Publication Highlights

The study estimated the prevalence and frequency of amphetamine use in a cohort of Australians aged in their mid‐30s and compared this with National Drug Strategy Household survey (NDSHS) estimates.

The study design was a cross‐sectional analysis of wave 10 data collected in 2014 from the Victorian Adolescent Health Cohort Study (VAHCS): a sample of 1435 persons originally selected in a stratified, random community survey of secondary school students that commenced in 1992.

The current results show that lifetime, current and frequent amphetamine use was common amongst adults in the fourth decade of life in this cohort, and associated with the experience of social disadvantage, poor mental health and living in a social context in which drug use is the norm.

Almost one in four (23.2%) VAHCS respondents reported lifetime amphetamine use and 6.5% were identified as currently (in the past 12 months) using amphetamines. A quarter (26%) of these current users (or 1.7% of all respondents) were using amphetamines weekly or more often. These figures are much greater than the published NDSHS estimates which report that 3.1% of 30–39‐year‐old Australians had used amphetamines in the past year, and only 15.3% of those using amphetamines in the past year (<0.5% of 30–39‐year‐old Australians) were using amphetamines weekly or more frequently. The results suggest that amphetamine use is much more common among adults in the fourth decade of life in contemporary Australia than suggested by household surveys.

These differences cannot be attributed to flaws in the VAHCS design, write the authors. The sample was randomly drawn from the population of Victorian high school students and early waves demonstrated a very high participation rate. Further, sample weights were generated using information on the schools recruited to the study and respondents themselves (including early adolescent smoking) to adjust for attrition over time and minimise potential bias. While the estimate of current (12‐month) amphetamine use is more than double that based on the NDSHS, this magnitude of difference is not evident for other, less‐stigmatised illicit drugs. For example, the VAHCS estimate of current cannabis use (14.3%, 95% CI 12.6–16.4%) is more consistent with the NDSHS estimate for the relevant age group (12.3%). This suggests the current findings are not simply a reflection of higher overall reported rates of substance use in the VAHCS sample.

The authors suggest that the ongoing participation in the VAHCS for over 20 years may have engendered greater trust and commitment amongst VAHCS respondents, which may have reduced social desirability bias and increased the accuracy with which respondents answered sensitive survey questions. The difference between the current findings from wave 10 of the VAHCS and the NDSHS results do reinforce concerns about the potential impact of the household sampling frame, low response rate and potential social desirability bias in household surveys. The current results may indicate that these biases can be reduced in well‐designed longitudinal studies and suggest further investigation of panel conditioning in this field of research is warranted.

Compared to the previous published report of amphetamine use in the VAHCS when respondents were aged 24 years, the current analysis found a lower level of 12‐month use (6.5% vs. 12%) but a greater proportion of frequent use (26% vs. 11%). Over a period of the life course when levels of drug use generally decline the stability or increasing rates of frequent amphetamine use likely reflects change in the form of drug used tied to a transition to crystal methamphetamine and increased drug purity. These would be expected to increase health, social and personal harms related to use.