Changing age profile and incidence of injecting drug use initiation among people in Australia who inject drugs: evidence from two national repeated cross-sectional studies.

Background: The age of people who inject drugs appears to be increasing in some high-income countries. We aimed to explore trends in the age of people injecting and recently starting to inject drugs in Australia, and to model incidence of initiation over time. Methods: We obtained data from the Illicit Drug Reporting System (IDRS) and the Australian Needle Syringe Program Survey (ANSPS), which comprise annual cross-sectional surveys with people who inject drugs (2000-2019). Outcome measures were current age, age of initiation, and time since initiation (both surveys), and drug first injected (IDRS only). We estimated time trends in age using quantile regression. We used the relative number of people initiating injecting drug use each year and existing population size estimates to model incidence of initiation. Findings: In total, 58,465 interviews with people who reported injecting drugs in the past month (33.7% women) were included. In both surveys, the median age increased over the study period (IDRS: 28-43 years; ANSPS 28-44 years). The median time since initiation also increased over the study period (IDRS: 8-22.5 years; ANSPS: 9-24 years), and the median age of initiation increased with calendar year of initiation (IDRS: 18-34 years; ANSPS: 20-30 years). Women tended to be younger and commence injecting drug use at a younger age than men. Modelling suggested that similar numbers of people injected drugs for the first time each year from 1980 (national estimates of 15,570 and 17,020 new initiates using IDRS and ANSPS data, respectively) to 1996 (IDRS: 16,360; ANSPS: 22,300 new initiates), followed by a decisive decline in incidence until 2012 (IDRS: 1110; ANSPS: 1830), whereafter it has fluctuated but remained low. Meth/amphetamine was consistently the predominant drug injected at initiation among IDRS participants, although there were peaks in heroin as the drug injected at initiation in the early 1980s and mid-1990s. Interpretation: In 2019, most people who were injecting drugs in Australia were part of a cohort that began injecting in the 1980s or 1990s. Consequently, the population was older and had been injecting drugs for longer compared to those injecting drugs in 2000. This has implications for health service delivery to people who inject drugs, with increasing age likely to be accompanied by a rise in chronic health conditions and an increase in injecting duration potentially resulting in higher incidence of injecting-related injuries and diseases. Funding: Australian Government Department of Health and Aged Care.