Long-term cannabis use: characteristics of users in an Australian rural area

Addiction (1998) 93(6), 837 ± 846

RESEARCH REPORT

Long-term cannabis use: characteristics ofusers in an Australian rural area

DAVID REILLY,1

PETER DIDCOTT,2

WENDY SWIFT3

& WAYNEHALL

3

1Northern Rivers Health Service, Lismore, New South Wales,

2Southern Cross University,

Lismore, New South Wales & 3National Drug and Alcohol Research Centre, University of

New South Wales, Australia

Abstract

Aim . To investigate the characteristics and patterns of cannabis and other drug use among long-te rm

cannabis users in an Australian rural area. Design. Cross-sectional survey of a ª snowballº sample of

long-term cannabis users. Setting. The North Coast of New South W ales is an area with high levels of

cannabis cultivation and use, and many long-term users. Participants. The study involved 268 long-te rm

cannabis users who had regularly used cannabis for at least 10 years. M easurements. A structured

interview schedule obtained information on: demographics, social circumstances, patterns of cannabis and

other drug use, contexts of use, perceptions about cannabis and legal involvement. Findings. The mean age

of the sample was 36 years and 59% were male. The median length of regular cannabis use was 19 years.

Most (94%) used two or more times a week and 60% used daily, with a median of two joints per day.

Two-thirds (67%) used cannabis in social setting s and two-thirds grew cannabis for their own use. The most

common reasons for using cannabis were for relaxation or relief of tension (61%) and enjoyment or to feel

good (27%). The most commonly reported negative effects were feelings of anxiety, paranoia, or depression

(21%), tiredness, lack of motivation and low energy (21%) and effects of smoke on the respiratory system

(18%). The majority drank alcohol (79%) and over one-third were drinking at hazardous levels. M ost were

current (64%) or ex-tobacco smokers (24%). One-quarter (25%) had been charged with possession of

cannabis, 11% for cultivation and 6% for supply, with non-drug offences low (8% or less). Overall,

three-quarters (72%) believed that the bene ® ts of cannabis use outweighed the risks, 21% felt there was an

even balance, and 7% said cannabis had done them more harm than good. Conclusions. Among long-te rm

cannabis users in this Australian rural area, cannabis use was an integral part of everyday life and it was

primarily used in social situations for the same reasons that alcohol use is used in the wider community .

Introduction

Cannabis is the most widely used illicit drug in

the United States (Adams & Martin, 1996) and

Australia, with one in three Australian adults

reporting that they had used cannabis at some

time in their lives (National Drug Strategy,

1996). Approximately 10% of people who ever

use cannabis continue to use it regularly for

some years, but the majority of these discontinue

their use in their mid to late twenties (Kandel et

Correspondence to: David Reilly, Northern Rivers Health Service, Locked Mail Bag 11, Grafton, New SouthWales 2460 , Australia. Tel: 1 61 2 6620 2124 ; Fax: 1 61 2 6621 7088.

Submitted 30th April 1997; initial review completed 6th August 1997 ; ® nal version accepted 5th November 1997 .

0965 ± 2140/98/060837 ± 10 $9.50 Ó Society for the Study of Addiction to Alcohol and Other Drugs

Carfax Publishing Limited

838 David Reilly et al.

al., 1986). It is the minority who use regularly

over years who put themselves at greatest risk of

experiencing the long-term health effects of can-

nabis (Hall, Solowij & Lemon, 1994; Hall,

1995).

There have been very few studies of long-term

cannabis users in naturalistic settings in Western

societies. This is surprising considering that can-

nabis use increased rapidly from the late 1960s

to become the most widely used illicit rec-

reational drug in the United States and, more

recently, in Australia (Donnelly & Hall, 1994).

Although there have been studies of chronic

users in Jamaica (Rubin & Comitas, 1975),

Greece (Stefanis, Dornbush & Fink, 1977) and

Costa Rica (Carter, Coggins & Doughty, 1980;

Page, Fletcher & True, 1988) only a limited

number of small sample studies of long-term

users were conducted in the United States over

10 years ago (Haas & Hendin, 1987; Hendin et

al., 1987; Rainone et al., 1987; Roffman & Barn-

hart, 1987). Consequently, there is limited infor-

mation on people who have used cannabis

regularly for 10± 20 years or more. We know very

little about their characteristics, life-style, drug

use patterns and the impact of cannabis on their

lives, families and the communities in which they

live.

The present study describes a sample of Aus-

tralians who had used cannabis regularly for 10

or more years, and provides data on their pat-

terns of cannabis use, the contexts within which

it was used, their social circumstances and in-

volvement with the legal system. Further details

on methods and results including health and

psychological characteristics of users, cannabis

dependence, and their partners’ perception

about cannabis use has been reported (Didcott et

al., 1997).

The NSW North Coast was chosen as the site

for study because it is widely reputed to be a

region where there is a concentration of long-

term cannabis users. The North Coast’ s repu-

tation as a centre for cannabis cultivation and

use is supported by historical evidence (Helli-

well, Reilly & Rippingale, 1992; Major, 1994)

and crime statistics which indicate a high rate of

cannabis seizures and charges for cannabis of-

fences in the region (NSW Bureau of Crime

Statistics and Research, 1996).

The present paper reports on two major objec-

tives of the study which were: to describe the

demographic, social and legal characteristics of

long-term cannabis users in a rural area with a

high prevalence of cannabis use; and to describe

their patterns of cannabis and other drug use,

their perceptions of cannabis use, and the con-

texts within which they used cannabis. The is-

sues of dependence have been reported in a

separate paper (Swift et al., 1998).

M ethod

The setting for the study was the North Coast

region of NSW. It is one of the fastest-growing

areas in Australia, with agriculture and tourism

the main industries. There is also a high level of

unemployment and socio-economic disadvan-

tage. There is a substantial traveller and transient

population and some areas have a high pro-

portion of people who pursue ª alternative life-

stylesº (North Coast Region, 1990; Department

of Urban Affairs and Planning, 1995).

The research was conducted in three main

phases. The ® rst consisted of an exploratory

phase of interviews with key informants and a

series of focus groups with long-term cannabis

users, which led to the development and pilot

testing of a draft questionnaire. The second

phase consisted of in-depth interviews with 268

long-term intensive cannabis users using a struc-

tured questionnaire. A third phase that involved

unstructured interviews with a sample of family

members and close friends of cannabis users will

be reported elsewhere.

Sampling

Since it was not possible to obtain a simple

random sample of cannabis users, subjects were

recruited by the chain-referral method known as

ª snowballº sampling. Snowball sampling in-

volves the researcher interviewing an initial set of

contacts and asking them to introduce the inter-

viewer to other users whom they know or with

whom they are acquainted. These contacts in

turn introduce others to the interviewer, and so

on. This technique has been used successfully to

access dif® cult-to-reach or hidden populations

(see, e.g. Biernacki & Waldorf 1981; Lambert,

1990; Lee, 1993).

In the present study, ® ve interviewers were

recruited who had knowledge of local networks

of cannabis users. They had several starting

points which in turn led to further contacts, and

so on. As Lee (1993) has pointed out, a snowball

Characteristics of long-te rm cannabis users 839

sample does not necessarily produce a long train

of contacts originating from the initial contacts.

It more often produces a slow accumulation of

interviews from a number of different starting

points rather than a geometrically increasing

number of respondents originating from a single

source. This was the case in the present research.

We interviewed many small chains of respon-

dents from the initial contacts that the

® eldworkers had within the social networks of

cannabis users in the rural communities involved

in the study. This strategy recruited the complete

sample, and avoided the problems of contacting

users through other approaches such as popu-

lation surveys, treatment services and media ad-

vertising.

The major criterion for entry into the study

was that the person had used cannabis

ª regularlyº for at least 10 years. ª Regularº use

was de® ned as using at least three times a week.

Use may have been less frequent on occasion

(e.g. if cannabis was in short supply) but users

ª usuallyº used cannabis with this frequency. Fe-

male users whose use was irregular during preg-

nancy or breast-feeding were included provided

they had at least 10 years of use. When recruiting

respondents, interviewers were deliberately

vague about the entry criteria so that volunteers

would not exaggerate their use in order to qualify

for the study.

Subjects who formerly met these criteria also

quali® ed for interview if their current use was

less than three times a week or if they were

currently abstinent from cannabis, provided that

they had ceased their use not more than 12

months before the date of interview. There were

25 of these former heavy users in the sample to

provide data on reasons for stopping and

dif® culties experienced in doing so. These sub-

jects have been excluded from statistical analyses

of current patterns of use.

Interview schedule

Data were collected by a personal interview

schedule that was derived from the research

literature and from pilot ® eldwork. The schedule

required self-reported responses about the re-

spondents’ behaviours and experiences. It cov-

ered: demographic characteristics; past and

current patterns of cannabis use, including rea-

sons for use, experiences of use and self-control

strategies; and other drug use. Some items were

categorized (e.g. duration, frequency, type of

cannabis) and other questions were open-ended

(e.g. Why do you currently smoke cannabis?

What do you like most about using cannabis?º

What do you like least about using cannabis?).

Problem drinking was assessed by the Alcohol

Use Disorders Identi® cation Test (AUDIT)

(Saunders et al., 1993), a 10-item scale that

assesses typical quantity and frequency of drink-

ing, alcohol-related problems and symptoms of

dependence. A score of 8 or more on this scale

has been found to predict an increased risk of

experiencing alcohol-related health and social

problems over a 2- year follow-up period (Coni-

grave et al., 1995).

Procedure

Interviews were carried out over an 11-month

period between August 1994 and June 1995 by

® ve interviewers. Interviewer training and quality

control of interview content continued through-

out the ® eldwork to ensure that the data were of

high quality and that the information collected

by different interviewers was consistent. Respon-

dents were interviewed in their homes, local

parks, cafes or in community facilities, such as a

private room in a health clinic.

All participants were briefed fully on the na-

ture and purpose of the study before taking part

and they were told that they could withdraw at

any time if they wished. All respondents signed a

consent form before the interview (although not

necessarily with their own name). The

con® dentiality and anonymity of information

supplied was stressed throughout the interview.

Subjects were reimbursed A$30 for expenses

incurred in attending the interview.

The sample was recruited from two areas

within the region situated about 250 km apart.

These areas are among the main cannabis grow-

ing areas on the North Coast, according to infor-

mation from the NSW Police Service Drug

Enforcement Agency’ s cannabis eradication pro-

gramme. Each area is a well-de® ned and socially

cohesive local government area. A total of 268

users were recruited for the user phase of the

study, 162 from one area and 106 from the

other.

Data analysis

The data analysis was of two kinds: descriptive


analyses of personal characteristics and their pat-

terns of cannabis and other drug use; and multi-

variate statistical analyses of relationships

between demographic characteristics and pat-

terns of cannabis and other drug use.

In the descriptive analyses, variables were tab-

ulated by gender but combined results are pre-

sented if there were no gender differences.

Continuous variables were summarized by

means when they were approximately normally

distributed and by medians when their distribu-

tions were skewed. Where possible, distributions

were compared with population data bases (e.g.

Census and National Health Survey data broken

down by Australian and North Coast samples)

or test norms by t-tests or c 2 tests.

Results

Demographic characteristics

The sample comprised 268 adult residents (59%

men and 41% women) of the NSW North Coast

with a mean age of 36.4 years (SD 5 7.5 years).

Men were older than women (38.9 years vs. 34.2

years, t 5 3.95, df 5 241, p , 0.001). More than

two-thirds of the sample (69%) were Australian-

born and 4% identi® ed themselves as Aboriginal

or Torres Strait Islanders.

The sample was well educated, with 62% hav-

ing obtained further educational quali® cations

since leaving school. More than half the sample

were employed full-time or part-time (58%),

while just under half (43%) reported that they

were unemployed or receiving government

bene® ts, including sickness or single parent al-

lowances. One in ® ve performed home duties

(21%), and 2% were retired. The main sources

of income were employment (48%) and govern-

ment bene® ts (42%). One in ® ve respondents

(19%) earned money from selling cannabis, and

6% earned at least half their income this way.

Two-thirds of the men and women in the

sample (64%) were currently in a relationship,

and just over half (57%) had children under 16

years of age in their part-time or full-time care.

The majority of respondents lived with others,

with 24% living alone (28% of men and 17% of

women). One -third (34%) were living with their

partner and children, 9% were living only with

their partner, and 19% were single parents. The

majority of the sample lived in a rural setting,

with 29% living in urban or suburban areas.

By comparison with Census (1991) data for

the Australian population and the NSW North

Coast, the cannabis users in this study had simi-

lar employment rates, were older and better edu-

cated, more likely to be single parents, less likely

to be married, and more likely to be living in

group households or in couples with children

than the Australian and North Coast popula-

tions.

Patterns of cannabis use

The snowball recruitment produced a sample of

adults with a long history of regular cannabis use

(see Table 1). The respondents ® rst initiated

cannabis use around the age of 17, later for

males (17.3 years) than for females (16.2 years)

(Mann± Whitney U-test, p , 0.01), and became

regular cannabis users 2 years later. The average

length of their regular cannabis use was 19 years,

with males having a longer history of regular use

(median of 20.2 years) than females (median of

17.4 years) (Mann± Whitney U-test, p , 0.001).

Most had ready access to cannabis, with two-

thirds growing cannabis for their own use, al-

most half growing most of the cannabis they

required.

Smoking was the usual method of cannabis

administration. Just under half (48%) usually

smoked a joint of cannabis and tobacco, 22%

smoked a cannabis joint without tobacco, 16%

used a waterpipe or ª bongº , 2% a chillum, and

12% used combinations of these methods. Only

one person usually ate cannabis.

The typical frequency of cannabis use was

every day in 60% of cases, four or more times a

week for 86%, and two or more times a week for

94%. This had been the typical pattern of use for

more than a year in 82%, and for more than 5

years in 62% of cases. The typical quantity con-

sumed per day ranged between 0.2 and 40 joints,

with a median of two joints per day. Half used

between one and four joints per day. A third

used cannabis throughout the day, while the

remainder restricted themselves to evenings or

other times.

Most of the sample (80%) had been daily

cannabis users at some time, with a median

duration of 9.9 years of daily use which repre-

sented over half (59%) of their cannabis using

history. Just under half (47%) reported that they

had occasional cannabis ª bingesº , usually at

ª harvest timeº or when with friends who were

using.


Table 1. Patterns of cannabis use

Males Females TotalN 157 111 268

Frequency of use (%)Daily 61.1 58.6 60.14± 6 days/week 26.1 25.2 25.71± 3 days/week 10.2 11.7 10.8, weekly 2.5 4.5 3.4

Median age 1st use 17.00 16.00 17(years)

Median age regular use 19.00 19.00 19Median years regular use 20.00 17.00 19Typical pattern (%)

, 6 months 12.8 12.6 12.76± 12 months 1.9 9.0 4.91± 5 years 19.1 22.5 20.5. 5 years 66.2 55.9 61.9

Usual method (%)Joint cannabis only 21.0 22.5 21.6Joint with tobacco 44.6 52.3 47.8Bong 19.7 11.7 16.4Other 4.4 1.8 3.3Combinations 10.2 11.7 10.8

Kind of cannabis (%)Heads 62.2 50.5 57.3Heads/leaf 19.9 22.5 21.0Other 17.9 27.0 21.6

The majority of participants used the more

potent cannabis preparations. Almost 60%

(57%) reported that they typically used the

ª headsº of the cannabis plant, and 21% smoked

a mixture of heads and leaf. Only 1% reported

using cannabis resin or ª hashº . The remainder

used cannabis leaf (4%), combinations of the

above (13%) or other forms (4%) of cannabis,

such as tips and leaf. Almost all said they pre-

ferred heads and only used less potent forms

when heads were unavailable, or they could not

afford to purchase them. Heads are the ¯ owering

tops of the female cannabis plant and have a high

content of delta-9-tetrahydrocannabinol or

THC. In Australia, typical cannabis contains

5± 6% THC and recent police seizures of hybrids

such as skunk have yielded THC levels of 12±

13% (Australian Bureau of Criminal Intelli-

gence, 1996).

Contexts of cannabis use

Cannabis was most often used in social settings

with two-thirds (67%) usually or always sharing

cannabis, and only 12% usually smoking can-

nabis alone. Over 90% had friends who used

cannabis most days, and three-quarters said that

most of their friends were cannabis users.

Three-quarters (73%) were members of a fam-

ily group in which some other members used

cannabis. Of those who were currently in a rela-

tionship, three-quarters (73%) of their partners

used cannabis, usually in shared sessions, on an

average of seven times a week. Just over half the

sample (54%) lived in a household containing

children under 16 years of age. Of these, half

(52%) said that they either used less cannabis

when their children were around, or they restric-

ted their use to times or places away from their

children.

The majority performed their normal daily

activities either during or after smoking can-

nabis: 20% did so often, and 41% always or

usually did so. Smoking before or during work

was also common, with 41% of those employed

in the past 2 years reporting that they did so.

One quarter (23%) smoked just before work,

22% smoked while working and 24% smoked

during work breaks. Just under half (48%)

smoked after ® nishing work.

Cannabis use often went together with driving

a motor vehicle. Ninety per cent of the sample


said they drove a vehicle at least occasionally

soon after using cannabis, and 70% reported

driving sometimes while using cannabis. A quar-

ter of respondents (23%) reported that they felt

con® dent when doing so, 18% said they felt no

different from usual, 25% said that they felt

slower or more relaxed, and 27% felt they were

more careful. More than half the sample (60%)

had operated machinery such as chainsaws and

power tools, or used guns, after using cannabis.

Users’ perceptions of cannabis use

The most popular reasons for using cannabis

were for relaxation or relief of tension (61%) and

for enjoyment or to feel good (27%). By far the

most commonly reported effect of cannabis was

relaxation or relief of tension. Almost half said it

was calming, slowed them down, or helped them

sleep. Other commonly reported effects were: a

feeling of pleasure or enjoyment, wellbeing, ela-

tion or euphoria; mood-enhancing or altering

(e.g. a sense of not taking life too seriously.); an

alteration of consciousness, or of the way the

world was perceived; an enhancement of every-

day experiences (e.g. of music or in nature);

stimulation of thinking processes and enhanced

creativity; and encouraging introspection.

Almost all respondents reported some negative

aspect of cannabis use, such as the illegality of

cannabis use (29%), the high costs of cannabis

(14%) and the social stigma of being a cannabis

user (11%). Substantial minorities reported a

variety of negative effects more directly attribu-

table to using cannabis. These included: negative

emotions, such as paranoia or depression (21%);

feeling tired, unmotivated or being low in energy

(21%); and being forgetful and less able to con-

centrate (11%). Nearly one in ® ve (18%) were

concerned about the effect of cannabis on their

lungs or respiratory system.

When asked to provide a global rating on the

overall effects of cannabis, almost three-quarters

(72%) believed that the bene® ts outweighed the

risks, while the remainder felt there was an even

balance (21%) or that cannabis had done them

more harm than good (7%).

Most participants employed self-control

strategies to limit their cannabis use. Almost half

(41%) did not use at work or before work,

one-third (30%) would not use around strangers,

parents or people who they knew would disap-

prove, and 12% said they did not use around

children. A minority (16%) said they had no

limits, rules or had been unsuccessful limiting

use in the past.

Other drug use

The most widely used drugs after cannabis were

alcohol and tobacco. Almost all had used alcohol

at some time in their lives (98%) and most

(79%) had used alcohol within the past month.

Using the cut-off of 8 on the AUDIT recom-

mended by Babor et al. (1989), 30% of the

respondents (28% of the men and 18% of the

women) were identi® ed as drinking ª at riskº .

Using a cut-off (Conigrave, Hall & Saunders,

1995) that takes account of gender differences in

consumption, 37% of the sample were so

classi® ed, 41% of men (who scored 7 or more)

and 30% of women (who scored 6 or more).

Forty-one per cent reported that they used al-

cohol and cannabis together on most days (14%)

or at least weekly (27%).

The majority of the sample (88%) were either

current (66%) or former (22%) tobacco smok-

ers. The average age at which former smokers

had quit smoking was 30 years. Among current

smokers, the median number of cigarettes

smoked per day was 15. Almost half the sample

(42% of men and half the women) reported that

they smoked more tobacco when cannabis was

not available.

Most of the sample had used other illicit drugs

at some time in their lives. Almost all (92%) had

used hallucinogens and more than half had used

amphetamines (75%), cocaine (71%), benzodi-

azepines (54%), opiates other than heroin (59%)

and designer drugs, such as ecstasy (56%). Just

under half had used heroin (42%) at least once

in their lives but only 23% had ever used in-

halants, such as amyl nitrate or nitrous oxide. At

some time in their lives the respondents had used

an average of 5.7 of the eight illicit drug classes

about which they were asked.

The use of these illicit drugs had occurred

primarily in the past. Only a quarter (24%) had

used any illicit drug other than cannabis in the

past month. None of these drug classes been

used by more than 10% of the sample in the past

month and, in most cases, fewer than 5% of the

sample had used any of these drugs in the past

month. On average, they had only used one of

the eight drug types in the past month.


Correlates of cannabis use patterns

There were small correlations between respon-

dent characteristics, other drug use and cannabis

use patterns. The age at which cannabis was ® rst

used was negatively correlated with the number

of illicit drug types that the person had used in

their lives (r 5 2 0.34, p , 0.05). That is, the

younger participants had ® rst used cannabis, the

more likely they were to have used other illicit

drugs over their lifetime.

The typical frequency of cannabis use was

only modestly correlated (r 5 0.19, p , 0.05)

with the quantity of cannabis that was typically

consumed (which was modelled as the logarithm

of the number of ª standard jointsº typically used

because the distribution was positively skewed).

This indicated that those who were daily can-

nabis users typically used more cannabis than

those who used less frequently.

A multiple regression analysis indicated that

typical quantity of cannabis used was modestly

predicted (R2 5 0.06, p , 0.01) by typical fre-

quency of use (B 5 0.17, p , 0.01) and a combi-

nation of past (B 5 0.12, p , 0.06) and current

use of other illicit drugs (B 5 2 0.11, p , 0.09).

This suggests that those who used other illicit

drugs in the past month typically used more

cannabis than those who had not used other

illicit drugs in the past month, and the more

cannabis that was typically used, the fewer other

illicit drugs were used in the past month.

Involvement with the legal system

One-quarter of the sample (26%) had ever been

charged with possession, 11% for cultivation and

6% for supply. One in 10 (12%) had been

arrested for other drug-related offences. Most of

these offences had occurred some years previ-

ously (mean 10.4 years), with only 22% of can-

nabis cases and 25% of other drug offences in

the past 3 years. Non-drug offences were low

with the most common drink-driving (8%),

stealing (6%), traf® c offences (6%), robbery and

assault (3%) and miscellaneous offences (8%).

Almost all (98%) said that the criminal penal-

ties in New South Wales for the personal use of

cannabis should be repealed. The majority sup-

ported factual education in schools (92%) and

availability of counselling (91%) and stop smok-

ing programmes (89%), although many were

sceptical that the counselling and education pro-

grammes would provide unbiased information.

Over half (56%) disagreed with the view that

there should be cannabis testing of drivers, such

as random breath testing for alcohol, compared

with 12% of the Australian general public (Bow-

man & Sanson-Fisher, 1994).

Discussion

The study identi® ed a sample of people in an

Australian rural area who had regularly used

cannabis for many years. Daily cannabis smoking

was common, and cannabis use was regarded an

integral and valued part of everyday life and

social relationships. Within these social net-

works, cultivation of cannabis was common and

an important way of ensuring a regular supply.

Study limitations

There are three limitations that affect the conclu-

sions that can be drawn from the study. First, it

is dif® cult to say whether the ª snowballº sam-

pling method succeeded in obtaining a represen-

tative sample of long-term cannabis. Ultimately,

this question can only be answered by the degree

to which our ® ndings are replicated in other

samples of long-term cannabis users.

Secondly, in any study of drug users that relies

upon self-reported data the question must be

asked: to what extent may respondents have

under-reported their cannabis use? The usual

guarantees of anonymity and con® dentiality were

given to minimize the risks of respondents not

feeling able to be as forthcoming as they might

about illegal and stigmatized activities. Special

efforts were also made to establish the credibility

of the project and to use as peer interviewers

persons who were well-known and respected in

the area. We believe that these measures maxi-

mized our respondents’ preparedness to be hon-

est and that was the view of the interviewers in

the ratings made of respondent honesty at the

end of each interview.

Finally, a cross-sectional survey of a snowball

sample of long-term cannabis users is limited in

the type of inferences that can be drawn from the

study ® ndings. All these limitations must be kept

in mind in interpreting our data.

The typical pro® le of individuals in this study

was of men and women in their middle 30s who

were better educated than their peers, with em-

ployment rates and social security bene® ts simi-

lar to the general population, although if


employed they were more likely to be self-

employed and working part-time. While all had

used cannabis for at least 10 years to be included

in the study, the sample had used for a median

of 19 years at least weekly use, and 80% had a

history of daily cannabis use at some time in

their lives.

Since most cannabis users in the broader

population discontinue their use in their late 20s

(Bachman et al., 1997), our sample over-repre-

sents that small minority of cannabis users who

use the drug regularly for several decades. Our

data are also consistent with earlier epidemiolog-

ical data (Kandel & Logan, 1984) in showing

that regular cannabis users begin cannabis use

earlier, are more likely to be cigarette smokers

and to have used a variety of other illicit drugs

than the general population. Thus, compared to

the general population sample interviewed in the

1995 Australian National Drug Household Sur-

vey, the North Coast sample had initiated can-

nabis use at an earlier age (mean of 18 vs. 20

years), were more likely to be current regular

smokers (66% vs. 23%) and were more likely to

have tried other illicit drugs (92% vs. less than

2% of respondents in the Household Survey who

had used illicit drugs other than marijuana) (Na-

tional Drug Strategy, 1995; Makkai & McAllis-

ter, 1997).

The reasons that regular users give for using

cannabis are similar to those offered by alcohol

users, namely that it relaxes, relieves tension and

stress and helps them to feel good and enjoy

themselves, similar to the ® ndings of other stud-

ies of long-term users (Hendin et al., 1987).

These reports are consistent with the fact that

most cannabis use occurs in a social context with

family and friends who are likely to be cannabis

users.

The majority of cannabis users in this study

were aware that cannabis can cause some health

and psychological problems. Previous studies

found similar concerns about reduced energy,

impaired concentration and memory, family

problems and respiratory symptoms (Rainone et

al., 1987; Roffman & Barnhart, 1987).

Driving a motor vehicle while using cannabis

or just after use was very common: almost 90%

of users had driven at some time after using it.

There is some evidence that psychomotor im-

pairment resulting from driving while ª stonedº

may be offset by driving at a slower speed and

taking fewer risks because the user is more aware

of his or her impairment. Driving is none the less

a potentially high-risk activity and emergencies

can arise which no amount of precautionary

behaviour can overcome (Adams & Martin,

1996; Solowij, Hall & Lemon, 1994). Moreover,

as it is the lives of other road users that cannabis

users put at risk, we should be educating them

not to drive when experiencing the acute effects

of cannabis.

The combination of cannabis and alcohol

poses a high risk for accidents when driving or

operating machinery. One in four in this study

had driven a vehicle within the previous month

after using cannabis and alcohol together. As

research indicates that driving after using can-

nabis and alcohol in combination is riskier than

driving after using either alone (Chesher, 1986),

cannabis users also need to be persuaded of the

risks of driving shortly after using cannabis and

alcohol together.

Although life-time use of a variety of illicit

drugs was high, current use of illicit drugs other

than cannabis was low. Alcohol and tobacco

were commonly used. Alcohol was regularly used

by just under half, often in combination with

cannabis, and it was used at hazardous levels by

a substantial minority of both men and women.

Tobacco smokers were over-represented in the

sample, as one would expect from data on pat-

terns of drug use.

The majority of participants who had involve-

ment with the criminal justice system had done

so because of their cannabis-related activities.

This supports the ® ndings of a Western Australia

study on cannabis offenders, which found that

the majority had low levels of criminal involve-

ment apart from their cannabis use (Lenton,

Ferrante & Loh, 1996).

Study implications

A survey such as ours inevitably raises many

questions that are deserving of further research.

The ® rst question is how representative is this

sample of long-term cannabis users? They are

arguably a special population who live in an area

well known for its high prevalence of cannabis

use and cultivation, and they live within social

networks that are very tolerant of cannabis use.

It remains to be seen to what extent their experi-

ences can be generalized to other long-term can-

nabis users, for example those who live in urban

areas.


When considered in conjunction with other

research the study ® ndings have a number of

implications for health educational information

directed at cannabis users and the broader com-

munity.

First, the misconception that regular cannabis

users rarely use other drugs needs to be dis-

pelled. The majority of regular cannabis users in

this and other studies (e.g. Kandel et al., 1986;

Rainone, et al., 1987; Wiesbeck et al., 1996) are

regular tobacco smokers, and a substantial min-

ority drink alcohol in hazardous and harmful

amounts. Secondly, the major health risks of

cannabis probably amplify the health risks of

tobacco and alcohol which so many regular can-

nabis users also use. Considering the methodo-

logical limitations, this study provides the ® rst

Australian data on the characteristics of long-

term cannabis users, and their patterns of can-

nabis and other drug use.

Acknowledgements

The study was funded by a grant from the Na-

tional Drug Strategy Research into Drug Abuse

Grants, Commonwealth Department of Health

and Family Services. Additional support was

provided by the Institute of Health and Re-

search, Northern Rivers Health Service and

Southern Cross University. We gratefully ac-

knowledge Sandra Heilpern and Graham Irvine

for their outstanding ® eldwork and advice and

also to Lynda Brenton and Steve Reeves.

ReferencesADAM S, I. B. & M ARTIN , B. R. (1996) Cannabis: phar-

macology and toxicology in animals and humans,

Addiction, 91, 1585 ± 1614 .AUSTRALIAN BUREAU OF CRIMINAL INTELLIGENCE

(1996) Australian Illicit Drug Report 1995 ± 96 (Can-berra, Australian Bureau of Criminal Intelligence).

BABOR, T. F., DE LA FUENTE , J. R., SAUNDERS, J. &GRANT , M. (1989) AUDITÐ the Alcohol Use Disorders

Identi® cation Test: guidelines for use in primary health

care (Geneva, World Health Organization, Divisionof Mental Health).

BACHM AN , J. G., WADSW ORTH , K. N., O’ MALLEY , P.M., JOHNSTO N, L. D. & SCHULENG ERG, J. E. (1997) .

Smoking, Drinking, and Drug Use in Young Adulthood(Mahwah, New Jersey, Lawrence Erlbaum Associ-ates).

BIERNACKI, P. & WALDORF , D. (1981 ) Snowball sam-pling: problems and techniques of chain-referralsampling, Sociological Methods and Research, 10, 141±163.

BOW MAN , J. & SANSON-FISHER, R. (1994) Public Percep-

tions of Cannabis Legislation, National Drug StrategyMonograph Series No. 28 (Canberra, AustralianGovernment Publishing Service).

CARTER , W. E., COGGINS, W. & DOUGHTY, P. L.(1980) Cannabis in Costa Rica: a study of chronic

marijuana use (Philadelphia, Institute for the Studyof Human Issues Press).

CH ESHER , G. (1986) The effects of alcohol and mari-juana in combination: a review, Alcohol, Drugs and

Driving, 2, 105± 119.CO NIGRAVE , K. M., HALL, W. D. & SAUNDERS , J. B.

(1995 ) The AUDIT questionnaire: choosing a cut-off score, Addiction, 90, 1349 ± 1356.

DEPARTMENT OF URBAN AFFAIRS AND PLANNIN G

(1995) North Coast Population of Development Monitor

Issue No. 17 (Sydney, Department of Urban Affairsand Planning).

D IDCOTT , P., REILLY, D., SWIFT, W. & HALL, W.(1997) Long-Term Cannabis Users on the New South

Wales North Coast, NDARC Monograph No. 30(Sydney, National Drug & Alcohol ResearchCentre).

DO NNELLY , N. & HALL, W. (1994) Patterns of Cannabis

Use in Australia, National Drug Strategy MonographSeries No.27 (Canberra, Australian GovernmentPublishing Service)

HAAS, A. P. & HENDIN , H. (1987 ) The meaning ofchronic marijuana use among adults: a psychosocialperspective, Journal of Drug Issues, 17, 333± 348.

HALL, W., SOLOW IJ, N. & LEMON, J. (1994) The Health

and Psychological Consequences of Cannabis Use, Na-tional Drug Strategy Monograph Series No.25 (Can-berra, Australian Government Publishing Service)

HALL, W. (1995) The public health implications ofcannabis use, Australian Journal of Public Health, 19,235± 242.

HELLIWELL, D., REILLY, D. & RIPPINGALE, C. (1992)Establishing a drug and alcohol service in an Aus-tralian rural area, Drug and Alcohol Review, 11, 371±378.

HENDIN , H., HAAS, A. P., SINGER, P., ELLNER , M. &ULMAN , R. (1987) Living High: daily marijuana use

among adults (New York, Human Sciences Press).KANDEL , D. B. & LO GAN, J. A. (1984) Patterns of drug

use from adolescence to young adulthood: periods ofrisk for initiation, continued use and discontinu-ation, American Journal of Public Health, 74, 660±666.

KANDEL , D. B., DAVIES, M., KARU S, D. & YAMAGUCHI,K. (1986) The consequences in young adulthood ofadolescent drug involvement, Archives of General

Psychiatry, 43, 746± 754.LAMBERT, E. Y. (Ed.) (1990 ) The Collection and In-

terpretation of Data from Hidden Populations, NIDAMonograph 98 (Rockville, MD, National Institute ofDrug Abuse).

LEE, R. S. (1993 ) Doing Research on Sensitive Topics

(London, Sage).LENTON , S., FERRANT E, A. & LOH , N. (1996) Dope

busts in the West: minor cannabis offences in theWestern Australia criminal justice system, Drug and

Alcohol Review, 15, 335± 341.M AJOR , C. (1994). Deadlocked-cannabis reform, Con-

nexions, 14, 8± 16.


MAKKAI, T. & MCALLISTER, I. (1997) Marijuana in

Australia: patterns and attitudes, National Drug Strat-egy Monograph Series No. 31 Canberra, AustralianGovernment Publishing Service).

NATIONA L DRUG STRATEG Y (1996) National Drug

Strategy Household Survey: survey report, 1995(Canberra, Australian Government Publishing Ser-vice).

NEW SOUTH W ALES BUREAU OF CRIME STATISTICS AND

RESEARCH (1996) New South Wales Recorded Crime

Statistics 1995 (Sydney, NSW Bureau of CrimeStatistics and Research).

NORTH COAST REGION (1990) The Regional Strategic

Plan 1990 ± 1995 (Lismore, NSW Department ofHealth, North Coast Region).

PAGE, J. B., FLETCH ER, J. & TRUE, W. R. (1988 )Psychosociocultural perspectives on chronic can-nabis use: the Costa Rican follow-up, Journal of

Psychoactive Drugs, 20, 57± 65.RAINONE , G. A., DERREN , S., HENM AN P. H. & W ISH,

E. D. (1987 ) Heavy marijuana users not in treat-ment: the continuing search for the ª pureº mari-juana user, Journal of Psychoactive Drugs, 19,353± 359.

ROFFMAN , R. A. & BARNHAR T, R. (1987) Assessing theneed for marijuana dependence treatment throughan anonymous telephone interview, International

Journal of Addictions, 1987, 639 ± 651.RUBIN, V. & COM ITAS, L. (1975) Ganja in Jamiaca: a

medical anthropological study of chronic marijuana use(The Hague, Mouton Publishers).

SAUNDERS , J. B., AASLAND , O. G., AMUNDSEN , A. &GRANT , M. (1993) . Alcohol consumption and re-lated problems among primary health care patients:WHO collaborative project on early detection ofpersons with harmful alcohol consumptionÐ I,Addiction, 88, 349± 362.

STEFANIS, C., DORNBU SH, R. & FINK , M. (1977) Hash-

ish: studies of long-term Use (New York, Raven Press).SWIFT, W. HALL, W. D IDCOTT , P. & REILLY, D. (1998)

Patterns and correlates of cannabis dependenceamong long-term users in an Australian rural area,Addiction, 93, in press.

W EISBECK, G. A., SCHUCKIT, M. A., KALM IJN, J. A.,TIPP, J. E., BUCHOLZ, K. K. & SMITH, T. L. (1996)An evaluation of the history of marijuana withdrawalsyndrome in a large population, Addiction, 91, 1469±1478.

Documents

Long-term cannabis use: characteristics of users in an Australian rural area