Characteristics of oxycodone-related ambulance attendances: analysis of temporal trends and the effect of reformulation in Victoria, Australia from 2013 to 2018

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Background and aims: In Australia, oxycodone has been associated with increasing rates of harm over time, despite reduced use, reformulation to a tamper-resistant form and in contrast to most prescription opioids. We explored characteristics of oxycodone-related ambulance attendances to understand whether presentation characteristics could explain increasing oxycodone harm. Design: Retrospective study of coded ambulance patient care records related to extramedical oxycodone use, January 2013 to September 2018. Setting: Victoria, Australia. Cases: A total of 2788 oxycodone-related ambulance attendances. Measurements: Primary outcomes were temporal changes in characteristics of oxycodone presentations over time (from 2013 to 2018) and following reformulation. Covariates include demographic characteristics, presentation severity, mental health, substance use and poisoning intent. Findings: Average age was 41.3 (± 16.4) years with females comprising 56.4% of attendances (n = 2788). The proportion of females in oxycodone-related attendances increased over time [an average increase in the odds ratios of 5% per year (OR) = 1.05, 95% confidence interval (CI) = 1.01–1.10]. Other temporal trends included a reduced likelihood of naloxone administration (OR = 0.92, 95% CI = 0.85–1.00), heroin involvement (OR = 0.81, 95% CI = 0.66–0.99), comorbid mental health symptoms (OR = 0.87, 95% CI = 0.82–0.92) and unknown intent poisoning (OR = 0.91, 95% CI = 0.85–0.96); and a greater risk of alcohol involvement (OR = 1.06, 95% CI = 1.01–1.11), non-opioid extramedical pharmaceutical use (OR = 1.05, 95% CI = 1.01–1.10), comorbid suicidal thoughts or behaviours (OR = 1.10, 95% CI = 1.05–1.15) and past history of psychiatric issues (OR = 1.22, 95% CI = 1.16–1.27). Interrupted time–series analysis showed that reformulation was associated with an immediate effect on sex, severity, accidental poisoning and unknown intent poisoning, although these were not sustained over time. Alcohol involvement in the attendance (OR = 1.43, 95% CI = 1.17–1.74, i.e. an average increase in the odds of 43% per year), Glasgow Coma Scale (OR = 1.28, 95%CI 1.04 – 1.57), a previous history of psychiatric issues (OR = 0.80, 95% CI = 0.70–0.92, i.e. an average decrease in the odds of 20% per year, heroin involvement (OR = 0.22, 95% CI = 0.05 – 0.98) and illicit drug use (OR = 0.45, 95% CI = 0.23 – 0.87) showed statistically significant relative changes following the reformulation. Conclusions: The characteristics of oxycodone presentations in Australian ambulances attendances appear to be changing over time, including more female presentations; increasing alcohol use, extramedical use of non-opioid pharmaceuticals and suicidal thoughts or behaviours and decreasing heroin and illicit drug involvement.

Original languageEnglish
Pages (from-to)2233-2241
Number of pages9
JournalAddiction
Volume116
Issue number8
DOIs
Publication statusPublished - Aug 2021

Keywords

  • Extramedical use
  • opioid analgesic
  • overdose
  • oxycodone
  • pharmacoepidemiology
  • temporal trends

Cite this