One year follow-up of cannabis dependence among long-term users in Sydney, Australia
Introduction
Cannabis use is typically initiated in late adolescence, used experimentally or intermittently, and often discontinued by the mid to late 20s (Chen and Kandel, 1995, Hall et al., 1999). A minority proceed to regular, long-term use. Approximately one tenth to one third of those who use on a monthly or more frequent basis at around 20 years of age report doing so in their early thirties (Bachman et al., 1997, Kandel and Davies, 1992). Compared to the rich longitudinal literature on the natural history of regular alcohol use, there is little research on the natural history of regular cannabis use.
Kandel and colleagues’ landmark prospective research on the consequences of adolescent drug use has contributed important information on the initiation, escalation and cessation of cannabis and other drug use in the general population. An early age of cannabis use onset, for example, was the major predictor of near-daily cannabis use at age 29 (Kandel and Davies, 1992). While the prevalence of daily use declined into the thirties, cannabis showed the highest persistence of the illicit drugs. Predictors of continued use included being male, frequent use, using to enhance positive and reduce negative feelings, early onset of use, and consumption of other illicit drugs (Chen and Kandel, 1998).
The Monitoring the Future Project (Bachman et al., 1997), studied successive, nationally representative cohorts comprising more than 33 000 young American adults over two decades. It found remarkable individual stability in cannabis use from late adolescence to early adulthood (correlations of approximately 0.9). As with alcohol, there was a modest age-related decline in use from the mid-twenties. Increases in cannabis and alcohol use after high school were attributed to the new freedoms associated with leaving home, while the subsequent decrease reflected new responsibilities such as marriage and parenthood. Because of the low overall prevalence and frequency of cannabis use in these studies, these findings were of little relevance to patterns of dependence.
Longitudinal research on cannabis dependence is rare. One of the few population studies to prospectively address dependence followed 1000 school children on five occasions over 12 years (Newcomb, 1992). Longitudinal analyses of a subset of this group investigated predictors of cannabis abuse, a term incorporating DSM-III-R dependence symptoms and items tapping other problem domains. Although cannabis consumption in late adolescence significantly predicted consumption in adulthood, it did not explain abuse, which was predicted by measures of psychosocial vulnerability, risk and protection (e.g. education, religiosity, availability, adult and peer use patterns). However, because this study was conducted on an adolescent population it does not necessarily relate to the experiences of older, more chronic users.
Consumption in adolescence (age 15) was also a strong predictor of consumption in early adulthood (age 21) in a recent prospective study (Poulton et al., 1997). Initiation and escalation of use was more common than remission or decrease in this age group. Almost half of those with a DSM-III-R cannabis dependence diagnosis at 18 were still dependent at 21. Finally, Kandel et al. (1997) calculated estimates of persistence of cannabis dependence from North American epidemiological research. They concluded that nicotine was the most addictive substance with the highest rate of persistence (0.45), followed by alcohol (0.32) and cannabis (0.25). Thus, one in four people with a lifetime diagnosis met criteria for cannabis dependence in the past year.
Unfortunately, the small body of prospective research on regular marijuana users (e.g. Weller and Halikas, 1982, Halikas et al., 1983, Halikas et al., 1984, Weller and Halikas, 1985, Haas, 1987, Hendin et al., 1987) has not addressed the issue of dependence, although it has provided valuable insight into the patterns and experiences of cannabis use over time.
For example, a group of 100 regular users and 50 non-using friends were followed over a 6–8 year period, and changes in cannabis and other drug use patterns, and psychological functioning, were examined (Weller and Halikas, 1982, Halikas et al., 1983, Weller and Halikas, 1985). Users displayed general stability in the reported effects of cannabis, although there was a small decrease in the frequency of desirable effects of use. Use patterns had polarised to either daily or occasional use, although they were significantly correlated with original frequency. There was a general decline in the use of other drugs with the exception of alcohol, and their use was also correlated with that reported initially.
In an attempt to supplement the sparse longitudinal literature on patterns of cannabis use and dependence this paper presents data collected from a 1-year follow-up interview of a sample of long-term cannabis users recruited in Sydney, Australia, as part of an exploratory study of the characteristics, correlates and measurement of cannabis dependence (Swift et al., 1998a, Swift et al., 1998b).
Section snippets
Aims
The aims of this study were to: (i) examine patterns and contexts of cannabis use and dependence in long-term users over a 1-year period, and (ii) investigate baseline factors that may be associated with these variables at follow-up.
Subjects
Eligible subjects were 200 long-term cannabis users who had participated in a comprehensive interview about their patterns and experiences of cannabis use and dependence in 1995–1996 and had provided permission to be contacted for a follow-up interview 12 months later. The recruitment and characteristics of this convenience sample are described elsewhere (Swift et al., 1998a, Swift et al., 1998b). Briefly, participants were recruited from advertisements or by word of mouth. Inclusion in the
Predictors of participation in follow-up
The baseline (n=200) and follow-up (n=162) samples were compared to investigate characteristics which predicted participation in the 1-year follow-up interview. Univariate analyses revealed a trend towards older participants remaining in the sample (χ2MH, 1 d.f.=3.4, P=0.07). Those who were re-interviewed were also more likely to be female (89 versus 75% of males; χ2, 1 d.f.=6.5, P=0.01), employed (86 versus 75% who were unemployed; χ2, 1 d.f.=3.4, P=0.06), in a relationship (86 versus 75% of
Discussion
A respectable 81% of the baseline sample was re-interviewed at 1 year. While younger, polydrug using males were significantly less likely to be interviewed at follow-up differential attrition had no effect on the predictors of the follow-up variables.
Consistent with other research (e.g. Halikas et al., 1984), cannabis use patterns remained fairly stable, with at least moderate correlations between frequency and quantity of use at the two interviews. Longitudinal multivariate analyses revealed
Acknowledgements
This research was funded by a National Drug Strategy Research Scholarship, Commonwealth Department of Health and Family Services. Professors Robin Room, of The Centre for Social Research on Alcohol and Drugs, Sweden, and Ronald Kessler, of the Harvard Medical School, USA, gave permission to use the ICD-10 measure of dependence and the short UM-CIDI, respectively. Dr Michael Lynskey provided valuable advice on the longitudinal analyses. We also thank the study participants who agreed to be
References (37)
- et al.
Predictors of cessation of marijuana use: an event history analysis
Drug Alcohol Depend.
(1998) - et al.
Regular marijuana use and its effect on psychosocial variables: a longitudinal study
Compr. Psychiatry
(1983) - et al.
Use of marijuana and other drugs among adult marijuana users: a longitudinal study
Compr. Psychiatry
(1984) - et al.
Dependence symptoms but no diagnosis: diagnostic “orphans” in a community sample
Drug Alcohol Depend.
(1998) - et al.
Prevalence and demographic correlates of symptoms of last year dependence on alcohol, nicotine, marijuana and cocaine in the US population
Drug Alcohol Depend.
(1997) - et al.
Alcohol and Other Drug Use Among Ontario Adults in 1994 and Changes Since 1977
(1994) - et al.
Comparative epidemiology of dependence on tobacco, alcohol, controlled substances, and inhalants: basic findings from the National Comorbidity Survey
Exp. Clin. Psychopharm.
(1994) - et al.
Smoking, Drinking, and Drug Use in Young Adulthood: the Impacts of New Freedoms and New Responsibilities
(1997) An introduction to sample selection bias in sociological data
Am. Soc. Rev.
(1983)- et al.
The natural history of drug use from adolescence to the mid-thirties in a general population sample
Am. J. Public Health
(1995)
Application to the alcohol dependence syndrome of a method of determining the sequential development of symptoms
Psychol. Med.
Severity of dependence and route of administration of heroin, cocaine and amphetamines
Br. J. Addict.
The Severity of Dependence Scale (SDS): psychometric properties of the SDS in English and Australian samples of heroin, cocaine and amphetamine users
Addict
Test-retest reliability of the Severity of Dependence Scale
Addict
Long-term outcomes of heavy marijuana use among adolescents
Drug Alcohol Abuse Child. Adolesc. Pediatr.
The epidemiology of cannabis use and its consequences
Living High: Daily Marijuana Use Among Adults
Cited by (66)
Perceptions of effectiveness and believability of pictorial and text-only health warning labels for cannabis products among Canadian youth
2019, International Journal of Drug PolicyTesting the biosocial cognitive model of substance use in cannabis users referred to treatment
2019, Drug and Alcohol DependenceDevelopment and initial validation of a marijuana cessation expectancies questionnaire
2017, Drug and Alcohol DependenceGender Differences in Cannabis Use Disorders
2017, Handbook of Cannabis and Related Pathologies: Biology, Pharmacology, Diagnosis, and TreatmentTreatment seeking in cannabis dependence: The role of social cognition
2017, Drug and Alcohol DependenceCitation Excerpt :The SDS-C is a five-item questionnaire that assesses the degree of cannabis dependence. It is sensitive to severity of cannabis dependence (Swift et al., 2000). The cannabis dependence cut-off is 3 (possible range: 0–15; Swift et al., 1998).
Social cognitive predictors of treatment outcome in cannabis dependence
2017, Drug and Alcohol DependenceCitation Excerpt :Responses are rated on a 4-point Likert scale (0 = Never to 3 = Always). The SDS-C has good test-retest reliability and is sensitive to severity of cannabis dependence (Swift et al., 2000). Using Australian normative data, a score of ≥3 is indicative of DSM-IV cannabis dependence (Swift et al., 1998).