Utilising data linkage to describe and explore mortality among a retrospective cohort of individuals admitted to residential substance use treatment

Drug Alcohol Rev. 2021 Nov;40(7):1202-1206. doi: 10.1111/dar.13279. Epub 2021 Sep 29.


Introduction: Individuals with moderate to severe substance use disorders may seek residential treatment, but outcomes after discharge for this hard-to-reach population are often unknown. Using linked data, we examined mortality outcomes among individuals after residential treatment.

Methods: We included 1056 individuals admitted to three residential treatment facilities across Queensland, Australia, from 1 January 2014 to 31 December 2016. Records were linked to Queensland death registration data and cause of death records from the Australian Coordinating Registry (1 January-31 December 2018). Standard mortality ratios were assessed, comparing participants to the Queensland, Australia, general population. Causes of death and years of potential life lost (YPLL) were examined.

Results: Thirty-six participants died (3.4%) in 3408 years of follow-up data. The age- and sex-adjusted standard mortality ratios were 3.96 (95% confidence interval: 2.78, 5.48) overall, 8.19 (3.74, 15.55) in females and 3.38 (2.23, 4.92) in males. Two-thirds of deaths were due to suicide/overdose. There was an average of 45.50 YPLL (SD 9.16).

Discussion and conclusions: This study used linked data to quantify mortality following residential substance use treatment. The YPLL and avoidable nature of deaths highlight the need for continuing care following discharge from residential services.

Keywords: data linkage; mortality; residential treatment; standardised mortality ratio; substance use disorder.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Australia / epidemiology
  • Cause of Death
  • Female
  • Humans
  • Information Storage and Retrieval
  • Male
  • Retrospective Studies
  • Substance-Related Disorders* / therapy