fitted in three steps, as follows:
i. In the first model, the fixed predictors presented in Table 1 (parental adjustment factors; abuse
exposure factors; individual factors; childhood and adolescent adjustment factors) were entered
in blocks, with forward and backward elimination of variables to identify a stable and
parsimonious set of predictors of illicit drug use and abuse/dependence. These models were of
Logit (Yit) = B0i + ΣBjXij+ B2At + Uit (EQ2)
where Yit was the measure of use or abuse/dependence at time t, B0i was a random-effects
intercept term, Xij were the set of fixed predictor variables in the fitted model, At was the
measure of the respondent’s age (centered at age 21), and Uit was the model disturbance term.
ii. The second model extended the first model by including the time-dynamic factors listed in
Table 1 (annual frequency of cannabis use ages 16-25; affiliation with substance-using peers
ages 16-25; frequency of alcohol use and cigarette smoking, ages 16-25), with the factors being
entered into the models simultaneously. The factors were represented as four-level categorical
variables in order to permit direct comparisons of effect sizes (see Methods). In addition, to
take into account age related changes in the strength of association between the predictors and
outcome measures, tests of age x predictor interaction terms were entered into the models in
forward and backward stepwise fashion, with only statistically significant interactions retained
in the final models. These models were of the form:
Logit (Yit) = B0i + ΣBjXij + ΣBkXikt + B2At + ΣBpAXipt + Uit (EQ3)
where Xij was the set of fixed predictors, Xikt was the set of time-dynamic predictors, and
AXipt was the set of potential interaction terms between age and the predictors X.
iii. The third model extended the second model by including a lagged cumulative dichotomous
measure of previous illicit drug use, in order to account for issues of reverse causality, in which
contemporaneous measures of cannabis use and other illicit drug use and abuse/dependence may
cause somewhat inflated estimates of association.
Tables 2 and 3 show the regression coefficients, standard errors, and tests of significance for each
model at each step for: (a) illicit drug use; and (b) illicit drug abuse/dependence. The Tables show:
1. In the first step of the analyses, several factors emerged as statistically significant predictors of
illicit drug use and abuse/dependence. These factors included exposure to childhood sexual
abuse (p < .0001), gender (p < .01), novelty-seeking (p < .0001), and conduct problems at ages
7-13 (p < .05). In addition, parental illicit drug use was a statistically significant predictor for
illicit drug use (parental illicit drug use, p < .05).
2. In the second step of the analyses, the results showed that:
a. Four of the time-dynamic factors were significant predictors of illicit drug use, while
two of the factors were significant predictors for illicit drug abuse/dependence. For
illicit drug use, these factors included annual frequency of cannabis use (p < .0001),
extent of affiliation with substance-using peers (p < .0001), frequency of alcohol use (p
< .0001), and frequency of cigarette smoking (p < .01). For illicit drug
abuse/dependence, significant predictors were annual frequency of cannabis use (p <
.0001), and affiliation with substance-using peers (p < .0001). In addition, for both illicit
drug use and abuse/dependence there were significant age x cannabis use interactions (p
< .0001), suggesting that the associations between cannabis use and illicit drug use and
abuse/dependence decreased in magnitude over time.
b. The results also showed that when the time-dynamic measures of cannabis use, peer
substance use, cigarette smoking, alcohol use, and interaction terms were entered into
the models along with the fixed predictors, the associations between the fixed predictors
and both illicit drug use and illicit drug abuse/dependence were reduced in magnitude,
with most associations being reduced to statistical non-significance. The exceptions to
this pattern were the associations between novelty-seeking and illicit drug use, which
remained statistically significant (p < .0001).
3. In the third step of the analyses, the addition of the lagged measure of illicit drug use reduced
the strength of the associations between the predictors and the measures of illicit drug use and
illicit drug abuse/dependence. The lagged measure of illicit drug use was a statistically
significant predictor for both illicit drug use and abuse/dependence (p < .0001).
The results of these analyses suggest that the time-dynamic substance use and peer factors mediated
the linkages between the predictors and both illicit drug use and illicit drug abuse/dependence,
indicating that the pathway to involvement with illicit drugs leads through the use of cannabis and
other substances, and through social processes including affiliation with substance-using peers. The
exception to this pattern was a persistent significant association between novelty-seeking and illicit
drug use. In addition, the finding of significant age by cannabis use interactions suggested that the
strength of these associations changed over time (see below).
INSERT TABLES 2 AND 3 HERE
3.4 Effect size estimates and evaluation of interaction effects
In order to illustrate the relative contributions of the factors to the final models presented in Tables
2 and 3, estimates of effect size were generated for the statistically significant time-dynamic
predictors and novelty-seeking. For the purposes of the present analyses, estimates of the odds ratio
(OR) and 95% confidence intervals were computed for each of the four levels of each time-dynamic
factor (cannabis use; peer substance use; alcohol use) and the quartile measure of novelty-seeking.
Tables 4 and 5 show estimates of the ORs and 95% confidence intervals for the statistically
significant predictors, for both: (a) illicit drug use; and (b) illicit drug abuse/dependence.
In addition, in order to illustrate the statistically significant age x cannabis use interaction
for both illicit drug use and abuse/dependence, ORs and 95% confidence intervals were calculated
for the data at ages 16-17, 20-21, and 24-25.
Examination of the Tables reveals that:
1. Annual frequency of cannabis use made the strongest contribution to both illicit drug use and
illicit drug abuse/dependence. Those using cannabis at least weekly at some point during the
period 16-25 years had odds of illicit drug use that ranged from 92.20 (95% CI: 46.53-182.72;
age 16-17) to 7.53 (95% CI: 4.48-12.43; age 24-25) times greater than those who did not use
cannabis, and had odds of illicit drug abuse/dependence that ranged from 117.92 (95% CI:
26.31-523.74; age 16-17) to 6.49 (95% CI: 2.19-19.20; age 24-25) times greater than those who
did not use cannabis. By contrast, the odds ratios for those individuals in the highest quartile for
affiliation with substance using peers, the highest quartile for novelty-seeking, and who used
alcohol almost every day had odds of illicit drug use and abuse/dependence that ranged from
1.62 to 5.87. The results of these analyses suggest that, amongst the predictors of illicit drug
use and abuse/dependence identified in the analyses presented in Table 2, the annual frequency
of cannabis use was the strongest predictor of both illicit drug use and illicit drug
2. The age x cannabis use interaction for both illicit drug use and illicit drug abuse/dependence
indicated that the associations between cannabis use and other illicit drug use and
abuse/dependence declined with increasing age. For example, the odds of illicit drug use for
those in the highest level of cannabis use declined from 92.20 at age 16-17 to 7.53 at age 24-25,
suggesting that while the association between cannabis use and other illicit drug use was very
strong in adolescence, this association tended to diminish greatly by early adulthood.
INSERT TABLES 4 AND 5 HERE
One of the important findings to emerge from these analyses was the relatively high rate of non-
cannabis illicit drug use, with over 40% of the cohort using these drugs on at least one occasion, and
over 10% meeting DSM-IV criteria for abuse or dependence. Most of the illicit drug use involved
so-called “party drugs” including ecstasy, amphetamines, and LSD. The use of harder drugs,
including cocaine and heroin, was uncommon, reported by less than 9% of the cohort. These
findings are generally consistent with findings reported in Australiasia (Coffey et al., 2000;
Fergusson and Horwood, 2000; Lynskey et al., 1999; Poulton et al., 2001; Swift et al., 2001), but
are slightly higher than comparable figures for studies conducted in North America and Europe
(Chilcoat and Anthony, 1996; Kandel et al., 1997; von Sydow et al., 2001).
To examine the contributions of a range of risk factors to the development of illicit drug use
and abuse/dependence, a three stage regression approach was used. In the first stage of the analysis
the role of various fixed childhood factors including parental adjustment, exposure to abuse in
childhood, individual factors including gender and novelty-seeking, and conduct and attention
problems in childhood and early adolescence was analysed. This analysis showed that risks of later
illicit drug were increased amongst those with parents who had used illicit drugs at least once, were
exposed to sexual abuse in childhood, were male, who had higher scores on novelty-seeking, and
who displayed higher levels of conduct problems in childhood. These findings are consistent with a
large literature that has identified a range of similar factors as predisposing factors in the
development of illicit drug use (for reviews see: Anderson, 2006; Bloor, 2006; Compton et al.,
2005; Galea et al., 2004; Hawkins et al., 1992; Mayes and Suchman, 2006; Rehm et al., 2006;
Ripple and Luther, 1996).
In the second and third stages of the analysis the statistical model was extended and refined
by the inclusion of a series of time-dynamic covariates and controls for reverse causality. The final
fitted model suggested that risks of later illicit drug use were determined by a series of factors that
include cannabis use, affiliation with substance-using peers, alcohol use, cigarette smoking, and
novelty-seeking. The risks of illicit drug abuse/dependence were determined by both cannabis use
and affiliation with substance-using peers. In general, the analyses suggested that, with the
exception of novelty-seeking, accounting for time-dynamic substance use and peer factors reduced
the associations between the childhood fixed factors and illicit drug use and abuse/dependence to
Of the time-dynamic factors included in the final models, cannabis use had the largest and
most complex associations. In particular, the study findings suggested an interactive relationship
between age and the use of cannabis in the development of other forms of illicit drug involvement.
In this relationship the effects of cannabis use were strongest at younger ages, and declined
progressively with age. Furthermore, the size of association depended on the extent of use of
cannabis. The net results of these findings is that risks of illicit drug use that were over 90 times
higher amongst 16-17 year olds who used cannabis at least weekly when compared to non-users of
cannabis. By the age of 25, these risks had reduced to nearly eight times higher. In addition, these
associations were controlled for reverse causality by including a lagged measure of other illicit drug
use in the model. These findings are consistent with the view that exposure to cannabis use
increases risks of other forms of illicit drug use and abuse/dependence, even when due allowance is
taken of childhood factors and possible reverse causal associations. This conclusion is consistent
with the previous analysis of these data using a fixed effects regression model (Fergusson et al.,
These findings on the role of cannabis in the development of other forms of illicit drug use
highlight three important points. The first is that the association was dose dependent, such that the
effects of cannabis on other illicit drug involvement depended on the amount used. The second is
that the association was age dependent such that young users were more susceptible to the effects of
cannabis than older users. The third important feature of the associations was that much of the
association between childhood factors and other forms of illicit drug use and abuse/dependence was
mediated via cannabis use. This finding is important in the light of claims that the association
between cannabis use and other forms of illicit drug use can be explained by common childhood
factors (Donovan and Jessor, 1985; Hays et al., 1987; Huba et al., 1981; Morral et al., 2002). The
present study suggests quite the opposite conclusion in which cannabis use mediated the effects of
childhood factors on later illicit drug abuse.
While the use of cannabis appeared to play a central role in the transition to various forms of
other illicit drug use and abuse/dependence, these risks were also influenced by other factors.
Specifically, there was also evidence to suggest that alcohol use may play a small gateway effect in
encouraging illicit drug use, with increasing alcohol use being associated with increasing illicit drug
use even after control for both the use of cannabis and reverse causal effects. This result is
consistent with findings that suggest that alcohol use may play a role in leading to involvement with
illicit drugs other than cannabis (Center On Addiction And Substance Abuse, 1994; Kandel and
An important factor contributing to the likelihood of other illicit drug use was the extent of
the young person’s affiliation with substance-using peers, with those having higher exposure to
such peers being more than twice as likely to use other illicit drugs, and more than five times more
likely to report other illicit drug abuse/dependence. It is likely that peer affiliation contributes to
illicit drug use in several ways including: (a) providing a source of information about illicit drugs;
(b) providing a source of supply of drugs; and (c) providing a social support system that encourages
the use of illicit drugs. These findings are consistent with previous findings that have suggested
that affiliation with substance-using peers may play an important role in the development of
substance use and abuse/dependence (Bloor, 2006; Duncan et al., 1998; Jenkins, 1996; Wills et al.,
There was also evidence to suggest that personality factors also played a role in the transition
to the use of illicit drugs, with those scoring high on novelty-seeking being more prone to use illicit
drugs than those scoring low on this dimension. These findings are consistent with previous
findings suggesting that variations in propensities to risk-taking may play a role in the development
of illicit drug use and abuse (Acton, 2003; Adams et al., 2003; Evren et al., 2007; Khan et al., 2005;
Staiger et al., 2007).
In summary, the findings of this 25-year longitudinal study suggest that the development of
illicit drug use and abuse/dependence in adolescence and young adulthood involves an
accumulative process that includes exposure to adversity in childhood, childhood adjustment,
personality and individual factors, the use of cannabis, affiliation with substance-using peers, and
alcohol use. Of these factors, the use of cannabis appears to play the strongest role, with this being
particularly evident for young users and heavy users of cannabis. These findings highlight the
importance of developing a better understanding of the processes that link the use of cannabis to the
increased use of other illicit drugs. While there have been frequent criticisms of the view that
cannabis acts as a gateway drug that increases risks of other forms of illicit drug use (Donovan and
Jessor, 1985; Hays et al., 1987; Huba et al., 1981; MacCoun, 1998; Morral et al., 2002), the
evidence gathered in this longitudinal study strongly points in that direction. As shown in the
present paper, cannabis use was the factor that is most strongly related to other forms of illicit drug
use. A previous study of the same birth cohort showed that the associations between cannabis use
and other illicit drug use in this cohort could not be explained away as being due to non-observed
sources of confounding (Fergusson et al., 2006) and persisted after control for reverse causality.
These findings clearly suggest that the use of cannabis plays a central but as yet poorly understood
role in facilitating the transition to other forms of illicit drug use.
These conclusions are of course subject to a number of caveats. First, the findings apply to
a particular cohort studied in a specific social context, and the extent to which these results apply to
other cohorts and social context is unknown. Second, the analyses are based on self-report data and
are subject to the usual limitations that apply to such data. Finally, the results of the regression
analysis are subject to the threats to validity from omitted sources of confounding.
However, within these limitations the findings of this 25-year longitudinal study suggest that
the development of illicit drug use and abuse in adolescence involved an accumulation of risk
factors including exposure to adversity in childhood, childhood adjustment, personality and
individual factors, the use of cannabis, affiliation with substance-using peers, and alcohol use, with
the most influential of these factors being the use of cannabis.
Acton, G.S., 2003. Measurement of impulsivity in a hierarchical model of personality traits:
implications for substance use. Subst Use Misuse 38, 67-83.
Adams, J.B., Heath, A.J., Young, S.E., Hewitt, J.K., Corley, R.P., Stallings, M.C., 2003.
Relationships between personality and preferred substance and motivations for use among
adolescent substance abusers. Am J Drug Alcohol Abuse 29, 691-712.
American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders
(4th ed.). American Psychiatric Association, Washington, DC.
Anderson, P., 2006. Global use of alcohol, drugs and tobacco. Drug Alcohol Rev 25, 489-502.
Andrews, G., Peters, L., 1998. The psychometric properties of the Composite International
Diagnostic Interview. Soc Psychiatry Psychiatr Epidemiol 33, 80-88.
Bahr, S.J., Hoffmann, J.P., Yang, X., 2005. Parental and peer influences on the risk of adolescent
drug use. J Prim Prev 26, 529-551.
Bloor, R., 2006. The influence of age and gender on drug use in the United Kingdom--a review. Am
J Addict 15, 201-207.
Boden, J.M., Fergusson, D.M., Horwood, L.J., 2006. Illicit drug use and dependence in a New
Zealand birth cohort. Aust NZ Psychiatry 40, 156-163.
Botvin, G.J., Scheier, L.M., Griffin, K.W., 2002. Preventing the onset and developmental
progression of adolescent drug use. In: Kandel, D.B. (Ed.), Stages and pathways of drug
involvement: examining the gateway hypothesis. Cambridge University Press, New York. pp. 115-
Carlin, J.B., Wolfe, R., Coffey, C., Patton, G.C., 1999. Tutorial in Biostatistics. Analysis of binary
outcomes in longitudinal studies using weighted estimating equations and discrete-time survival
methods: Prevalence and incidence of smoking in an adolescent cohort. Stat Med 18, 2655-2679.
Center On Addiction And Substance Abuse, 1994. Cigarettes, alcohol, marijuana: Gateways to
illicit drug use. Center On Addiction And Substance Abuse, Columbia University, New York.
Chilcoat, H.D., Anthony, J.C., 1996. Impact of parent monitoring on initiation of drug use through
later childhood. Journal of the American Academy of Child and Adolsecent Psychiatry 35, 91-100.
Christoffersen, M.N., Soothill, K., 2003. The long-term consequences of parental alcohol abuse: a
cohort study of children in Denmark. J Subst Abuse Treat 25, 107-116.
Cloninger, C.R., 1987. A systematic method for clinical description and classification of personality
variants. A proposal. Arch Gen Psychiatry 44, 573-588.
Coffey, C., Lynskey, M., Wolfe, R., Patton, G.C., 2000. Initiation and progression of cannabis use
in a population-based Australian adolescent longitudinal study. Addiction 95, 1679-1690.
Compton, W.M., Thomas, Y.F., Conway, K.P., Colliver, J.D., 2005. Developments in the
epidemiology of drug use and drug use disorders. Am J Psychiatry 162, 1494-1502.
Conners, C.K., 1969. A teacher rating scale for use in drug studies with children. Am J Psychiatry
Conners, C.K., 1970. Symptom patterns in hyperkinetic, neurotic and normal children. Child Dev
Deas, D., Brown, E.S., 2006. Adolescent substance abuse and psychiatric comorbidities. J Clin
Psychiatry 67, e02.
Donovan, J.E., Jessor, R., 1985. Structure of problem behavior in adolescence and young adulthood.
J Consult Clin Psychol 53, 890-904.
Dube, S.R., Anda, R.F., Felitti, V.J., Edwards, V.J., Williamson, D.F., 2002. Exposure to abuse,
neglect, and household dysfunction among adults who witnessed intimate partner violence as
children: implications for health and social services. Violence Vict 17, 3-17.
Duncan, S.C., Duncan, T.E., Biglan, A., Ary, D., 1998. Contributions of the social context to the
development of adolescent substance use: a multivariate latent growth modeling approach. Drug
Alcohol Depend 50, 57-71.
Evren, C., Evren, B., Yancar, C., Erkiran, M., 2007. Temperament and character model of
personality profile of alcohol- and drug-dependent inpatients. Compr Psychiatry 48, 283-288.
Felitti, V.J., Anda, R.F., Nordenberg, D., Williamson, D.F., Spitz, A.M., Edwards, V., Koss, M.P.,
Marks, J.S., 1998. Relationship of childhood abuse and household dysfunction to many of the
leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study. Am J Prev
Med 14, 245-258.
Fergusson, D.M., Boden, J.M., Horwood, L.J., 2006. Cannabis use and other illicit drug use:
Testing the cannabis gateway hypothesis. Addiction 101, 556-569.
Fergusson, D.M., Horwood, L.J., 1993. The structure, stability and correlations of the trait
components of conduct disorder, attention deficit and anxiety/withdrawal reports. J Child Psychol
Psychiatry 34, 749-766.
Fergusson, D.M., Horwood, L.J., 1999. Prospective childhood predictors of deviant peer affiliations
in adolescence. J Child Psychol Psychiatry 40, 581-592.
Fergusson, D.M., Horwood, L.J., 2000. Cannabis use and dependence in a New Zealand birth
cohort. NZ Med J 113, 156-158.
Fergusson, D.M., Horwood, L.J., 2001. The Christchurch Health and Development Study: Review
of findings on child and adolescent mental health. Aust NZ Psychiatry 35, 287-296.
Fergusson, D.M., Horwood, L.J., Lloyd, M., 1991. Confirmatory factor models of attention deficit
and conduct disorder. J Child Psychol Psychiatry 32, 257-274.
Fergusson, D.M., Horwood, L.J., Ridder, E.M., 2007. Conduct and attentional problems in
childhood and adolescence and later substance use, abuse and dependence: Results of a 25 year
longitudinal study. Drug Alcohol Depend 88S, S14-S26.
Fergusson, D.M., Horwood, L.J., Shannon, F.T., Lawton, J.M., 1989. The Christchurch Child
Development Study: A review of epidemiological findings. Paediatr & Perinatal Epidemiol 3, 278-
Fergusson, D.M., Lynskey, M.T., 1997. Physical punishment/maltreatment during childhood and
adjustment in young adulthood. Child Abuse Negl 21, 617-630.
Fergusson, D.M., Lynskey, M.T., Horwood, L.J., 1996. Childhood sexual abuse and psychiatric
disorder in young adulthood: I. Prevalence of sexual abuse and factors associated with sexual abuse.
J Am Acad Child Psychiatry 35, 1355-1364.
Fergusson, D.M., Woodward, L.J., Horwood, L.J., 1999. Childhood peer relationship problems and
young people's involvement with deviant peers in adolescence. J Abnorm Child Psychol 27, 357-
Flory, K., Milich, R., Lynam, D.R., Leukefeld, C., Clayton, R., 2003. Relation between childhood
disruptive behavior disorders and substance use and dependence symptoms in young adulthood:
Individuals with symptoms of attention-deficit/hyperactivity disorder and conduct disorder are
uniquely at risk. Psychol Addict Behav 17, 151-158.
Galea, S., Nandi, A., Vlahov, D., 2004. The social epidemiology of substance use. Epidemiol Rev
Gibbons, R.D., Hedeker, D., Waternaux, C., Davis, J.M., 1988. Random effects regression models:
A comprehensive approach to the analysis of longitudinal psychiatric data. Psychopharmacological
Bulletin 24, 438-443.
Hawkins, J.D., Catalano, R.F., Miller, J.Y., 1992. Risk and protective factors for alcohol and other
drug problems in adolescence and early adulthood: implications for substance abuse prevention.
Psychol Bull 112, 64-105.
Hayatbakhsh, M.R., Kinner, S.A., Jamrozik, K., Najman, J.M., Mamun, A.A., 2007. Maternal
partner criminality and cannabis use in young adulthood: prospective study. Aust N Z J Psychiatry
Hays, R.D., Widaman, K.F., DiMatteo, M.R., Stacy, A.W., 1987. Structural equation models of
current drug use: Are appropriate models so simple(x)? J Pers Soc Psychol 52, 134-144.
Huba, G.J., Wingard, J.A., Bentler, P., 1981. A comparison on two latent variable causal models of
adolescent drug use. J Pers Soc Psychol 40, 180-193.
Jenkins, J.E., 1996. The influence of peer affiliation and student activities on adolescent drug
involvement. Adolescence 31, 297-306.
Johnson, J.L., Leff, M., 1999. Children of substance abusers: overview of research findings.
Pediatrics 103, 1085-1099.
Kandel, D.B., 2003. Does marijuana use cause the use of other drugs? JAMA 289, 482-483.
Kandel, D.B., Chen, K., Warner, L.A., Kessler, R.C., Grant, B., 1997. Prevalence and demographic
correlates of symptoms of last year dependence on alcohol, nicotine, marijuana and cocaine in the
US population. Drug Alcohol Depend 44, 11-29.
Kandel, D.B., Yamaguchi, K., 2002. Stages of drug involvement in the US population. In: Kandel,
D.B. (Ed.), Stages and pathways of drug involvement: examining the gateway hypothesis.
Cambridge University Press, New York. pp. 65-89.
Kandel, D.B., Yamaguchi, K., Chen, K., 1992. Stages of progression in drug involvement from
adolescence to adulthood: Further evidence for the gateway theory. J Stud Alcohol 53, 447-457.
Kandel, D.B., Yamaguchi, K., Klein, L.C., 2006. Testing the gateway hypothesis. Addiction 101,
Keyes, M.A., Iacono, W.G., McGue, M., 2007. Early onset problem behavior, young adult
psychopathology, and contextual risk. Twin Res Hum Genet 10, 45-53.
Khan, A.A., Jacobson, K.C., Gardner, C.O., Prescott, C.A., Kendler, K.S., 2005. Personality and
comorbidity of common psychiatric disorders. Br J Psychiatry 186, 190-196.
Kilpatrick, D.G., Acierno, R., Saunders, B., Resnick, H.S., Best, C.L., Schnurr, P.P., 2000. Risk
factors for adolescent substance abuse and dependence: data from a national sample. J Consult Clin
Psychol 68, 19-30.
Lynskey, M., White, V., Hill, D., Letcher, T., Hall, W., 1999. Prevalence of illicit drug use among
youth: results from the Australian School Students' Alcohol and Drugs Survey. Aust N Z J Public
Health 23, 519-524.
Lynskey, M.T., Heath, A.C., Bucholz, K.K., Slutske, W.S., Madden, P.A.F., Nelson, E.C., Statham,
D.J., Martin, N.G., 2003. Escalation of drug use in early-onset cannabis users vs co-twin controls.
JAMA 289, 427-433.
MacCoun, R., 1998. In what sense (if any) is marijuana a gateway drug?
http://www.fas.org/drugs/issue4.htm#gateway. Accessed on 9 March 2005.
MacCoun, R., 2006. Competing accounts of the gateway effect: The field thins, but still no clear
winner. Addiction 101, 470-476.
Mayes, L.C., Suchman, N., 2006. Developmental pathways to substance abuse. In: Cicchetti, D.,
Cohen, D. (Eds.), Developmental Psychopathology. John Wiley and Sons, New York. pp. 599-619.
Merikangas, K.R., Stolar, M., Stevens, D.E., Goulet, J., Preisig, M.A., Fenton, B., Zhang, H.,
O'Malley, S.S., Rounsaville, B.J., 1998. Familial transmission of substance use disorders. Arch Gen
Psychiatry 55, 973-979.
Miller, B.A., Maguin, E., Downs, W.R., 1997. Alcohol, drugs, and violence in children's lives.
Recent Dev Alcohol 13, 357-385.
Molina, B.S.G., Pelham Jr, W.E., 2003. Childhood predictors of adolescent substance use in a
longitudinal study of children with ADHD. J Abnorm Psychol 112, 497-507.
Morral, A.R., McCaffrey, D.F., Paddock, S.M., 2002. Reassessing the marijuana gateway effect.
Addiction 97, 1493-1504.
Mulder, R.T., 2002. Alcoholism and personality. Aust N Z J Psychiatry 36, 44-52.
Pears, K., Capaldi, D.M., Owen, L.D., 2007. Substance use risk across three generations: the roles
of parent discipline practices and inhibitory control. Psychol Addict Behav 21, 373-386.
Poulton, R., Moffitt, T.E., Harrington, H., Milne, B.J., Caspi, A., 2001. Persistence and perceived
consequences of cannabis use and dependence among young adults: implications for policy. NZ
Med J 114, 544-547.
Putnam, F.W., Trickett, P.K., 1993. Child sexual abuse: a model of chronic trauma. Psychiatry 56,
Rehm, J., Taylor, B., Room, R., 2006. Global burden of disease from alcohol, illicit drugs and
tobacco. Drug Alcohol Rev 25, 503-513.
Ripple, C.H., Luther, S.S., 1996. Familial factors in illicit drug abuse: an interdisciplinary
perspective. Am J Drug Alcohol Abuse 22, 147-172.
Rodham, K., Hawton, K., Evans, E., Weatherall, R., 2005. Ethnic and gender differences in
drinking, smoking and drug taking among adolescents in England: a self-report school-based survey
of 15 and 16 year olds. J Adolesc 28, 63-73.
Rutter, M., Tizard, J., Whitmore, K., 1970. Education, Health and Behaviour. Longmans, London.
Simpson, T.L., Miller, W.R., 2002. Concomitance between childhood sexual and physical abuse
and substance use problems. A review. Clin Psychol Rev 22, 27-77.
Staiger, P.K., Kambouropoulos, N., Dawe, S., 2007. Should personality traits be considered when
refining substance misuse treatment programs? Drug Alcohol Rev 26, 17-23.
Stallings, M.C., Cherny, S.S., Young, S.E., Miles, D.R., Hewitt, J.K., Fulker, D.W., 1997. The
familial aggregation of depressive symptoms, antisocial behavior, and alcohol abuse. Am J Med
Genet 74, 183-191.
Straus, M., 1979. Measuring intrafamily conflict and violence: The conflict tactics (CT) scale.
Journal of Marriage & the Family 41, 75-88.
Substance Use and Mental Health Services Administration, 2007. Results from the 2006 National
Survey on Drug Use and Health: National findings. In: Office of Applied Studies (Ed.), NSDUH
Series H-32. United States Department of Health and Human Services, Rockville, MD.
Swift, W., Hall, W., Teesson, M., 2001. Cannabis use and dependence among Australian adults:
results from the National Survey of Mental Health and Wellbeing. Addiction 96, 737-748.
van der Zwaluw, C.S., Scholte, R.H., Vermulst, A.A., Buitelaar, J.K., Verkes, R.J., Engels, R.C.,
2008. Parental problem drinking, parenting, and adolescent alcohol use. J Behav Med.
Viveros, M.P., Marco, E.M., File, S.E., 2006. Nicotine and cannabinoids: parallels, contrasts and
interactions. Neurosci Biobehav Rev 30, 1161-1181.
von Sydow, K., Lieb, R., Pfister, H., Hoefler, M., Sonntag, H., Wittchen, H.U., 2001. The natural
course of cannabis use, abuse and dependence over four years: a longitudinal community study of
adolescents and young adults. Drug Alcohol Depend 64, 347-361.
Widom, C.S., Marmorstein, N.R., White, H.R., 2006. Childhood victimization and illicit drug use in
middle adulthood. Psychol Addict Behav 20, 394-403.
Wills, T.A., Windle, M., Cleary, S.D., 1998. Temperament and novelty seeking in adolescent
substance use: convergence of dimensions of temperament with constructs from Cloninger's theory.
J Pers Soc Psychol 74, 387-406.
Wilsnack, S.C., Vogeltanz, N.D., Klassen, A.D., Harris, T.R., 1997. Childhood sexual abuse and
women's substance abuse: national survey findings. J Stud Alcohol 58, 264-271.
World Health Organization, 2000. Guide to Drug Use Epidemiology. World Health Organization,