Illicit drug consumption is a worldwide problem that is difficult to accurately measure. In recent years, quantifying excreted drug metabolites in raw wastewater (wastewater analysis) has become a complementary tool to epidemiological methods for estimating consumption of illicit drugs within a population. The overall objective of this PhD research was to apply wastewater analysis to measure illicit drug use in selected Australian communities in South East Queensland and an Asian metropolitan city. Specifically, this project aimed to (a) refine the back calculation methodology for more reliable estimation; (b) compile data on different scales of temporal and spatial resolution; and (c) investigate characteristics of consumption during the Australian peak holiday season and in an annual music festival.
All the samples in this research were collected at the inlets of wastewater treatment plants. Representative composite samples were collected by following best practice sampling techniques: continuous flow-proportion, discrete volume-proportion and high-frequency time-proportion sampling. Samples were preserved at pH 2 onsite and frozen until analysis. The analytical method applied the use of isotope-labelled compounds as surrogates to compensate for errors and variations from sample preparation and analysis. Targeted drug residues in the samples were analysed by liquid chromatography coupled with tandem mass spectrometry.
For refinement of the back calculation methodology, this study provided a demonstration of systematically identifying uncertainty components, quantifying the uncertainties (relative standard deviation) and reducing the overall uncertainties through normalising to the daily estimated populations derived from selected highly used pharmaceuticals in the wastewater (Chapter 2). This normalisation approach can circumvent uncertainties associated with flow measurement because the population sizes estimated from the selected pharmaceuticals and the estimated consumed masses of illicit drugs are both based on the same flow data. With the best sampling technique, the method uncertainty for estimating consumed masses of illicit drugs was approximately 20–30%. Normalising to the daily estimated population derived from one prescription pharmaceutical (atenolol) reduced the method uncertainty for per capita consumption to approximately 14–24%. This normalisation approach was further refined by a multiple-substance model using Bayesian inference in that ten pharmaceuticals were used to estimate population changes in the samples, which further reduced the method uncertainty to approximately 10–15% (Chapter 7).
The developed methodology was then applied to compare illicit drug use in an urban community in November 2009 and 2010 during the celebratory period for school leavers (Chapter 3). The results from this temporal study revealed a change in illicit drug patterns: methamphetamine > cocaine > MDMA in 2009 vs. methamphetamine > MDMA > cocaine in 2010. This study suggests that the methamphetamine market is larger than the cocaine and MDMA markets. Preliminary results from ongoing monitoring in this catchment show a temporal change (generally increasing) in consumption of these three illicit drugs from February 2011 to June 2012 (Chapter 7). Daily trends showed higher consumption during weekends than weekdays (Chapter 3 & Chapter 7).
Another application evaluated spatial differences in illicit drug use (types and amounts) among urban, semi-rural and vacation communities in southeast Queensland (Chapter 4). Cannabis and methamphetamine were the most commonly consumed drugs in all these communities. Usage of these two drugs was generally stable and higher in both urban and semi-rural communities than that in the vacation community. Cocaine and MDMA consumption were more abundant in both urban and vacation communities than in the semi-rural community. The resultant drug use profiles partially agreed with current national drug availability data. Also, specific holiday effects on illicit drug use were investigated in these three communities (Chapter 4). Compared to the control period, distinctive changes in drug use profiles during the holiday period were observed for the urban and vacation communities but not necessarily for the semi-rural community. During the holidays, cocaine and MDMA increased in all communities whereas cannabis and methamphetamine only increased in the urban and vacation communities. New Year celebrations were peak consumption events during the monitoring period. The results suggest drug consumption (types and amounts) is influenced differently by the specific holiday across the studied communities.
This PhD research also applied wastewater analysis to examining illicit drug use at an annual music festival (Chapter 5). The drug consumption pattern at the music festival was cannabis > MDMA > methamphetamine > cocaine. Only MDMA consumption was greater at the festival than in the nearby urban community, confirming its association with music events. Emerging illicit drugs (benzyl piperazine, methylone and mephedrone) were also detected but in fewer samples and in much lower amounts than the other four conventional illicit drugs. Their use generally decreased over the two festival years.
Finally, this research study applied wastewater analysis to an Asian metropolitan city (Hong Kong) (Chapter 6). The detected patterns of illicit drug use (ketamine > methamphetamine > cocaine > MDMA) were consistent with findings from Hong Kong’s drug reporting system. Diurnal data (every two hours) revealed higher mass loads of illicit drug residues in the mornings but on only occasional evenings. Cocaine consumption in Hong Kong was higher than in the Australian communities, while methamphetamine consumption resembled the Australian urban and semi-rural communities.
This thesis demonstrates that wastewater analysis can readily capture illicit drug use in a variety of communities. The usage patterns detected by wastewater analysis broadly agreed with epidemiological findings and expectations. Both wastewater analysis and existing epidemiology methods can be complementary to each other to obtain more comprehensive insight into trends and patterns of illicit drug use within and between populations.