Prevalence of opioid dependence in Scotland, 2015-2020: results from a Multi Parameter Estimation of Prevalence model

Background: In Scotland, there is an ongoing public health emergency in opioid drug-related deaths, with the number having more than doubled between 2011 and 2021, rendering the population rate among the highest in Europe. Robust estimates of the number of people with opioid dependence are urgently needed to inform policy makers and aid evaluation of interventions. In this study, we estimated the number of people with opioid dependence aged 15 to 64 years resident in Scotland, for each year 2014/15 - 2019/20. We defined people with opioid dependence as those with the potential to benefit from Opioid Agonist Treatment (OAT), irrespective of current receipt of such treatment.
Methods: We used a Bayesian Multi Parameter Estimation of Prevalence (MPEP) model to estimate prevalence jointly from opioid-related mortality and hospitalisations data. We used linked administrative data from the Scottish Public Health Drug Linkage Programme, which includes OAT prescription records and records of all opioid-related deaths and hospitalisations in Scotland.  Prevalence is estimated through an assumption that adverse event rate models estimated from all people in recent (last five years) but not current OAT apply also to the “unobserved” (i.e. no recent OAT) people with opioid dependence. The model is described more fully in a related abstract. While our primary results were based on joint modelling of mortality and hospitalisations data, we will also discuss estimates based on each data source separately, as a check on the consistency of the evidence and robustness of conclusions. 
Results: The estimated number and prevalence of people with opioid dependence in Scotland in 2019/20 were 47,100 (95% Credible Interval, CrI, 45,700 to 48,600) and 1.32% (95% CrI 1.28% to 1.37%), respectively. Estimated prevalence in Glasgow, Scotland’s most populous city, was 1.77% (95% CrI 1.69% to 1.85%). OAT coverage in Scotland in 2019/20 was estimated as 61%, with 74% having received OAT at some point in the last five years. There was evidence of a small reduction in prevalence since 2014/15 (estimated reduction -0.07%, 95% CrI -0.14% to 0.00%). Results also show that the population is ageing, with estimates of the number of people with opioid dependence aged 15-34 reducing from 17,200 (1.24%) in 2014/15 to 12,100 (0.87%) in 2019/20, and numbers aged 50-64 increasing from 4,600 (0.43%) to 7,400 (0.65%) over the same period. Sensitivity analyses showed some small differences in estimates based on each data source separately, mostly driven by differences in the estimated number of men aged 15-34 years.
Conclusions: Despite the sustained high prevalence of opioid dependence in Scotland, there is no evidence of an increase between 2014/15 and 2019/20. Increased drug-related deaths can therefore be attributed to increased risk among people with opioid dependence, rather than an increasing population size.