A comparison of different methods for estimating the prevalence of problematic drug misuse in Great Britain

Abstract

Aims. The European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) has produced methodological guidelines for national drug prevalence estimation. This paper pilots the methods to estimate prevalence for Great Britain and provides a commentary on the methods and resulting estimates. Design. Three types of methodology were used to estimate prevalence: (a) the multiple indicator (MI) method, (b) multipliers applied to (i) drug-treatment records (ii) HIV estimates and (iii) mortality statistics and (c) the British/Scottish Crime Surveys. Setting. England, Scotland and Wales. Participants. Aggregated data on people recorded on databases and respondents in household surveys. Measurements. Prevalence estimates of different forms of problematic drug use. Findings. The estimates are 161 133 (range: 120 850-241 700) for people at risk of mortality due to drug overdose; 161 000-169 000 for people who have ever injected drugs; 202 000 (range: 162 000-244 000) problem opiate users and 268 000 problem drug users (all types). Conclusions. The multiple indicator method offers a comprehensive approach to estimating the prevalence of problematic drug use in the United Kingdom. Simple multiplier methods and household surveys also provide a range of estimates corresponding to different types of drug use in the United Kingdom. The current study suggests that previous national estimates of 100 000-200 000 were conservative. The new estimate of 161 000-266 000 should enable a more focused response. For further development of this method, reliable and timely estimates of anchor points are required for speci c geographical areas such as cities or Drug Action Teams (DAT), as well as routine aggregation of drug indicators for these areas.

Comparing population estimates for drug users, the prevalence of blood-borne viruses and the provision of services in one London borough: Evidence of a gap in provision?

“…The methods used in this study, particularly the capture-recapture and multiplier methods, have been employed elsewhere, both for local and national (including Home Office) estimates of drug use (Hartnoll, 1985;Hickman, 1999, Frischer, 2001. However, for a small area study such as this we would expect variation in the different estimates, and as such we have used cross validation and comparison with other studies to corroborate our findings.…”

“…The multiplier in this example, although derived from a different population, is robust enough to use in other settings (Frischer, 2001;Hartnoll, 1997).…”

“…One of our estimates for injecting users (617) was derived from an HIV dataset and is in fact lower than the number of known injectors (1000). Limitations of the HIV multiplier method include assumptions that all HIV-infected injectors are recorded, and that these individuals are still injecting (Frischer, 2001). Inclusion of this value has reduced our overall estimate for injecting users which was derived from three sources.…”

Acute pain management for opioid dependent patients

“…A recent comparison made of different methods of estimating the number of 'problematic opioid users' in Great Britain, arrived at the figure of about 202 000 [9].…”

Prevalence of Problem Drug Use and Injecting Drug Use in Luxembourg: A Longitudinal and Methodological Perspective

“…Indirect estimation techniques tend to build upon existing data and their mutual relationship in order to estimate the number of non-observed cases in a given population. The latter include capture-recapture techniques [1,2,3,4,5,6,7,8], benchmark-multiplier methods [1,9,10,11], the truncated Poisson method [1,12,13], and the multivariate indicator method [14,15,16,17,18,19]. However, none of currently known methods is considered to be the gold standard since they all rely on a series of assumptions and limitations [20].…”