Elsevier

Drug and Alcohol Dependence

Volume 75, Issue 3, 6 September 2004, Pages 287-299
Drug and Alcohol Dependence

Effects of age at first substance use and psychiatric comorbidity on the development of substance use disorders

https://doi.org/10.1016/j.drugalcdep.2004.03.013Get rights and content

Abstract

In this paper, we examine the effects of age at first substance use, and history of psychiatric disorders, on the development of substance use disorder (SUD) by age 16. We use a prospective, longitudinal design to disaggregate the effects of age at first use and time since first use on the development of adolescent SUD. Second, we test the hypothesis that adolescent SUD is an unlikely progression from early substance use unless children also show other early conduct problems. A population sample of 1420 children from the Great Smoky Mountains Study (GSMS) was assessed annually between ages 9 and 16. Logistic regression models were applied within the hierarchical Bayesian framework, where the covariate effects were described by time-varying parameters having a first-order auto-regressive prior distribution. Posterior analyses based on a Gibbs sampling approach revealed that, controlling for years of exposure, the risk of transition to SUD increased with age at onset for onsets before age 13, but began to fall for onset at 14. Among users, use alone, without early conduct problems, led to a 11% prevalence of SUD by age 16. Past conduct disorder (CD) had a strong additive effect at ages 13–15, but at age 16, when substance use and abuse became more normative, the excess risk from prior CD decreased. Boys, but not girls, with a history of depression were at increased risk of SUD. Anxiety increased the risk of SUD in girls at age 16, but not before that. Results only partially support the study hypothesis; early use was a major predictor of adolescent SUD even in the absence of CD.

Introduction

In this paper, we first examine what Anthony and Petronis called the “uncertainty whether early-onset drug use exerts an independent effect on risk of later drug problems, even though it might presage and serve as a vulnerability marker for later problems”(Anthony and Petronis, 1995). Using retrospective data, they showed that the risk of adult drug problems was linearly related to age at onset of drug use, being twice as high in adults reporting first use before age 13 as in those with first use after age 17. However, the time from first use to problem use was around 4 years, irrespective of age at first use among problem users. Here we look at the situation in adolescence, but using prospective data. Many previous research findings have established that early onset of drug use is associated with later problem use (Andreasson et al., 1992, Boyle et al., 1992, DeWit et al., 2000, Kandel et al., 1986, Kaplan et al., 1986, Mills and Noyes, 1984, Robins and Murphy, 1967, Robins and Przybeck, 1985, Yamaguchi and Kandel, 1984). We have found no prospective study testing for a linear relationship between age at onset and probability of later problem drug use in adolescents, but retrospective analysis of the Ontario Mental Health Supplement data (DeWit et al., 2000)found a nonlinear relationship, with the highest risk in those reporting first use at 13–14 for alcohol abuse and 11–12 for alcohol dependence.

Second, we explore the effects of early psychiatric disorders, including conduct disorder (CD), on both the likelihood and the timing of the transition from substance use to substance use disorders (SUD). It is well established that early problem behavior is associated with later drug use and abuse (Anderson et al., 1989, Jessor and Jessor, 1977, Kandel et al., 1978, Kaplan, 1980, Monnelly et al., 1983, Smith and Fogg, 1979, Vaillant and Milofsky, 1982, Wingard et al., 1979). However, most of the studies cited start in adolescence. In this paper we examine an earlier developmental period, beginning at age 9 and following children through to age 16. We examine the effects of a range of psychiatric disorders with onset by age 15 on risk for substance use disorders by age 16.

In her analysis of the (retrospective) Epidemiologic Catchment Area (ECA) data, Robins proposed a more specific hypothesis than one of a general association between conduct problems and SUD.

Abuse is extremely rare for those free of (early) conduct problems, no matter how early substance use began. At every other level of conduct problems, however, the earlier that use begins, the greater is the likelihood of substance abuse …. For those first using substances before age 20, the number of conduct problems was an even better predictor of substance abuse than was age of onset (before or after age 15). Among those beginning substance use before age 15 with seven or more conduct problems, more than half developed substance abuse; with only one conduct problem, only 5% did so. When first use occurred between ages 15 and 19, there is still a large effect from number of conduct problems, but the control for age of first use somewhat reduced their impact.”(Robins and McEvoy, 1990) (p. 196).

If it is the case that early CD increases the risk of adult SUD, a logical extension is to expect early CD also to drive the onset of adolescent SUD. Here we test this hypothesis, and also examine the impact of other psychiatric disorders on the transition from use to abuse/dependence.

The present study also adds to the literature in the following ways: (1) It follows a representative population sample of youth prospectively from childhood into adolescence. (2) It uses structured psychiatric interviews with parent and child to generate a range of DSM-IV psychiatric diagnoses. (3) The sample contains both boys and girls, unlike many previous studies. (4) The statistical methodology incorporates the effects of psychiatric comorbidity on transitions from the prior use of substances to SUD. (5) Use of a hierarchical Bayesian modeling approach permits the analysis of a complex, stratified data set that addresses some issues of dynamic age and covariate effects.

Section snippets

Study population

The Great Smoky Mountains Study (GSMS) is a longitudinal study of the development of psychiatric disorder and substance use and abuse in a predominantly rural area of the southeastern United States. Full details of the study design and instruments used can be found elsewhere (Costello et al., 1996). Briefly, a representative sample of 9, 11, and 13-year olds was recruited from 11 counties in western North Carolina using a household equal probability design. A two-phase sampling procedure was

Rates of onset

Table 1 shows the reported rates of age specific incidence for psychiatric disorders, substance use, and substance use disorder at each age. The rate of onset at a given age was computed by dividing the weighted number of reported onsets by the weighted total number of subjects interviewed since the age of onset after removing subjects who had onsets previously. There was no SUD before 12, but onsets increased rapidly thereafter, affecting 9.25% of the sample by age 16. Among the substance use

Conclusions

The goal of these analyses was to test some of the ideas about the development of substance use disorders that scholars have derived mainly from the scrutiny of studies using retrospective recall of ages at onset of both substance use and other psychiatric disorders. We adopted a hierarchical Bayesian dynamic modeling approach to assess the precise developmental impact of risk factors on SUD, showing that the effects evolved over time. For example, if a static regression such as GEE had been

Acknowledgements

This research was supported by grants from the National Institute of Drug Abuse (1R01DA11301-01) and the William T. Grant Foundation, and a National Institutes of Health Independent Scientist Award to the last author.

References (59)

  • American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders Fourth Edition (DSM-IV)....
  • T. Anderson et al.

    Patterns of adjustment problems and alcohol abuse in early adulthood: a prospective longitudinal study

    Dev. Psychopathol.

    (1989)
  • S. Andreasson et al.

    Antecedents and covariates of high alcohol consumption in young men

    Alcohol Clin. Exp. Res.

    (1992)
  • A. Angold et al.

    A test-retest reliability study of child-reported psychiatric symptoms and diagnoses using the Child and Adolescent Psychiatric Assessment (CAPA-C)

    Psychol. Med.

    (1995)
  • A. Angold et al.

    The Child and Adolescent Psychiatric Assessment (CAPA)

    Psychol. Med.

    (1995)
  • Bernardo, J., Smith, A., 1994. Bayesian Theory. John Wiley and Sons, New...
  • M.H. Boyle et al.

    Predicting substance use in late adolescence: results from the Ontario Child Health Study Follow-up

    Am. J. Psychiatr.

    (1992)
  • J. Brook et al.

    Consequences of adolescent drug use on psychiatric disorders in early adulthood

    Ann. Med.

    (2000)
  • C. Cargnoni et al.

    Bayesian forecasting of multinomial time series through conditionally Gaussian dynamic models

    J. Am. Stat Assoc.

    (1997)
  • G. Casella et al.

    Explaining the Gibbs sampler

    Am. Stat.

    (1992)
  • E.J. Costello et al.

    The Great Smoky Mountains Study of Youth: goals, designs, methods, and the prevalence of DSM-III-R disorders

    Arch. Gen. Psychiatr.

    (1996)
  • D. Cox

    Statistical analysis of time series: some recent developments

    Scand. J. Stat.

    (1981)
  • L. Dalagija-Ramic et al.

    Zavod za alkohilizam i druge toksikomanje Kantona Sarajevo

    Medicinski Arch.

    (2001)
  • B.S. Dansky et al.

    Comorbidity of bulimia nervosa and alcohol use disorders: results from the National Women’s Study

    Int. J. Eat. Disord.

    (2000)
  • G.L. De Bernardo et al.

    Comorbid psychiatric and alcohol abuse/dependence disorders: psychosocial stress, abuse, and personal history factors of those in treatment

    J. Addict. Dis.

    (2002)
  • N. Dedobbeleer et al.

    Outcomes of an ecological and participatory approach to prevent alcohol and other drug “abuse” among multiethnic adolescents

    Substance Use Misuse

    (2001)
  • D. DeWit et al.

    Age at first alcohol use: a risk factor for the development of alcohol disorders

    Am. J. Psychiatr.

    (2000)
  • Diggle, P.J., Liang, K.Y., Zeger, S.L., 1994. Analysis of Longitudinal Data. Clarendon Press,...
  • D.M. Fergusson et al.

    The short-term consequences of early onset cannabis use

    J. Abnorm. Child Psychol.

    (1996)
  • Cited by (150)

    • Moderate adolescent chronic intermittent ethanol exposure sex-dependently disrupts synaptic transmission and kappa opioid receptor function in the basolateral amygdala of adult rats

      2021, Neuropharmacology
      Citation Excerpt :

      Young adolescence appears to be a crucial time for the initiation of problematic alcohol use; many individuals who begin drinking during early adolescence tend to rapidly escalate their use, with more than ten percent accelerating to meet alcohol use disorder (AUD) criteria within the first year of initiation (Faden, 2006; Forman-Hoffman et al., 2017; Masten et al., 2009). This early introduction to alcohol use has been associated with numerous long-lasting alterations to brain function, brain structure, and behavior (Crews et al., 2016; Spear, 2018; Squeglia et al., 2014), including increased rates of anxiety, depression, and AUD later on (Duncan et al., 1997; Grant and Dawson, 1997; Hawkins et al., 1997; Pitkanen et al., 2005; Sung et al., 2004; Wolford-Clevenger and Cropsey, 2019). Our understanding of the effects of adolescent alcohol comes largely from studies that have used exposures yielding BECs of 150–250 mg/dL, far exceeding the definition of binge-level exposure [80 mg/dL (NIAAA, 2004),].

    View all citing articles on Scopus
    View full text