Who becomes cannabis dependent soon after onset of use? Epidemiological evidence from the United States: 2000–2001

https://doi.org/10.1016/j.drugalcdep.2004.11.014Get rights and content

Abstract

In this paper we estimate the risk of becoming cannabis dependent within 24 months after first use of cannabis and examine subgroup variation in this risk. The study estimates are based on the National Household Survey on Drug Abuse conducted during 2000–2001, with a representative sample of U.S. residents ages 12 and older (n = 114,241). A total of 3352 respondents were found to have used cannabis for the first time within a span of up to 24 months prior to assessment. An estimated 3.9% of these recent-onset users developed a cannabis dependence syndrome during the interval since first use (median interval duration ∼12 months). Excess risk of cannabis dependence was found for those with cannabis onset before late-adolescence, those with family income less than US$ 20,000, and those who had used three or more drugs before the first use of cannabis (i.e., tobacco, alcohol, and other drugs). While these findings generally support previous study results, this study's focus on recent-onset users more closely approximates prospective and longitudinal research on the incidence (risk) of becoming cannabis dependent soon after onset of cannabis use, removing the influence of users with long-sustained or persistent cannabis dependence developed years ago.

Introduction

As one of the most commonly consumed illegal drugs worldwide, cannabis has become a public health concern in recent decades (e.g., see Dennis et al., 2002, Patton et al., 2002). Recent epidemiological studies in Australia, Germany, New Zealand, and the United Kingdom suggest that about one to two thirds of young people in these countries have tried cannabis at least once prior to young adulthood (e.g., see Fergusson and Horwood, 2000, Swift et al., 2001; von Sydow et al., 2001, Boys et al., 2003). Within the U.S., during calendar year (CY) 2001, an estimated 21 million community-dwelling residents aged 12 or older used cannabis at least once, and over 50,000 adolescents started to use cannabis in the same year (SAMHSA, 2002). The more recent surveys of cannabis use also have assessed an array of cannabis-related problems—especially problems in the domain of clinical features of cannabis dependence. Accordingly, in this paper, the main aims are to estimate the risk of becoming a case of cannabis dependence during an interval within 24 months after the start of first use, and to investigate characteristics and conditions associated with becoming cannabis dependent soon after onset of cannabis use. These estimates will help shed light on the suspected causal determinants of cannabis dependence in the first months and initial 2 years of cannabis use. To achieve these aims, we have turned to recently available epidemiological data with nationally representative samples drawn from the 50 U.S.

Over several decades, a number of research teams have investigated the extent of cannabis-related problems in the U.S. (e.g., Anthony and Helzer, 1991, Anthony et al., 1994, Newcomb, 1996, Chen et al., 1997, Kandel et al., 1997). For example, based upon lifetime history data from the National Comorbidity Survey conducted between 1990 and 1992, an estimated 4% of non-institutionalized civilians age 15–54 had become cannabis dependent by the date of assessment; moreover, for every 11 people who used cannabis at least once, one user (9%) developed the full DSM cannabis dependence syndrome, mainly during the first 10 years after onset of cannabis use, with a peak risk value some 1–3 years after first use (Anthony et al., 1994).

The same methodological approach with lifetime history data indicates that some 1–3% of users developed cannabis dependence within 12 months after onset of cannabis use, and another 1–2% of users developed cannabis dependence in the next year (i.e., 2–5% within 24 months after onset). Nonetheless, this approach requires users to reconstruct their histories of cannabis use and problems over a recall span across decades of life (Wagner and Anthony, 2002a). Cross-sectional ‘point prevalence’ estimates from the 1991–1993 National Household Surveys on Drug Abuse (NHSDA) suggested that approximately 7% of recently-active cannabis users now have a cannabis dependence syndrome (i.e., among users with at least one occasion of use in the year prior to assessment; see Chen et al., 1997). A prospective study of youths in California, with a start date in the late 1970s, found that about one in four cannabis users had developed the full-blown DSM-III-R syndrome of cannabis dependence during the 12 year span from adolescence through young adulthood (Newcomb, 1996). If these risks were to be distributed equally over all 12 years of follow-up, then one might project about 4% had become dependent every 24 months of the study interval.

Of course, these transition probabilities depend in part upon reinforcing functions served by cannabis and its chemical constituents, but they also depend upon conditions and processes that will change from time to time and from place to place. Some observers claim that increasing delta-9-tetrahydrocannabinol (THC) content of cannabis means that cannabis use now might be more likely to result in cannabis dependence than was true in the past (e.g., see Ashton, 2001). Crucial evidence on this topic requires a comparison of two risk estimates: (1) the risk of becoming cannabis dependent soon after onset of cannabis use as observed in recent years, which is a focus of the present study, and (2) the corresponding risk estimates observed in past years. Making such a comparison, Compton and colleagues (2004) found some evidence suggestive of an increased occurrence of cannabis dependence in years 2000–2001 as compared to corresponding estimates 10 years ago.

Several population-based studies have investigated suspected causal determinants of cannabis dependence. Using data from the 1991–1993 NHSDA, Kandel et al. (1997) identified an array of demographic characteristics associated with prevalence of being a recently active case of cannabis dependence, including age, sex, and race/ethnicity group. Other research teams have found that higher prevalence of cannabis dependence is associated with male sex, higher levels of education, unemployment, as well as prior levels of alcohol and tobacco use (Offord et al., 1996, Chen et al., 1997, Swift et al., 1998, Swift et al., 2001, Wagner and Anthony, 2002a; Young et al., 2002). Elsewhere, some research teams have tried to identify the causal determinants of becoming cannabis dependent among cannabis users. For instance, based upon epidemiological data prospectively gathered in Australia, Coffey et al. (2003) found that males and persistent tobacco smokers were more likely to develop the full-blown syndrome of cannabis dependence. von Sydow et al. (2002) examined an inclusive list of socioenvironmental, drug use, and intrapersonal and interpersonal correlates in relation to the risk of becoming cannabis dependent among adolescents and young adults in Germany and found that male sex, younger age, deprived socio-economic status, and other illegal drug involvement were prominent predictors for the subsequent onset of cannabis dependence. Prospective studies of this type are underway in the U.S., as exemplified by Newcomb's follow-up of adolescents in California (e.g., Newcomb, 1996), and our own work in the mid-Atlantic States (e.g., Rosenberg and Anthony, 2001). However, none of the published prospective studies has had an epidemiological sample of national scope within the U.S.

It is important to consider in research of this type that the prevalence of recently active cannabis dependence is the product of two processes; one process leads to the occurrence of an incident case of cannabis dependence, and the other process leads to the persistence of cannabis use, perhaps after cannabis dependence has already formed. Hence, most prior studies have disclosed useful information about the correlates of being cannabis dependent, but the study evidence does not indicate whether these characteristics are associated only with persistence of cannabis use once dependence develops, and not with risk of becoming dependent.

With nationally representative samples drawn in the U.S., in this study we try to shed new light on the risk of becoming cannabis dependent, with a focus upon the risk of becoming an incident case of cannabis dependence within the first 24 months after first use, and to constrain the already-mentioned survey errors that become most prominent with a long-span retrospective approach. The methods used in this study are based upon an approach that members of our research team and others have described in a series of papers (Gfroerer and Brodsky, 1992, Anthony and Petronis, 1995, Chen and Anthony, 2004, O’Brien and Anthony, in press). The method of identifying recent-onset cannabis users is one that involves taking differences between a respondent's age at assessment and his or her age of onset of first cannabis use. With this focus on an interval of up to 24 months after onset of cannabis use, we also are able to narrow the investigation to the incidence process, while reducing focus on cases with long-sustained use. The conceptual advantage is one that constrains a possible reciprocity—that is, at the start of cannabis use, there is no cannabis dependence. However, once cannabis use occasions mount up in number, the dependence process begins to influence the likelihood of each subsequent occasion of use and becomes a causal determinant of persistent use (Brady, 1989, Chen and Anthony, 2003).

Section snippets

Participants

The data for this study are from public use data sets produced and released after two recent years of the National Household Surveys on Drug Abuse: 2000–2001, with an aggregate community sample size of 114,241. The NHSDA (recently re-named the National Surveys on Drug Use and Health; NSDUH) are ongoing cross-sectional surveys of the drug experiences of U.S. citizens aged 12 years and over, and detailed methodological descriptions have already been published (SAMHSA, 2001, SAMHSA, 2002). In

Results

In the 2000–2001 NHSDA sample, about one in three respondents (n = 40,438; 35%) had used cannabis at least once in their lifetimes. Sociodemographic characteristics of subsamples within the study population, stratified by recency of cannabis onset, are shown in Table 1 (weighted). Overall, roughly 1% of the study population had become a recent-onset cannabis user, with a 95% CI = 1.05%, 1.15% (unweighted number of recent-onset users, n = 3352).

The distinction between risk of becoming a cannabis user

Discussion

Based upon data from recent nationally representative samples of civilians in the U.S., the main findings of this population-based epidemiological study may be summarized as follows. First, approximately one in 25 cannabis users developed a cannabis dependence syndrome within the 24 month interval after the first use (3.9%; 95% CI = 3%, 5%). This value lies within the interval of 2–5% that Wagner and Anthony (2002a) derived by focusing on the first 24 months of cannabis use in a lifetime

Acknowledgements

The work was supported by NIDA awards R01DA09897, T32DA07292, K05DA015799 (JCA), and a Development Award from the Johns Hopkins University Center for Adolescent Health Promotion and Disease Prevention (CYC). Data reported herein come from national survey data collected under the auspices of the Office of Applied Studies, Substance Abuse and Mental Health Services Administration.

References (52)

  • Anthony, J.C. The epidemiology of cannabis dependence. In Roffman, R., Stephens, R. (Eds.). Cannabis dependence: Its...
  • J.C. Anthony et al.

    Syndromes of drug abuse and dependence

  • J.C. Anthony et al.

    Do I do what I say? A perspective on self-report methods in drug dependence epidemiology

  • J.C. Anthony et al.

    Comparative epidemiology of dependence of tobacco, alcohol, controlled substance, and inhalants: basic findings from the National Comorbidity Survey

    Exp. Clin. Psychopharmacol.

    (1994)
  • C.H. Ashton

    Pharmacology and effects of cannabis

    Br. J. Psychiatry

    (2001)
  • A. Boys et al.

    Psychiatric morbidity and substance use in young people aged 13–15 years: results from the Child and Adolescent Survey of Mental Health

    Br. J. Psychiatry

    (2003)
  • Brady, J.V., 1989. Issues in human drug abuse liability testing: overview and prospects for the future. In: Fischman,...
  • J.S. Brook et al.

    Early adolescent marijuana use: risks for the transition to young adulthood

    Psychol. Med.

    (2002)
  • B.J.L. Buchan et al.

    Cannabis use: consistency and validity of self-report, on-site urine testing and laboratory testing

    Addiction

    (2002)
  • C.-Y. Chen et al.

    Possible age-associated bias in reporting of clinical features of drug dependence: epidemiological evidence on adolescent-onset marijuana use

    Addiction

    (2003)
  • C.-Y. Chen et al.

    Epidemiological estimates of risk in the process of becoming dependent upon cocaine: cocaine hydrochloride powder versus crack cocaine

    Psychopharmacology

    (2004)
  • C. Coffey et al.

    Adolescent precursors of cannabis dependence: findings from the Victorian Adolescent Health Cohort Study

    Br. J. Psychiatry

    (2003)
  • W.M. Compton et al.

    Prevalence of marijuana use disorders in the United States: 1991–1992 and 2001–2002

    JAMA

    (2004)
  • M. Dennis et al.

    Changing the focus: the case for recognizing and treating cannabis use disorders

    Addiction

    (2002)
  • D.M. Fergusson et al.

    Early onset cannabis use and psychosocial adjustment in young adults

    Addiction

    (1997)
  • D.M. Fergusson et al.

    Cannabis use and dependence in a New Zealand birth cohort

    NZ Med. J.

    (2000)
  • Cited by (0)

    An alternate presentation of unweighted data can be found by accessing the online version of this paper at http://dx.doi.org by entering doi:10.1016/j.drugalcdep.2004.11.014.

    View full text