Mortality in the SuperMIX cohort of people who inject drugs in Melbourne, Australia: a prospective observational study

AIMS: To measure mortality rates and factors associated with mortality risk among participants in the SuperMIX study, a prospective cohort study of people who inject drugs. DESIGN: A prospective observational study using self‐reported behavioural and linked mortality data. SETTING: Melbourne, Australia. PARTICIPANTS/CASES: A total of 1209 people who inject drugs (67% male) followed‐up between 2008 and 2019 for 6913 person‐years (PY). MEASUREMENTS: We linked participant identifiers from SuperMIX to the Australian National Death Index and estimated all‐cause and drug‐related mortality rates and standardized mortality ratios (SMRs). We used Cox regression to examine associations between mortality and fixed and time‐varying socio‐demographic, alcohol and other drug use and health service‐related exposures. FINDINGS: Between 2008 and 2019 there were 76 deaths in the SuperMIX cohort. Of those with a known cause of death (n = 68), 35 (51%) were drug‐related, yielding an all‐cause mortality rate of 1.1 per 100 PY [95% confidence interval (CI) = 0.88–1.37] with an estimated SMR of 16.64 (95% CI = 13.29–20.83) and overall accidental drug‐induced mortality rate of 0.5 per 100 PY (95% CI = 0.36–0.71). Reports of recent use of ambulance services [adjusted hazard ratio (aHR) = 3.77, 95% CI =1.78–7.97] and four or more incarcerations (aHR = 2.78, 95% CI = 1.55–4.99) were associated with increased mortality risk. CONCLUSIONS: In Melbourne, Australia, mortality among people who inject drugs appears to be positively associated with recent ambulance attendance and experience of incarceration.