Prevalence of problem drug use in Scotland: 2015/16 Estimates
A review of definitions and statistical methods

About this release

This release by Public Health Scotland (PHS) provides a follow-up to previously published estimates of the prevalence of Problem Drug Use (PDU) in Scotland between April 2015 and March 2016. The original definition was limited to opioids and benzodiazepines. Using data collected for the original study, broader definitions of PDU have been constructed for Scotland based on the addition of substances such as cocaine (including crack cocaine), amphetamines (including amphetamine-type substances) and cannabis (including synthetic cannabinoids).

This release presents patterns in the resulting estimates at Scotland level and explains limitations in the wider application of the estimation approach in a review of the underlying statistical methods and data sources.

Main points

Definition 1: Opioids (including illicit and prescribed methadone use) and/or the illicit use of benzodiazepines

Definition 2: As definition 1, plus illicit use of cocaine and amphetamines / amphetamine type substances

Definition 3: As definition 2, plus illicit use of cannabis / synthetic cannabinoids

• It is possible to generate estimates of prevalence for wider definitions of Problem Drug Use using log-linear capture/recapture techniques but the robustness of the estimates may decrease.
• Estimated prevalence of problem drug users in 2015/16 increased from 1.62% under the original definition of opioids and benzodiazepines (definition 1), to 1.91% with the further inclusion of cocaine and amphetamines (definition 2), and to 2.51% with the inclusion of cannabis (definition 3).
• The percentage of problem drug users that are male increased from 71% (definition 1) to 74% (definition 2) and 77% (definition 3) as the criteria progressively widened.
• The percentage of problem drug users that are in the oldest age category (35-64) decreased from 64% (definition 1) to 58% (definition 2) and 52% (definition 3) as the criteria progressively widened.
• Future studies should aim to capture representative samples of the target population using routine data, and avoid being heavily correlated.
• It is recommended that future studies are not only restricted to replicating the current methodology and data sources.

Background

Although the study has been replicated a number of times in Scotland since 2000, no previous attempt has been made to systematically estimate the national prevalence of Problem Drug Use in a way that captures the extended problematic use of other substances. As part of the commission of the 2015/16 study, Scottish Government asked PHS to consider, in addition to the established definition of PDU, the feasibility of estimating prevalence for a wider definition of drug types. This report contains further estimates for 2015/16 and recommends further methodological considerations for the design of future studies based on the experience of this work.

Further information

Data from this publication are available from the topic page (external website).