Andreas Markoulidakis, Matthew Hickman, Megan Glancy, Andrew McAuley, Lee R Barnsdale, Tara Shivaji, Craig Collins, Jaroslaw Lang, Femke de Wit, Gordon Hunt, Sharon J Hutchinson, Louisa Degenhardt, Hayley E Jones
We estimate there were 43,600 (95% credible intervals 42,000-45,400) people with opioid dependence in Scotland in 2023/24, which is 1.13% (95% credible intervals 1.09%-1.18%) of those aged 15-69. Opioid agonist therapy coverage was estimated as 61%. Estimated prevalence was higher among those aged 50-69 (0.76%) than 15-34 (0.67%). Prevalence varied across regions, being highest in Greater Glasgow & Clyde (1.46%). Over the 10-year period, the estimated population size decreased by 4100 (95% credible intervals 6800 -1500).
INTRODUCTION: Scotland remains in the midst of a public health crisis driven by drug-related deaths, with rates reaching 2.0 per 100 person-years in 2020-22. Estimates of the prevalence of opioid dependence are critical for supporting policy decisions and better understanding the factors behind this rise.
METHODS: We applied a Bayesian multi-parameter estimation of prevalence approach to estimate the prevalence of opioid dependence from 2014/15 to 2023/24. We used linked opioid agonist therapy prescriptions, mortality and hospital admission data from the Substance Use and Health Intelligence Linked Dataset.
RESULTS: We estimate there were 43,600 (95% credible intervals 42,000-45,400) people with opioid dependence in Scotland in 2023/24, which is 1.13% (95% credible intervals 1.09%-1.18%) of those aged 15-69. Opioid agonist therapy coverage was estimated as 61%. Estimated prevalence was higher among those aged 50-69 (0.76%) than 15-34 (0.67%). Prevalence varied across regions, being highest in Greater Glasgow & Clyde (1.46%). Over the 10-year period, the estimated population size decreased by 4100 (95% credible intervals 6800 -1500).
DISCUSSION AND CONCLUSIONS: In Scotland, prevalence of opioid dependence remains high. There are signs of a modest decline but, notably, the reduction is less than the number of opioid-related poisoning deaths over the period modelled.