Direct measurements of the prevalence of the more common forms of drug use can be obtained from population surveys, such as the British Crime Survey (3) presented in Tables A1-A2. However, problem drug use, in particular injecting behaviour, heroin use and crack-cocaine is seriously under-estimated through population surveys. It is doubtful whether it can be measured reliably because it is too rare and because of sampling and non-response errors. For example, sub-groups of the population that are more likely to be problem drug users are less likely to respond to population surveys or to live in households that are sampled by population surveys.

Instead, indirect methods have been proposed as a way of Estimating Problem Drug Use (4). A number of these studies have been carried out in a selection of London boroughs, for example, using capture-recapture in 1992-95 for Lambeth, Southwark, Lewisham, Newham, Camden and Islington (5). This study suggested that the prevalence of problem drug use in areas of Inner London in the early to mid 1990s was as high as 3.3% (1 in 30 individuals) in people aged 15-49. And that the prevalence of problem opiate use, at least, in Lambeth, Southwark and Lewisham was greater than 1.0% (1 in 100) in the same age-group. These equate to over 55,000 problem drug users, and over 20,000 opiate users in Inner London as a whole.

Another more recent study’s estimates suggest that there were between 281,125 and 506,025 Class A problem drug users in England and Wales (15). Assuming a quarter of problem drug users live in London based on data on numbers entering treatment and drug seizures would give an estimate of approximately 70,000 problem drug users in London. More reliable indications of problem drug use in London should be available later this year.

Back to top ^^

Finally, details of the latest study are below:

A feasibility study undertaken by Frischer, Heatlie, and Hickman attempted to estimate the prevelance of problematic and injecting drug use for Drug Action Team Areas in England.

A Multiple Indicator Method (MIM) was used to produce prevalence estimates for problematic and injecting drug use for Drug Action Teams (DATs) in England in 2001. The MIM required drug indicators for all DATs and information on drug prevalence in selected DATs. For the purposes of the project, problematic drug use was defined as current use of illicit opiates, crack-cocaine or benzodiazepines. Injecting drug use covers all drug use where injecting was the method of drug administration.

Eight quantitative drug indicators were obtained from the DAT template (information collated by the Home Office) and other sources. [1: episodes of possession of controlled drugs, 2: episodes of supply of controlled drugs, 3: adults entering drug treatment services, 4: people arrested and referred for drug treatment, 5: drug abuse related deaths, 6: methadone prescriptions, 7: episodes of hospitalisation for drug abuse, 8: material deprivation measured by Townsend score].

Findings

From the model developed the estimated prevalence rate of problematic drug use in the 149 DATs varied from 0.2 per cent to 1.5 per cent of the population. The estimated rate of injecting drug use varied from nought to one per cent of population. Overall, the estimates for England in 2001 based on the model were:

Although this was a pilot project, it was large-scale in the sense that it provided estimates for all English DATs. The study’s authors suggested that DATs be given the opportunity to critically evaluate the estimates for their area and provide feedback. The feedback will help in the process of developing robust estimates in future years.

Further details can be found by clicking here.

Back to top ^^

This study by Hickman et al compared and developed existing indirect estimation techniques, here we report on the capture-recapture technique, in order to estimate the prevalence of injecting in three areas of England (Liverpool, Brighton and part of London). Estimates of the prevalence of problem opiate use also were generated for London (and Liverpool, and in London for problem drug and crack-cocaine use.

The estimation exercise involved collecting data about problem drug users in contact with specialist drug treatment, arrest referral, needle exchange, and Accident and Emergency departments, and conducting a community-recruited survey in the three areas.

Findings

The "best" estimates for London are summarised below and were obtained using covariate capture-recapture techniques - further details in main report.

Best estimates of the prevalence of injecting drug use using capture-recapture in London, including the 95 per cent confidence intervals are shown below.

Back to top ^^

Site	Total Population	Total Number of Injectors (95% CI)	Prevalence of Injecting Drug Use (95% CI)
12 London Boroughs*	1,361,267	16,782 (13,793 - 21,621)	1.20% (1.00% - 1.60%)
4 Outer London	476,411	2,099 (1,554 - 3,743)	0.40% (0.30% - 0.80%)
8 Inner London	884,856	14,684 (10,744 - 29,203)	1.70% (1.20% - 3.30%)

*Boroughs of Brent, Camden, City of Westminster, Ealing, Hammersmith and Fulham, Harrow, Hounslow, Islington, Kensington and Chelsea, Lambeth, Lewisham, Southwark

Overall, the study estimated that the prevalence of injecting drug use in those aged 15 to 44 years was 1.2 per cent in 12 London boroughs, and 1.7 per cent in the eight inner London boroughs.

The study’s authors note that prevalence is high and clearly a cause for concern, but state that the estimates are credible, consistent with each other and fit with the available evidence from some other public health indicators.

Table 2 shows the study’s current estimates for the number and prevalence of problem opiate users and problem drug use and crack or cocaine use in London.

Overall, the study estimated that the prevalence of problem opiate use was 2.1 per cent in London, and the prevalence of problem drug use and crack or cocaine use was 3.4 per cent and 0.8 per cent respectively.

The authors state that the estimates for crack/cocaine should be treated cautiously and perhaps as minimum estimates, but that, the estimates are still an important step forward in measuring the spread of crack cocaine use and the first for this substance in the United Kingdom

Back to top ^^

Site	Total Population	Total Number of Injectors (95% CI)	Prevalence of Injecting Drug Use (95% CI)
London (12 Boroughs *)
Opiate	1,361,267	28,979 (22,368 - 43,022)	2.1% (1.6% - 3.2%)
Crack/Cocaine	1,361,267	11,033 (10,176 - 12,074)	0.8% (0.7% - 0.9%)
Problem	1,361,267	146,156 (35,326 - 64,705)	3.4% (2.6% - 4.8%)

* Boroughs of Brent, Camden, City of Westminster, Ealing, Hammersmith and Fulham, Harrow, Hounslow, Islington, Kensington and Chelsea, Lambeth, Lewisham, Southwark

Further details can be found by clicking here.

Back To Top

More up-to-date estimates of the prevalence of problem drug use, injecting drug use and heroin use will be available shortly, through the University of Glasgow.

Back to top ^^