Commentary on Ort et al. (2014): What next to deliver on the promise of large scale sewage-based drug epidemiology?

Abstract

Ort and colleagues 1 present an impressive demonstration of the power of wastewater to provide systematic information about the extent of drugs consumed across 21 countries, resulting from the Sewage Analysis CORe group Europe (SCORE) collaboration. In a rapidly advancing field, significant effort has been placed on reducing uncertainties in the data set through painstaking attention to understanding sewer characteristics and ensuring international consistency of laboratory techniques. Clearly, systems such as this provide an unprecedented insight into the full extent of substance consumption in the population as a whole. A number of interesting challenges lie ahead to realize the full potential of large-scale wastewater analyses to inform those reading the data across the spectrum of health, law enforcement and policy perspectives. It would be beneficial to integrate drug purity data into the existing frameworks to facilitate comparisons between regions and trends over time. As the authors note, it is difficult to interpret differences (between sites or over time) in the existing system, as an increase in mass load may reflect any combination of the following scenarios: an increase in the number of people consuming a drug of consistent purity; a stable number of people consuming a drug with increased purity; or a stable number of people consuming a greater amount of a drug of consistent purity. There is clear evidence of substantial differences in drug purity levels across Europe 2, and evidence that consumers do not necessarily respond to low purity by purchasing larger quantities, or vice-versa (in fact, the opposite typically applies 3, 4). As such, integration of purity data by, for example, conducting a purity audit of all drugs seized in the catchment area during the week-long assessment period would dramatically enhance how informative these data are to understand the consumer side of the market. However, while this would be analytically trivial, such integration may be administratively difficult because it requires secure transport and storage processes of illegal substances to contributing SCORE laboratories, or rapid analysis of all law enforcement seizures and careful quality assurance across forensic laboratories. Systematic continuous wastewater samples have demonstrated that substance use tends to fluctuate over time, with greater levels of use in warmer months 5 and with particular events 6. There are two implications of this for the SCORE system. First, while using a single-week census across sites is appropriate, it is also important for the study to be conducted at a consistent time every year to minimize the impact of this variability. Secondly, bi-annual or quarterly assessments would enhance the information available from the system, particularly in relation to strategic early warning of substantial change in particular regions. The wastewater analysis field has made dramatic advances in identification and reduction of errors due to uncertainties arising from sampling, flow measurement, analysis and in-sewer biotransformation. Two key components still require dedicated research attention for wastewater analysis to fully deliver on its potential. Population values are the denominator in the final estimation of drug load in wastewater. Populations contributing to catchments can increase dramatically over the resident population with the daily influx of commuting workers or visitors into capital cities. Substantial advances may be made in improving population estimates via co-examination of pharmaceuticals or other biomarkers 7, 8, but these may require advanced analytical capacity or secondary data for application. Emerging indirect approaches, such as using mobile phone signal load 9, may also offer scope to improve population estimates. In order to back-estimate the quantity of a drug consumed, the load of the drug identified through wastewater analysis is multiplied by the molar fraction of the parent drug (or metabolite) that is excreted by consumers. These data are based on a surprisingly scant database of human pharmacological studies 10, 11. For example, a recent meta-analysis used to provide excretion fractions for 3,4-methylenedioxy-N-methylamphetamine (MDMA) 10 is based on just four studies, involving a total of 32 individuals, all of whom were male, with 25% of these MDMA-naive, and just one of Asian descent 12-15. The oral dose range studied varied between 40 and 150 mg, and was suggestive of non-linear pharmacokinetics 12. It is difficult to generalize these estimates to typical real-world situations where both males and females may be ingesting MDMA in stacked (multiple) doses, potentially beyond this dose range, via different routes, or in conjunction with multiple other substances—all of which may affect the metabolism profile of the drug and hence the accuracy of excretion fractions applied. This is an area that addiction scientists and pharmacologists have substantial scope to contribute to addressing. Such large-scale wastewater studies such as SCORE are potentially extremely powerful tools for understanding changing drug markets. Refining the processes underpinning such studies and enhanced collaborations with addiction scientists and epidemiologists will enable these methods to maximize their potential to inform drug policy. None.