Portal de Periódicos da CAPES

Peering into dirty waters: the potential and implications of a new approach to monitoring drug consumption

2008; Wiley; Volume: 103; Issue: 8 Linguagem: Inglês

10.1111/j.1360-0443.2008.02204.x

1360-0443

Norbert Frost, Paul Griffiths, Roberto Fanelli,

Analytical chemistry methods development

New things have a tendency to generate a polarized response, especially when they represent a radically different approach to a problem. The advantages can be overstated, especially when the emerging science attracts the interest of the media, and exaggerated claims will be dismissed, often with little understating of the basis from which they where originally derived. This risk is highest when the topic touches on issues of political or social sensitivity and can impede a clear-headed assessment of the advantages and implications of the developing science. All this is true of new methods to test for drug residues in wastewaters. Recent developments in this area have caused concern at high levels, not only in Europe but also in the United States and at the United Nations; generated media reports that have questioned existing estimates of drug use levels; have been dismissed out of hand by some experts; and raised privacy concerns. We are thus in murky waters—but as the dirt slowly settles it is becoming clear that new developments in our ability to detect drugs and their metabolites in wastewater are likely to have important implications for the approaches we adopt to monitoring drug consumption trends over time and the debate on what kind of surveillance approaches are ethically and politically acceptable. This approach has been made possible by recent improvements that have increased the sensitivity of analytical laboratory equipment. Developments in chromatographic and mass-spectrometric analysis have made it possible to identify urinary excretion of illicit drugs and their main metabolites in wastewater at very low concentrations, down to nanogram per litre ranges 1. The initial focus of work in this area was on the monitoring of environmental contamination caused by prescribed drugs and the related question of the effectiveness of water treatment plants 2-4. An example of this can be found in the work carried out by Ternes and coworkers in the 1990s 2, 3, who investigated the extent to which therapeutic drugs were detectable in rivers and wastewater systems in Germany. The main concern for this type of environmental monitoring is the extent to which a known level of drug prescription is likely to produce residues that are excreted into the environment. A number of researchers realized quickly that by standing this question on its head it might be possible to assess levels of illicit drug use by detecting their residues in the environment 1, 4-6. In the last few years, research teams have demonstrated the ability to detect drug residues in wastewater 1-6. In 2005, researchers at the Milan Mario Negri Institute in Italy reported that they had detected illicit drug residues in water taken from the Po river basin 6. Furthermore, by surveying water daily over 3 non-consecutive weeks at a large wastewater treatment plant in Milan, the same group began to explore how this technique might be used to detect changes in drug consumption patterns over time 1, 6. Testing for a range of drugs and their metabolites, a statistically significant peak for the main cocaine metabolite, benzoylecgonine, was noted at the weekend, whereas morphine and THC-COOH levels remained relatively constant throughout the week. Interesting as these findings are, if this approach is to have value as an epidemiological tool for the analysis of patterns of illicit drug use a number of issues need to be addressed. The potential and probable limitations of this approach were the focus of discussion of a multi-disciplinary meeting of experts held in April 2007 in Lisbon 7. One of the conclusions of this group was that analysing drug residues in river water is likely to be of limited value for drug epidemiology purposes. Illicit drug monitoring requires sampling based on a known population and this requires what can be described as closed water systems; for example, wastewater entering a sewage plant that serves a known population. Such closed water systems provide the possibility of calculating population-level consumption estimates but do not provide a direct measure of individual prevalence. Among the main conclusions from this meeting was that the evidence now supports the view that within a closed water system, when factors such as a drug's pharmacokinetics and metabolism and the environmental fate of excretion products are taken into account, the environmental load (i.e. the amounts entering a wastewater plant over time) of a drug and/or its major metabolites can be considered an indicator of the drug's consumption by the local population. The need to focus upon closed water systems also highlights the potential ethical and privacy issues that this type of surveillance approach raises. The unit of analysis could be, for example, the wastewater exiting a particular building, a university campus, military establishment, local neighbourhood or even a parliamentary chamber. The approach requires no individual consent and may be viewed by some as representing an unacceptable form of intrusion. In some respects, the novel nature of the technique may make it particularly threatening—and concern has been expressed that people might view this as the government spying on them through their lavatories. Leaving such political and ethical concerns aside, a number of technical questions remain to be resolved if this approach is to be developed further. At first sight these appear formidable, but as research continues none appears insurmountably so. Understanding is growing of the extent to which wastewater systems can be considered to represent truly closed systems, and what factors can affect this—such as storm surges and leakages. The impact of environmental degradation and contamination are also being explored 5, 8. Consensus is growing on what approaches are likely to be most fruitful, as illustrated by a move towards the selection of local sewage intake pipes as the potentially most appropriate sites for sample collection. Although this technique may have potential for many types of drugs, to date it has been used most commonly for exploring cocaine consumption levels. This is because the metabolite benzoylecgonine is formed only after human or animal consumption of cocaine, and the possible uncertainty that results may be influenced by the dumping of unconsumed cocaine in the environment is ruled out. Possible consumption of the drug in research or for pharmacological development purposes cannot be dismissed so easily. If a wider array of substances is to be studied, other sources that may lead to drugs or their metabolites being present in the environment need to be considered. As information on human excretion rates for different substances is important for calculating the original amount of drugs consumed, knowledge of typical values and the factors that may affect these is important. Some data are available here to act as a guide for estimates, but the evidence base is currently far from comprehensive. Similarly, although different routes of administration will produce different blood levels and different peak times and excretion patterns, how this impacts on the ratio of drugs consumed to drugs excreted is poorly understood. These factors also complicate any attempt to move from population consumption estimates to the epidemiologically more important question of prevalence estimation. However, here the more fundamental and obvious problem of modelling patterns and frequencies of use arises. This is complicated further by the fact that closed water systems may serve transient human populations, an issue important for estimating drug prevalence where use may be concentrated in specific settings. Where, then, does this leave us? Wastewater monitoring has potential but, like other approaches in this area, presents us with a number of methodological and practical challenges to overcome. It increases the importance of developing a better understanding of some basic science questions, particularly those concerning the excretion of illicit substances. Studies in this area will need to pay particular attention to some complex and somewhat novel sampling questions—but then again, when has sampling not been a, or even the, critical issue for the study of illicit drug use? We can conclude, with some confidence, that this technique will not replace more conventional approaches to the estimation of drug prevalence. That said, existing techniques in this area are not free of methodological problems, and having an additional information source to build into the model will be beneficial. Moreover, standardized wastewater analysis in an advanced stage has the advantage that repeat real-time estimates can be made both cheaply and quickly in comparison to other approaches and therefore may open up some new possibilities for the monitoring of trends over time. From an ethical and privacy perspective, some of the more difficult issues arise if this new approach is applied for purposes other than scientific inquiry. From a purely research perspective, ensuring the anonymity of individuals applies equally to any study that generates collective data. The issue of informed consent is more problematic but arguably applies equally to the monitoring of waters for environmental purposes. Again, the general conclusion must be that wastewater studies, like other research endeavours, require appropriate ethical scrutiny. In conclusion, we are in the early days of the development of a new approach that appears to have considerable potential and may open up some new possibilities for the way drug consumption trends are studied. It is a mistake, in our view, to see wastewater analysis as an alternative approach to drug monitoring; rather, what is emerging is an additional tool that both complements and extends the possibilities of existing approaches to the fundamentally difficult problem of monitoring a complex, hidden and stigmatized behaviour. While the views expressed here are those of the authors, we would like to acknowledge that the catalyst for this editorial was the discussions that occurred recently in the first multi-disciplinary meeting on wastewater analysis attended by: Damià Barceló, Instituto Investigaciones Químicas y Ambientales, Barcelona, Spain; Fátima Pina, Faculdade de Medicina da Universidade do Porto, Portugal; Fritz Sörgel, Institut für Biomedizinische und Pharmazeutische Forschung, Germany; Joerg Rickermann, Department of Geography, San Diego State University, San Diego, CA, USA; Julian Vicente, EMCDDA, Lisbon; Linda Montanari, EMCDDA, Lisbon; Lucas Wiessing, EMCDDA, Lisbon; Matthew Hickman, Department of Social Medicine, University of Bristol, United Kingdom; Norbert Frost, EMCDDA, Lisbon; Paul Griffiths, EMCDDA, Lisbon; Roberto Fanelli, Istituto di Ricerche Farmacologiche Istituto Mario Negri, Italy; Stefano Salvadori, Consiglio Nazionale delle Ricerche, Italy. We would also like to thank Peter Fay and Shane Darke for their helpful comments during the preparation of this manuscript.