EU Drug Markets: Overview of data and methods

The analysis presented in EU Drug Markets: In-depth analysis is based primarily on data and information reported to the EMCDDA and Europol. Global data are sourced from the UNODC and the INCB.

EMCDDA statistics are based on core monitoring and non-routine data collection. Core monitoring data are derived from annual data-collection exercises, using standardised reporting tools, conducted by the EMCDDA via its network of national focal points across the 27 EU Member States, Norway and Turkey. These include drug seizures, drug prices and purity, drug law offences, production, and qualitative descriptions of national trends and developments. Composite indicators are also being used, including market size estimates and drug affordability. In instances where significant national data gaps exist, this is noted in the analysis.

Estimates of the illicit drug market in the European Union in 2020 have been constructed using a demand-side approach, based on a methodology established by the EMCDDA (2019). These estimates are likely to underestimate the size of the market due to misreporting and under-reporting of prevalence and use (Udrisard et al., 2022). One way to compensate for this is to apply correction factors. Accordingly, the analysis and discussion presented here explore correction factors and their impact on the initial estimates. In addition, alternative scenarios are examined – utilising national estimates for a particular drug; supply-side strategies or novel methods to estimate drug market size, such as wastewater-based analysis.

Affordability of drugs is a composite measure that incorporates drug purity and accounts for differing national economic conditions as quantified in the price level indices (PLI) (for fuller detail and limitations, see Groshkova et al., 2018).

In addition to the established statistical data, open-source information was utilised as well as data on individual significant drug seizures with relevance for the EU market, provided to the EMCDDA by national authorities. Also utilised were international databases, key periodical analysis and ad hoc research.

The analyses presented here are also informed by new, more innovative monitoring methods. These include wastewater analysis, syringe residue analysis, web survey data and monitoring of darknet markets. These novel methods have their own strengths and limitations, some of which are discussed below.

Wastewater analysis

Wastewater-based drug epidemiology involves sampling a source of wastewater and allows the estimation of the quantity of drugs consumed by a community (Zuccato et al., 2008). While this approach offers a valuable complementary data source for monitoring the quantities of drugs used at population level, it does not provide information on prevalence and frequency of use, main classes of users and purity of the drugs. This may be further complicated by the fact that large cities often have several wastewater treatment plants and not all of them are usually covered in the wastewater sampling.

Additional challenges arise from uncertainties associated with the behaviour of the selected biomarkers in the sewers, different back-calculation methods and different approaches to estimate the size of the population being tested (Castiglioni et al., 2013; EMCDDA, 2016). While wastewater monitoring has its limitations, efforts are being made to enhance this approach. For example, work has been undertaken on overcoming uncertainty related to estimating the number of people present in a sewer catchment at the time of sample collection (Thomas et al., 2017).

Syringe residue analysis

Data on drug residues in used syringes can be a valuable indicator of the range of substances used by people who inject drugs in a certain area, and can allow the monitoring of changes in patterns of use over time. However, there are some challenges and limitations related to the use of this data source, including issues related to representativeness, syringe re-use, sample validity and integrity.

For example, the number of syringes collected and tested cannot be translated into a number of individual users, since a small number of users could have contributed a disproportionately large number of syringes (for example, users who may return syringes to collection facilities in bulk). The method therefore does not measure prevalence of injecting nor does it necessarily provide the relative prevalence of use for different substances among people who inject drugs.

Drugs in syringes may also degrade over time and become undetectable. The time lag between injection and collection of syringes can therefore have an impact on the analysis of residual substances. Furthermore, drugs found in syringes may originate from blood drawn into the syringe during an injection, that is, from drugs consumed prior to the injection and possibly through other modes of administration. It is also not possible to distinguish a syringe containing residues of multiple drugs that had been used once from a syringe that had been reused by one or several users.

Web survey data collection

Web-based surveys can be a useful tool for gathering information efficiently from large number of drug users to complement our understanding of patterns of use, including drug-purchasing behaviours, particularly among recreational users. However, there are some notable limitations in the use of web surveys. Some important limitations revolve around sampling and coverage biases, the extent and nature of which can be difficult to ascertain since the characteristics of those not responding to the survey, and how they differ from survey respondents, are not known. Web survey results are not readily generalisable and representable across the population, as no sampling frame exists. Consequently, these surveys cannot directly be used to estimate the prevalence of drug use. Moreover, some studies indicate that web survey respondents can differ significantly from those recruited via probabilistic sampling methods. In relation to web surveys for drug data collection, some evidence indicates that respondents are generally younger, male and more frequent users.

While web surveys are not representative of the general population, some statistical methods have shown promise in making the results of web surveys more generalisable. For example, some research has been undertaken on combining the results of web surveys with those of general population surveys by using matching procedures, raking and regression. Overall, when carefully conducted and combined with traditional data-collection methods, web surveys can help paint a more detailed picture of drug use and drug markets in Europe.

The main web survey used by the EMCDDA is the European Web Survey on Drugs (EWSD), which since 2016 has supported the EMCDDA and national decision-makers in gathering in-depth data on the drug situation in Europe. The latest wave of the survey collected data between March and April 2021, involving 30 countries (21 of which are EU Member States) and translated into 27 different languages. To be eligible for participation, respondents had to be 18 years or older, reside in the country where the survey was undertaken, and have used at least one of the drugs covered in the survey (cannabis, cocaine, ecstasy/MDMA, amphetamine, methamphetamine, heroin or new psychoactive substances).

Darknet markets monitoring

Data on drug sales on darknet markets provide insights into the extent of online drug trade and improves our understanding of the related health and security threats. This monitoring relies on ‘web crawlers’ that systematically gather and process information from darknet markets. The information collected includes drug listings, quantity, price and buyers’ feedback reports as a proxy for sales activity. While this is a valuable data source, it covers a volatile environment, which presents challenges to interpreting trends over time. Other online platforms are also relevant, though gathering data from these is harder to achieve. Darknet markets data are likely to represent only a small fraction of online drug sales.

As such, when interpreting findings from darknet monitoring, it is important to recognise the methodological and practical difficulties of conducting research on darknet markets. Studies may consider only a subset of markets or geographical locations, and it is difficult to ensure that a ‘scrape’ of an online darknet market is complete; thus, it cannot be assumed that the results are necessarily representative. In addition, these markets are dynamic and change occurs rapidly. Despite these limitations, the available data are informative and also highlight areas requiring further investigation and monitoring.

References

Castiglioni, S., Bijlsma, L., Covaci, A., Emke, E., Hernández, F., Reid, M., et al. (2013), ‘Evaluation of uncertainties associated with the determination of community drug use through the measurement of sewage drug biomarkers’, Environmental Science and Technology 47(3), pp. 1452-1460, doi:10.1021/es302722f.

EMCDDA (2016), Assessing illicit drugs in wastewater: advances in wastewater-based drug epidemiology, EMCDDA Insights 22, Publications Office of the European Union, Luxembourg.

EMCDDA (2019), Estimating the size of the main illicit retail drug markets in Europe: an update, Technical report, Publications Office of the European Union, Luxembourg.

Groshkova, T., Cunningham, A., Royuela, L., Singleton, N., Saggers, T. and Sedefov, R. (2018), ‘Drug affordability: potential tool for comparing illicit drug markets’, International Journal of Drug Policy 56, pp. 187-196.

Thomas, K. V., Amador, A., Baz-Lomba, J. A. and Reid, M. (2017), ‘Use of mobile device data to better estimate dynamic population size for wastewater-based epidemiology’, Environmental Science and Technology 51(19), pp. 11363-11370.

Udrisard, R., Esseiva, P. and Zobel, F. (2022), Improving the estimation of the size of the European drug market, report commissioned by the EMCDDA, contract CT.20.SDI.0139.1.0.

Zuccato, E., Chiabrando, C., Castiglioni, S., Bagnati, R. and Fanelli, R. (2008), ‘Estimating community drug abuse by wastewater analysis’, Environmental Health Perspectives 116(8), pp. 1027-1032.

Loading