Table Prevalence-4. Prevalence of problem drug use (rate per 1000 aged 15-64) estimated by various methods, from 1995 to 2002. Part (ii) Prevalence of problem drug use assumed to be injecting use

Country	Year		Population size age 15-64 ^(b)	Multiplier mortality ^(a) data		Multiplier HIV-AIDS ^(a)		Capture-recapture		Multivariate indicator		Other

Austria	2000		5483440	2.19-4.19

Belgium	1995		6707199			2.92	⁽¹⁾
	1997		6705928			3.46-4.24	⁽²¹⁾

Czech Republic	n.a.

Denmark	1996		3535829	3.50-5.22	⁽³⁶⁾

Finland ⁽³⁾	1997		3420598	1.17-2.48

France	1995		37736396			3.74-4.69	⁽¹⁾
	1999		38131287			3.2

Germany	1996		55846564	1.43-2.01
	2000		55915209	2.27-3.03		1.61-2.83	⁽²⁴⁾

Greece	n.a.

Ireland	1995 / 96		2342670	2.0-3.37

Italy	1996		39099069			8.34

Luxembourg	1999		287096	7.27-7.51 [7.39]		6.2
	2000		291010			5.89

Netherlands	n.a.

Norway	1997		2835512	2.54-3.63	⁽¹²⁾
	1998		2852913	3.15-4.56
	2000		2900748	3.62-4.83	⁽¹⁴⁾
	2002		2968000	3.71-5.10	⁽²⁹⁾

Poland	n.a.

Portugal	1999		6778900	2.72-5.44		3.34-4.96
	2000		6920368	2.30-4.60	⁽⁹⁾	4.28-6.35	⁽⁹⁾

Slovenia	n.a.

Spain	1998	⁽¹¹⁾	26923202	3.12

Sweden	n.a.

United Kingdom	1996	⁽¹⁶⁾	37007584	4.35		4.36
	2001		38593141									3.07-3.34 (3.20)	⁽³⁵⁾
U.K. (Scotland)	2000		3352022					5.10-13.70 (7.30)	⁽³¹⁾
U.K. (N. Ireland)	n.a.
U.K. (England + Wales)	2001		32292156							2.89	⁽³³⁾

Notes:
The individual reference numbers shown against the studies relate to detailed comments on method or scope of the study given below should be consulted when interpreting the data presented in the tables.
Rates are represented with three significant digits in order to preserve precision of the original estimates which mostly have been provided with three or more significant digits. However, the precision of the estimates is best represented by the intervals given and where no interval has been provided precision is unknown.
In several cases [see project report CT.00.RTX.23, Table 4] original estimates were based on age range 15-54 and the rates have been recalculated using total population age 15-64 as the denominator - this assumes that the proportion of drug users aged 55-64 is very small and can be neglected.
Midpoints have been calculated in cases where no central estimate was provided to facilitate interpretation and comparisons.

(a) Mortality and HIV/AIDS multipliers are assumed to give estimates of injecting drug use, wich is only part of the problem drug use. The police and treatment multiplier methods are assumed to result in estimates to problem drug use (the EMCDDA definition of problem drug is "injecting drug use or long-duration/regular use of opiates, cocaine and/or amphetamines").
(b) Population sizes taken from Eurostat NewChronos database February'2003, except for Greece and UK (2001 data not available) [http://intracomm.cec.eu-admin.net/newcronos/].
(1) Revised estimate.
(2) Problematic drug users.
(3) Problem opiate and amphetamine users.
(4) Estimates based on 3 and 4 registers.
(6) Demographic multiplier method, results are less reliable and not included in overall country range.
(9) Combined interval from multiple capture-recapture and truncated Poisson estimates.
(10) ‘High-risk consumption problem drug users’.
(11) Problematic opiate users.
(12) Mean of the years 1995,1996,1997 number of drug related deaths used to avoid the effect of random variations and outliers.
(13) Method based on survey within police and public health services.
(14) Opiate users. For the number of drug related deaths the mean of the years 1998,1999,2000 was used to avoid the effect of random variations and outliers.
(15) Mostly amphetamine users. Estimates according to EMCDDA definition of problem drug use. 1992: figures include an estimated 5800-6950 (point estimate 6450) opiate users (opiates as primary or secondary drug). 1998: country report not available, figures include an estimated 11500-13400 (12200) opiate users. Data provided by Borje Olson and based on: B. Olsson, C.A. Wahren, S. Byqvist, Det tunga narkotikamissbrukets, omfattning i Sverige 1998, CAN, Stockholm 2001.
(16) Country report not available, data taken from Frischer et al 2001 (Frischer M, Hickman M, Kraus L, Mariani M, Wiessing L. A comparison of different methods for estimating the prevalence of problematic drug misuse in Great Britain. Addiction 2001; 96: 1465-1476).
(17) Method not based on HIV but on use of syringes and substitution substance.
(18) 6308: estimate for “problematic opiate users with medical problems”, 13735: estimate for “opiate users with both medical and social/legal problems” . The higher figure only is compatible with the EMCDDA definition of problem drug use and is used for country comparisons.
(19) Truncated Poisson model, combined results of Chao and Zelterman estimators.
(20) Problematic drug users using as data source treatment admissions data. Multiplier method for opiates and demographic method (low and high bound) for cocaine. Multiplier data for opiates has been corrected for admissions reporting both opiate and cocaine use.
(21) The lower estimate is based on extrapolation from self-reported prevalence of HIV among IDUs and the higher on serological based prevalence.
(22) Problem opiate users. Two-sample capture-recapture between police and substitution treatment sample, potential interactions cannot be accounted for. Possibly an overestimation due to systematic bias in the registries.
(23) Problem amphetamine or opiate users. 1999/00 and 2003 estimates: Multiplier method, using as benchmark the Treatment Demand register and for the in-treatment rate capture-recapture and nomination techniques at the local level. 2002 estimate: Multiplier method, benchmark - number of pervitin (methamphetamine) and opiate users in contact with low-threshold facilities in 2002. Annual Report on Drug Situation 2002 – Czech Republic.
(24) Kraus L, Augustin R, Frischer M, Kümmler P, Uhl A, Wiessing L. Estimating prevalence of problem drug use at national level in countries of the European Union and Norway. Addiction 2003; 98: 471-85.
(25) Drug abusers having a persistent use of illegal drugs, including cannabis, which leads to physical, psychological and social consequences. Estimate for 2001 is estimated to contain 6000 problem users of cannabis only. 2-sample capture-recapture using the National Registry of Patients and the National Register of Drug Abusers Undergoing Treatment. The estimate for 1996 is not directly comparable with the estimates for 1998 and 2001. For 1996 the National Register of Drug Abusers undergoing Treatment is not as complete as it is for the following years. Reference: Hammerby E. Nye Tal fra Sundhedsstyrelsen nr. 16 2003.
(26) Problem heroin users. 3-sample capture-recapture using 3 different treatment sources. Estimate cannot be compared with 2001 due to different methodology.
(27) Problem opiate users. 3-sample capture-recapture using 1) Clients on the Central Methadone Treatment List in 2000, 2) Hospital discharges in 2000 with ICD 9 codes 304.0, 304.7 and 305.5, 3) Police database, individuals known by the police to be opiate users in 2000. Reference: Kelly A, Carvalho M, Teljeur C. A 3-Source Capture Recapture Study of the Prevalence of Opiate Use in Ireland 2000-2001. Key Findings Summary Tables. Dublin: National Advisory Committee on Drugs, 2003. Report available online at www.nacd.ie.
(28) Problem drug users. Police multiplier based on: Ministry of the Interior data on arrests for drug offences. Treatment multiplier and capture-recapture based on: Ministry of Health data on IDUs attending public treatment services (Ser.T). Multivariate Indicator Method based on: Ser.T., Ministry of the Interior data on arrests for drug offences and HIV/AIDS register data. All intervals are 95% Poisson C.I. based on provincial area breakdown estimates. Reference: Epidemiological Section of the Italian Observatory on Drugs and Drug Addiction, 2003. (unpublished data). Injecting drug users 2002: no details available, source NFP.
(29) Opiate users. Reference: Bretteville-Jensen, A.L., personal communication, 2002.
(30) Problem drug users. Two-sample capture-recapture based on police arrests and drug treatment data. These are preliminary results that need further checking of validity and reliability.
(31) Current injectors (24696) or problematic opiate and benzodiazepine users (55800). Aggregate national estimate from multiple local capture-recapture estimates using data from treatment agencies, general practitioners, social enquiry reports (probation), HCV postive drug injectors. Local estimates are available for 92% of Scotland. These estimates have been scaled up to give a 'Scotland' estimate. Reference: CDMR (University of Glasgow) & SCIEH 'Estimating the National and Local Prevalence of Problem Drug Misuse in Scotland'.
(32) Problem opiate users. Capture-recapture using data from treatment agencies, addicts index/hospital inpatient data, arrests for heroin and acquisitive offences. Reference: McElrath, K. (2002) Prevalence of Problem Drug Use in Northern Ireland. Belfast, Department of Health, Social Services and Public Safety.
(33) Current injecting (93185) or current use of illicit opiates, crack-cocaine or benzodiazepines (287670). Multivariate indicator method using possession of controlled drugs, supply of controlled drugs offences in year, treatment for drug use, arrest referrals, drug related deaths, methadone prescriptions, hospital admissions, deprivation measure (8 indicators). Reference: M Frischer, H Heatlie and M Hickman. Estimating the prevalence of problematic and injecting drug use for drug action team areas in England: A feasibility study using the multiple indicator method. London: Home Office Drug and Alcohol Research Unit.
(34) Persons dependent on opiates and/or crack cocaine, in contact or potentially in contact with police or health care services for drug-related problems. Reference 1999: Smit F, Toet J, van Oers H, Wiessing L. Estimating Local and National Problem Drug Use Prevalence from Demographics. Addiction Research and Theory 2003; 11: 401-413. Reference 2001: Smit F et al. (in press).
(35) Weighted average of estimates for England, Scotland, Wales and Northern Ireland derived by different methods and assuming that the PDU and IDU prevalence rates of England apply to Wales and that the ratio PDU/IDU for England applies to Northern Ireland. Source: national focal point.
(36) The assessment of the number of problematic drug users (hidden population) was done with the use of the ‘benchmark’ method within the framework of a countrywide population survey of 2002. Three estimations were conducted on the basis of the following registers: residential drug treatment, out-patient drug treatment and HIV cases.

Sources:
National focal points through EMCDDA project: National prevalence estimates of problem drug use in the European Union, 1995-2000, CT.00.RTX.23, Lisbon, EMCDDA, 2003. In addition where specific reports have been published these are also given.

Download