All-cause mortality and overdose deaths among 4192 people who inject drugs in Stockholm: a 10-year register-based cohort study

Abstract Background People who inject drugs (PWID) face premature mortality, particularly from opioid overdose. In Sweden, harm reduction has expanded, including increased access to opioid agonist therapy (OAT) and the 2018 introduction of Take-Home Naloxone (THN). This study aimed to examine trends in all-cause and cause-specific mortality and to estimate predictors of all-cause and opioid overdose mortality among PWID in Stockholm. Methods We conducted a retrospective cohort study from April 2013 to March 2023. Data from the national Cause of Death Register were linked to Stockholm Needle and Syringe Program (NSP) records. Causes of death were categorised as opioid overdoses, external causes, internal/natural causes, or other/unknown. Crude mortality rates and age- and sex-standardised mortality ratios were calculated. Time-dependent Cox regression models estimated risk of all-cause death, and Fine and Gray subdistribution hazard models estimated opioid overdose mortality, accounting for competing risks. Person-time began at first NSP visit and ended at death, study end or censoring (> 365 days without a visit). Results Among 4192 participants, 685 (16%) died. The crude mortality rate declined from 36.75 to 27.04 deaths per 1,000 person-years and the standardised mortality ratio from 17.24 to 10.94. In multivariable models, reporting opioids as the latest injected drug was the strongest driver for both all-cause and opioid mortality. Other significant predictors included infrequent injecting and current contact with social services, psychiatry or addiction care, while OAT participation was associated with a lower risk. Male sex and age over 56 at enrolment were associated with a higher risk of all-cause mortality. Opioid overdose was the most common cause of death (53%). However, the opioid overdose mortality rate declined from 29.40 to 5.88 deaths per 1,000 person-years over the study period, coinciding with the 2018 introduction of THN and declining reported opioid injecting drug use among NSP clients. Conclusions All-cause mortality among PWID in Stockholm declined over the study period, alongside significant reductions in opioid overdose deaths, during a period of broadened harm reduction and reduced reporting of opioid injecting drug use. Our findings support continued scale-up of OAT and THN and consideration of supervised consumption sites to further reduce preventable deaths.