Explanatory notes

The 1998 National Drug Strategy Household Survey

The 1998 National Drug Strategy Household Survey was the sixth in a series which commenced in 1985. The Australian Institute of Health and Welfare was commissioned in October 1997 by the Commonwealth Department of Health and Family Services to manage the survey. The Institute was assisted in this task by a Departmental Policy Reference Group and a Technical Advisory Committee which comprised members from the Australian Institute of Aboriginal and Torres Strait Islander Studies, the Australian Bureau of Criminal Intelligence, the National Drug and Alcohol Research Centre, the Australian National University's Research School of Social Sciences, and the Department's Research and Marketing Group.

Between June and September 1998 over 10,000 Australians aged 14 years or older selected from a geographic stratified random sample of 8,357 private dwellings, completed the survey. Note that this did not include institutionalised or homeless people. Persons aged 14 or 15 years participated with the consent of a parent or guardian. The survey instrument comprised items on licit and illicit drug-related awareness, knowledge, attitudes and behaviours. Approximately 40% of the sample completed the survey through personal interviews and a self-completion booklet for the more sensitive issues. The balance of the sample self-completed the entire questionnaire. The response rate achieved across the whole sample was 56%.

Data presented in this release are based on estimates derived from responses weighted to the total Australian population aged 14 years and over. Comparative data for 1995 were drawn from the National Drug Strategy Household Survey of that year.

Because the results are based on survey responses, they are subject to sampling variability. As an example, the number of 14-19 year old persons recently using marijuana/cannabis is shown as 547,000. Strictly speaking, we can only be 95% certain that the value is in the range 499,000 to 595,000. The size of the range varies with the magnitude of the estimate, the size of the population for the which the estimate is made, and the degree of certainty we want to attach to the range. Caution is therefore recommended in interpreting low prevalence results: in general, prevalence estimates below 1% are not statistically robust because the margin of error is very large.

Drug-related mortality and morbidity

Estimates of drug-related deaths and hospital separations are based on the application of aetiological fractions to data contained in the Institute's National Mortality Database and National Hospital Morbidity Database.

The aetiological fraction, which is also known as the attributable proportion or attributable risk, is a form of indirect quantification of drug-caused morbidity and mortality. Indirect methods involve the estimation of a probability measure of the likelihood of drug causation which is then applied to the total numbers of deaths, illnesses or injuries

due to a specific cause. The distinguishing characteristic of indirect methods is that individual drug caused cases are not identified. For example, if the probability that a case of low birthweight is caused by illicit drug use is .32 (that is a 32% likelihood), then the product of this aetiologic fraction and the total number of low birthweight babies in a population results in an estimate of all low birthweight cases in that population that can be attributed to illicit drug use. Estimates in this release are based on the English and Holman (1995) derivations for the year 1992.

Reference: English, DR, Holman, CDJ, Milne, E, et al. 1995. The quantification of drug caused morbidity and mortality in Australia, 1995 edition. Canberra: Commonwealth Department of Human Services and Health.

31 March 1999


Further information: Geoff Sims, ph. 02 6244 1168, or Paul Williams, ph. 02 6289 6851.
For general media enquiries: Lyn Elliott, ph. 02 6244 1034.