Estimating prevalence of problem drug use in Scotland

Ongoing
Scotland has one of the highest rates of drug-related deaths in Europe, with the number of these more than doubling between 2011 and 2020. At 250-300 per million population in 2021-22, Scotland’s rate of drug-related deaths was sixteen times higher than the average in the European Union and on par with rates in North America. 

As part of the response to the public health emergency in drug-related deaths, our ongoing research project, commissioned by the Scottish Government and in collaboration with Public Health Scotland and Glasgow Caledonian University (GCU), seeks to understand how many people are dependent on opioids in Scotland and whether this number is increasing or decreasing. This can’t be measured accurately through traditional household or general population surveys, so statistical models are needed to estimate this population size.  

To estimate how many people aged 15 to 64 years old are dependent on opioids, we extended and applied a Bayesian statistical modelling approach developed at the University of Bristol, the Multi-Parameter Estimation of Prevalence (MPEP) approach. The analyses utilise routinely collected data from the Substance Use and Health Intelligence Linked Dataset (SHIeLD), held by Public Health Scotland. Prevalence is estimated by fitting statistical models to administrative data records describing prescriptions of opioid agonist therapy (treatment), opioid-related deaths and hospitalisations for opioid overdose.

Project aims

To establish robust MPEP methodology that can be routinely applied to estimate the number of people, and prevalence (i.e. number as a percentage of the general population) who are dependent on opioids in Scotland.  

What we hope to achieve

Over the coming year, we will extend the model to provide estimates up to 2023/24We additionally aim to incorporate an additional data source into the MPEP model – data from the Scottish Ambulance Service describing administration naloxone (the antidote to opioid overdose). This will help us to take a closer look at the assumptions made in the MPEP model and make improvements if we find any problems. It will also allow us to give more detailed estimates for different regions. Over this time period we will also scope the feasibility of developing an MPEP model to produce estimates of the prevalence of cocaine dependence. 

Lead researchers

  • Professor Hayley Jones
  • Professor Matthew Hickman
  • Dr Andreas Markoulidakis
  • Prof Sharon Hutchinson (Glasgow Caledonian University)
  • Lee Barnsdale (Public Health Scotland) 

Partners

  • Public Health Scotland
  • Glasgow Caledonian University

Papers:

Estimated prevalence of opioid dependence in Scotland

Prevalence of opioid dependence in Scotland 2015–2020: A multi-parameter estimation of prevalence (MPEP) study