Using primary care data to understand Opioid Prescribing, Policy Impacts and Clinical Outcomes: A protocol for the OPPICO Study

Study type

• Study type:
  Non-interventional study
• Scope of the study:
  Disease epidemiology

  Drug utilisation

Non-interventional study

• Non-interventional study design:
  Cohort

Population studied

• Age groups:
  • Adolescents (12 to < 18 years)
  • Adults (18 to < 46 years)
  • Adults (46 to < 65 years)
  • Adults (65 to < 75 years)
  • Adults (75 to < 85 years)
  • Adults (85 years and over)
• Estimated number of subjects:
  700000

Study design details

• Main study objective:
  To evaluate the impact of recent Australian policies on opioid prescribing in primary care, understand the impact of recent Australian primary care opioid prescribing guideline recommendations on specific conditions, use exploratory analysis to examine the different patterns and outcomes of opioid cessation, and explore the use and correlates of non-opioid and non-pharmacological interventions.
• Outcomes:
  Opioid exposure: Opioid prescribing (e.g. change in primary care in opioid dose, opioid initiation and cessation, and prescribing patterns within specific populations), Utilisation of non-opioid medicines: Average daily doses of non-opioid types of medicines commonly used among people prescribed opioids (e.g. analgesics, gabapentinoids, benzodiazepines). Utilisation of non-pharmacological interventions: Non-pharmacological interventions will be defined as referrals to relevant healthcare providers.
• Data analysis plan:
  To explore the changes in opioid prescribing after policy change, interrupted time series analyses will be used. Descriptive statistics will be used to describe the demographic and other characteristics of the cohort prescribed opioids prior to and after the policy change. Multivariable logistic regressions will be then used to examine predictors of receiving an opioid prescription pre and post policy change. Differences between groups in terms of odds of opioid use will be expressed as Odds Ratios (ORs) derived from logistic regressions. To explore the patterns of opioid cessation among long term opioid users, group-based trajectory modelling will be used to determine determine key subgroups of people prescribed and ceased opioids and compare outcomes for each of the trajectories via stand‐alone trajectory modelling. Multinomial regression and descriptive analysis will be used to examine our sample characteristics that are associated with trajectory groups.