Journal of the New Zealand Medical Association, 23-May-2008, Vol 121 No 1274
Trends in population drug use in New Zealand: findings from national household surveying of drug use in 1998, 2001, 2003, and 2006
Chris Wilkins, Paul Sweetsur
Drug use imposes a range of health and social costs on New Zealand including death, illness, mental health problems, injuries from accidents, domestic violence, family and relationship breakdown, and child neglect.1 Monitoring population trends in drug use is important in developing responses to emerging drug problems, and also for evaluating the effectiveness of existing responses.1
There have been several changes in drug use and drug policy in New Zealand over the past 10 years or so. There has been a liberalisation of the alcohol environment since 1989, which culminated in the lowering of the alcohol purchase age to 18 years old in 1999.2,3
The Government has implemented a range of initiatives to reduce the use and social impacts of tobacco use—including increased taxation of tobacco products; mass media campaigns encouraging users to seek help to stop use; and the imposition of smoke free areas in workplaces, restaurants, and bars.4
The Government’s response to the emergence of methamphetamine has included the reclassification of methamphetamine to the highest Class A category of the Misuse of Drugs Act 1975, additional resources and powers given to Police and Customs involved in drug enforcement, the negotiation of a Memorandum of Understanding with the New Zealand Chemical Industry Council to monitor the sale of chemical precursors that can be used to clandestinely manufacture methamphetamine, and the development at a local level of protocols with pharmacies to notify police of suspicious purchases of medicines containing pseudoephedrine.5
Trends in the use and regulation of high-profile substances should be viewed in the wider context of trends for other drug types in New Zealand. While national data on drug use in New Zealand has been collected fairly regularly over the past 8 years using the same survey methodology,6–8 extensions to the original survey design have made comparisons over the entire period of national surveying somewhat difficult. The age ranges of the samples surveyed have been extended in recent waves and some new drug types have been included.
The national surveys of drug use have regularly collected data concerning changes in the level of use of each drug type, but the nature of the data collected has made it difficult to assess overall changes in the level of use of substances between survey waves.
The aim of this paper is therefore to compare the national population prevalence, and change in levels of use, of 13 drug types—including alcohol, tobacco, cannabis, amphetamines, and ecstasy (MDMA)—among the general New Zealand population aged 15–45 years old over the past 8 years.
National household surveying of drug use was conducted in New Zealand in 1998, 2001, 2003, and 2006. The 1998 National Drug Survey (NDS) interviewed a national sample of 15–45 year olds using a Computer Assisted Telephone Interview (CATI) survey methodology.6 The 2001 NDS retained the same questionnaire and CATI methodology as the 1998 NDS and extended the age range of the sample to 13–45 year olds.7 The 2003 Health Behaviours–Drug Use (2003 HBS-Drug Use) retained the core sections of the questionnaire and same CATI survey methodology as the two previous NDS.8 The age range of the 2003 HBS-Drug Use sample was extended further to include 13–65 year olds. The 2006 data was collected as part of the 2006 national survey of benzylpiperazine (BZP) party pill use.
The party pill survey used the same CATI methodology as the previous NDS and the 2003 HBS-Drug Use.9 The drug prevalence section used in the NDS and the 2003 HBS-Drug Use was included in the BZP party pill questionnaire and was asked of all those aged 13–45 years who were contacted for the survey whether or not they used BZP party pills.
All four waves of random digit dialling (RDD) sampling employed the same CATI sampling methodology. Telephone numbers were selected using a stratified random digit dialling method so that each household, of a particular stratum nationwide, had an equal chance of being called. The country was divided into several strata based on telephone exchanges to represent the different socioeconomic characteristics of the population. A proportionate sample from each stratum was then taken. Within each household, one person was randomly selected for an interview.
Each telephone number was tried at least 10 times on different dates and times of the day in an effort to reach those seldom at home. Respondents were informed that the study was being conducted on behalf of the Ministry of Health and that everything they said would be confidential.
In each survey wave, participants were asked the same questions concerning whether they had ever used a drug type for recreational purposes, and whether they had used that drug type in the past 12 months. Questions concerning the use of nitrous oxide were added to the list of drug types asked about in the 2003 survey wave and were included in all subsequent survey waves. In 2006, BZP party pills were included as a drug type asked about.
The drug types were described to the respondents by the interviewer in the same way in each survey wave. The questions about amphetamine referred to the broad class of amphetamines, which the interviewer described as ‘amphetamines, uppers, speed, methamphetamine’. There was a separate category for ‘ice’ which was described by the interviewer as ‘crystal methamphetamine’.
All of those respondents who reported using a drug type in the past 12 months were asked whether they were using ‘more’, ‘less’, ‘the same’ or had ‘stopped’ using the drug type compared to a year ago. Those who had used tobacco had not been asked this question in previous surveys.
The age range of the survey samples for the 2001 NDS, 2003 HBS-Drug Use and 2006 national household survey of party pill use were truncated to those aged 15–45 years old to allow valid comparisons back to the 1998 NDS. The general population RDD samples of those aged 15–45 years old from each survey wave were then compared. The respective sample sizes for each survey wave were: 5475 in 1998; 5504 in 2001, 3042 in 2003, and 1902 in 2006. The response rates for the survey waves were 79% in 1998, 80% in 2001, 68% in 2003, and 69% in 2006.*
(*The response rates quoted are for the original age ranges of the surveys. It was not possible to recalculate the response rates for the different surveys for the truncated age range as we cannot distinguish the non-response by age.)
The sample data were weighted by eligible household size to adjust for the selection of only one person from each household. Comparisons of drug measures were made between the 2006 survey and the three other survey waves conducted in 2003, 2001, and 1998. Significance testing controlled for multiple tests with an overall alpha level of 0.05. Mean prevalence and 95% confidence intervals for prevalence variables were calculated using logistic regression. To ensure reliable statistical comparisons we restricted our analysis to the drug types which included 10 or more respondents in the 2006 survey wave.
To provide an overall quantitative measure of the change in level of use we calculated the mean score for each drug type by enumerating the scale provided. These values were defined as follows:
One-way ANOVAs were used to test for differences in the mean score for a question between the 2006 survey and the other survey waves. One-way ANOVAs assume the samples tested form a normal distribution. With scale-type questions such an assumption can never be met as the scores are based on discrete data. However, frequency tables show the distribution of data as being mound shaped, providing an approximation of a normal probability distribution. All analysis was completed in the SAS statistical environment and controlled for the effects of weighting and stratification.
Lifetime use of different drug types—Alcohol tobacco and cannabis remained the most commonly tried drug types in all four survey waves (Table 1). In 2006, BZP party pills emerged as the fourth most widely tried drug type.
Table 1. Survey respondents who ever tried different drug types: 1998, 2001, 2003, 2006
n/a: not applicable (question was not asked in survey wave);
BZP: benzylpiperazine; LSD: lysergic acid diethylamide; MDMA: methylenedioxymethamphetamine.
A higher proportion of the sample had tried alcohol in their lifetimes in 2006 compared to 2003 (89.5% vs 83.7%, p<0.0001) and compared to 2001 (89.5% vs 86.4%, p=0.0038). A higher proportion of the sample had tried alcohol in 2006 compared to 1998 (89.5% vs 87.6%) and this difference was close to being statistically significant (p=0.0683).
A lower proportion of the sample had tried tobacco in 2006 compared to 2001 (57.6% vs 63.9%, p<0.0001) and compared to 1998 (57.6% vs 64.4%, p<0.0001). A lower proportion of the sample had tried cannabis in 2006 compared to 2003 (44.1% vs 53.8%, p<0.0001) and compared to 2001 (44.1% vs 52.1%, p<0.0001) and compared to 1998 (44.1% vs 50.4%, p<0.0001). These differences in lifetime cannabis use were due to a particularly low level of lifetime cannabis use reported in the 2006 survey.
A higher proportion of the sample had tried amphetamine in 2006 than in 1998 (9.3% vs 7.6%, p=0.0315). A higher proportion of the sample had tried crystal methamphetamine in 2006 compared to 1998 (1.8% vs 0.2%, p<0.0001). A higher proportion of the sample had tried kava in 2006 compared to 2003 (7.6% vs 5.9%, p=0.0303). A higher proportion of the sample had tried nitrous oxide in 2006 than in 2003 (7.4% vs 4.3%, p<0.0001). A higher proportion of the sample had tried cocaine in 2006 compared to 2003 (4.5% vs 3.1%, p=0.0225) and compared to 2001 (4.5% vs 3.3%, p=0.0289).
Use of different drug types in the past year—Alcohol, followed by tobacco and cannabis, were the drug types most commonly reported used in the past year in all the survey waves (Table 2). In 2006, 16.1% of the survey sample reported using BZP party pills in the past year and this approached the level of cannabis use in that survey wave. The next drug types most commonly used in the past 12 months in 2006 were ecstasy (MDMA), amphetamines, and nitrous oxide.
Table 2. Survey respondents’ last year use of different drug types: 1998, 2001, 2003, 2006
n/a: not applicable (question was not asked in survey wave);
BZP: benzylpiperazine; LSD: lysergic acid diethylamide; MDMA: methylenedioxymethamphetamine.
Change in level of use of different drug types—There was a relative increase in the level of alcohol drinking in 2006 compared to 2003 (2.8 vs 2.7, p<0.0001) and compared to 2001 (2.8 vs 2.3, p<0.0001) and compared to 1998 (2.8 vs 2.3, p<0.0001) (Table 3).
Table 3. Recent change in level of use for all drug types: 1998, 2001, 2003, 2006
The statistical power of our survey waves to detect changes in drug use can be summarised as follows. If the last year use of amphetamine increased by 2% in 2006 compared to 2003 (e.g. from 4.0% to 6.0%) then, based on the sample sizes available, we would have an 89% chance of detecting the increase. Alternatively, if amphetamine use fell by 2% we would have a 98% chance of detecting the decrease.
Alcohol, tobacco, and cannabis remained the most widely used drug types in New Zealand over the entire 8-year period of national surveying. We found mixed evidence of rising levels of alcohol use. The lifetime use of alcohol appeared to be tracking upwards while the last year prevalence of alcohol use was largely stable through the four waves of surveying. A higher proportion of alcohol drinkers reported drinking ‘more’ alcohol and a lower proportion reported drinking ‘less’ or to have ‘stopped’ drinking in 2006 compared to the earlier survey waves.
Census statistics show the total volume of alcohol available for consumption in New Zealand has increased almost every year since 1998, reaching its highest recorded level in 2006 since the series was begun in 1986.10 Increasing lifetime use of alcohol is likely to reflect a range of environmental factors including more liberal public attitudes to alcohol use, the lowering of the drinking age, the expansion of places where alcohol can be purchased and consumed, the liberalisation of alcohol advertising, and the introduction of ready-to-drink (RTD) alcohol products marketed at young people.2
Huckle at al11 found evidence of increased alcohol-related social disorder and driving offences among young people in New Zealand from 1990 to 2003 following the liberalisation of alcohol policy during the 1990s.
We found a consistent decline in the lifetime use of tobacco in each of the four waves of surveying. The decline in lifetime levels of tobacco use appears to reflect societal shifts in tolerance to smoking, increases in tobacco prices through taxation, and the impact of stricter regulation of smoking in semi-public areas such as workplaces, public transport, restaurants, and bars.4
The rapid emergence of BZP party pills, as measured in the 2006 wave of surveying, introduced a fourth most widely used drug type in New Zealand. The extent of BZP party pill use in New Zealand at this time was to our knowledge unique in the world.
Prior to 2006, BZP had been prohibited in several countries including the United States, Sweden, Denmark, Belgium, Greece, and most states in Australia.12,13,14 Low levels of BZP use were occurring in some other European countries during this time, including the United Kingdom.15
The use of BZP party pills in New Zealand has been linked with a number of hospital Emergency Department presentations which ranged from minor problems involving anxiety, headaches, and insomnia, to more serious incidents such as collapse, seizures, and renal failure.12,16,17
A national household survey of BZP party pill use in New Zealand conducted in 2006 found most last year users of BZP pills reported fairly minor physical problems from BZP use, such as ‘insomnia’ (50% of last year users), ‘poor appetite’ (41%), ‘nausea’ (22%), and ‘headaches’ (22). However, some users reported potentially more serious physical problems such as ‘heart palpitations (15%), ‘vomiting’ (12%), ‘chest pains’ (4%), ‘passing out’ (0.8%), and ‘seizures’ (0.3%).18
Eighty-six percent of the BZP users combined their use of BZP with other drug types, including alcohol (91%), tobacco (40%), cannabis (22%), and 5-HTP (5-hydroxytryptophan) ‘recovery’ pills (9%) and ecstasy (MDMA) (5%).
In November 2006, the Government signalled its intention to follow the recommendation of the Expert Advisory Committee on Drugs (EACD) to schedule BZP party pills as a Class C controlled drug under the Misuse of Drugs Act 1975 (i.e. the same category as cannabis).19 The New Zealand Parliament passed the bill to schedule BZP as a Class C drug in March 2008. In the same month, the Council of the European Union announced that BZP would be subject to ‘control measures and criminal provisions’ across all European Union Member States.25
We found a general trend toward greater use of amphetamine type stimulant (ATS) drug types—such as amphetamine, crystal methamphetamine, and ecstasy (MDMA)—in New Zealand since 1998. The use of amphetamine appears to have peaked in 2001 and levelled off after that following the introduction of greater legal penalties, increased law enforcement, and greater public awareness of the health risks of methamphetamine use.5,20
Ecstasy (MDMA) has sustained a more consistent increase in use over this time with more of the sample having used ecstasy (MDMA) in the past 12 months in 2006 compared to 1998. The situation with ecstasy in New Zealand is somewhat confused, however, by the presence of a large legal market for BZP. Drug dealers sometimes sell BZP as ecstasy to earn higher black market prices, and chemical analysis of alleged ecstasy pills has uncovered a range of substances, including BZP. Another stimulant, cocaine, was also found to have higher levels of last year use in 2006 compared to 2003.
We found a surprising decline in both the lifetime and last year use of cannabis in 2006 compared to the previous survey waves. It is difficult to know at this early stage whether the 2006 result will translate into a sustained trend toward lower levels of cannabis use in New Zealand. The emergence of several ‘new’ synthetic stimulant drugs over the past 5 years—such as methamphetamine, ecstasy, and BZP pills—may have impacted negatively on levels of cannabis use.
Stimulants are currently more consistent with underlying cultural trends among young people which value productivity and success in both the social and professional spheres.21
Cannabis, on the other hand, is often associated with the counter culture of the 1960s and 1970s, and a desire to opt out of mainstream ambitions.22 Recent research in Australia has found that young people now place negative social and health connotations on cannabis use, and associate cannabis smoking with the health risks of tobacco smoking.23
There was an increase in the level of use of both amphetamine and crystal methamphetamine by last year users of these drug types in 2006 compared to 2003. These findings suggest that there has been some entrenchment in amphetamine use in recent years since the reported levelling out of its prevalence of use in 2003.20 Heavy dependent users are less responsive to stiffer legal penalties and evidence of health harms from use. New Zealand has relied on voluntary agreements with chemical suppliers and pharmacies to control precursor chemicals used to manufacture methamphetamine. Evidence from overseas indicates that stricter more formal controls of precursor chemicals can have an impact on levels of methamphetamine use and harm.26
The entrenchment of amphetamine use can be viewed as part of the natural lifespan of this drug trend where occasional and experimental users cease use due to growing awareness of health harms and increasing legal pressures, and leave a residual user population of heavy and dependent users.24 (pp 287–90).
Paradoxically, this type of entrenchment can lead to growing social costs related to the use of a drug type even when its prevalence of use is in decline24 (pp 287–90).
Competing interests: None known.
Author information: Dr Chris Wilkins, Senior Researcher; Paul Sweetsur, Statistician; Centre for Social and Health Outcomes Research and Evaluation (SHORE), Massey University, Auckland
Acknowledgements: We gratefully acknowledge the different funding sources involved in each of household surveys. The 1998 and 2001 National Drug Surveys were funded by contestable research grants from the Health Research Council (HRC) and partially by direct funding from the Ministry of Health. The 2003 HBS-Drug Use was directly funded by the Ministry of Health and carried out as part of the Public Health Intelligence (PHI) Health Behaviours Survey Monitor.
The legal party pill survey was funded from the 2005/6 round of the National Drug Policy Discretionary Fund (NDPDF). The NDPDF is jointly managed by the Inter-Agency Committee of Drugs (IACD) and the Ministerial Committee on Drug Policy (MCDP). The national household comparison analysis presented in this paper was funded from the 2006/7 NDPDF.
In addition, we acknowledge all the researchers and interviewers who worked on the different survey waves and all those members of the New Zealand public who participated in the surveys.
Correspondence: Dr Chris Wilkins, Centre for Social and Health Outcomes Research and Evaluation (SHORE), Massey University, PO Box 6137, Wellesley Street, Auckland, New Zealand. Fax: +64 (0)9 3665149; email: email@example.com
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads)
Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals