Drug use can negatively affect future health and wellbeing, including mental health, family relationships, educational achievement and employment opportunities.1 The social cost of drug-related harm in New Zealand (excluding alcohol) was recently estimated to be $1.5 billion per year, with the government spending $78.5 million on drug-related health interventions each year.2 Monitoring levels of drug use is important to guide the delivery of health services and inform policy responses to drug-related harms. Levels of drug use in New Zealand are commonly compared to those in Australia, as the two countries are from the same global region and share similar socio-economic characteristics.3 Traditionally the monitoring of drug use in the population has relied heavily on national social surveys where respondents self-report drug use. While these social surveys provide fairly good measures of drug prevalence in the population they are known to suffer from a number of limitations, including sample under-coverage depending on the means used to contact potential respondents (eg, people without landline telephones or those seldom at home), declining response rates due to high levels of commercial market surveying, and the under-reporting of drug use due to illegality and social stigma.4–6 National social surveys of drug use are also expensive and time-consuming to conduct and analyse.5 A common strategy to address the challenges of monitoring illegal drug use is to triangulate from a range of data sources, including health statistics (eg, drug-related hospital admissions and poisonings), police statistics (eg, drug seizures and arrests) and by conducting targeted studies of ‘at risk’ populations with high drug use and related harm (eg, police arrestees and injecting drug users).7,8 Wastewater-based epidemiology (WBE) is an emerging new methodology which can provide objective measures of drug consumption based on the detection of drug residues in pooled wastewater (ie, sewage) sampled at the inlet pipe of a wastewater treatment plant (WWTP).9–12 Sampling from a WWTP ensures all the dwellings in the WWTP’s catchment are automatically covered by the estimates; thereby avoiding issues of under-coverage and under-reporting. WBE also minimises privacy issues related to drug surveying, as pooled wastewater guarantees individual anonymity.5 Numerous WBE studies have been conducted in cities in Europe, North America, Asia and Australia in recent years.13–17 Most recently, WBE has been applied to identify spatial variation in drug use across all Australian states, and to identify temporal changes in methamphetamine consumption from 2009 to 2016.18–20 WBE is beginning to be utilised in New Zealand with a pilot study completed in Auckland in 2014,21 and a larger pilot programme commissioned by New Zealand Police and conducted by the ESR in 2017.22 However, no comparisons of WBE findings have yet to be made between New Zealand and Australian.

The aim of this paper is therefore to compare levels of methamphetamine, cocaine, MDMA, codeine and methadone measured using WBE from two urban WWTPs in the Auckland Region with eight urban WWTPs from four Australian states (ie, Queensland, New South Wales, Australian Capital Territory [ACT] and Victoria).

Methods

For the purposes of comparisons, a week of wastewater sampling from 10 urban WWTPs was selected: two in the Auckland region; one from Australian Capital Territory (ACT); two from Victoria; two from Queensland; and three from New South Wales (NSW). Urban sites were defined as WWTP with catchments of more than 150,000 people and located in major cities. The specific communities involved in the comparison were anonymised to protect confidentiality with only Region/State identified (eg, NSW-A). The selected data used for the city comparisons is taken from larger data sets, which have been previously published.19,21

Auckland region sampling

Wastewater sampling was completed on nearly every day from 2 May to 18 July 2014 at two urban WWTPs in the Auckland region: Auckland WWTP-A and Auckland WWTP-B. Twenty-four hour composite samples were collected at the Auckland WWTP-A using time-proportional sampling (a collection of 100mL of influent wastewater every 15 min) and at the Auckland WWTP-B using volume-proportional sampling (a collection of 200mL of wastewater in every 1,000m3 influent wastewater) (see Lai et al21 for sampling details).

Over the monitoring period, a total of 65 24-hour composite samples were collected at the Auckland WWTP-A. For the purposes of comparisons for this paper we selected a full week of samples from Tuesday 6 May to Monday 12 May 2014 (there was only one other instance where seven consecutive sampling days were completed over the two and half months of sampling).

A total of 40 24-hour composite samples were collected at the Auckland WWTP-B using volume-proportional sampling mode. Sampling was not routinely conducted at Auckland WWTP-B on Fridays and Saturdays. The most consecutive run of daily samples available for a week comparison were from Monday 23 June to Tuesday 1 July (Friday and Saturday samples were not available). Note, sampling on a day represented samples from the previous night. Consequently, Sunday samples covered the previous Saturday night and so on.

Australian sampling

A week of samples was selected from the 11–17 March 2014 at urban WWTPs in the ACT and Victoria (VIC). Daily wastewater samples were collected at the inlet of the WWTPs using flow proportional sampling (sampling frequency of the autosampler proportional to the actual flow of the influent wastewater). A further week of samples was selected from urban WWTPs in Queensland (QLD), New South Wales (NSW) and Victoria using the time proportional sampling (QLD), volume proportional sampling (VIC) and flow proportional sampling (NSW) respectively. Details of the sampling have been previously reported elsewhere.19

Chemical analysis

Drug residues (parent drugs and metabolites) in samples were measured using an internationally validated analytical method.14,23 The wastewater samples were filtered and spiked with deuterated chemical standards for correcting potential instrumental variability and matrix effects during analysis. Concentrations of the drug residues in the samples were identified and quantified using liquid chromatography coupled with tandem mass spectrometry.

Back-calculation

To obtain the daily mass load (mg/day) of the drug residues in the samples, the measured concentration (µg/L) of the drug residues was multiplied by the daily wastewater flow volume (ML/day). The estimated mass load of the drug residues was then corrected by the average fraction of the drug residue excreted by humans11 to back-calculate the amount consumed (mg/day). This was further normalised to the catchment population size so as to allow comparison of data (mg/day/1,000 people) between catchments.

Statistical analysis

The Mann Whitney test was used to test for differences in estimated drug use between the Auckland region and each Australian state WWTP. For the purposes of statistical comparison with the Australian WWTP, data on the population-normalised consumption from the two Auckland sites were combined.

Results

Occurrence and consumed mass loads of drug residues

Methamphetamine was detected on every day of the selected week at both the two Auckland and eight Australian WWTPs over the selected week of sampling. MDMA was detected at the Auckland WWTP-A and Auckland WWTP-B on only one day over the sampled week, both on Sunday, representing use from the previous Saturday night. In contrast, MDMA was detected on all seven days of the selected week at all eight Australian sites. Cocaine was not detected at either the Auckland WTTP-A or Auckland WWTP-B on any of the days selected for comparison. In contrast, cocaine was detected on every day of the week at all eight Australia WWTP. Codeine and methadone were detected on every sampled day at both the Auckland and Australian sites.

Overall levels of consumption

A mean of 322mg of methamphetamine was estimated to have been consumed per day per 1,000 people at the Auckland WWTP-A, and an estimated mean of 402mg per day per 1,000 people at the Auckland WWTP-B (Table 1). The estimated levels of methamphetamine consumption at the Auckland WWTPs were higher than the ACT, but lower than QLD-A, VIC-B, VIC-A, NSW-B and QLD-B WWTP (Figure 1). There was no statistically significant difference in levels of methamphetamine consumption between Auckland and NSW-C and NSW-D WWTPs. Only low levels of MDMA consumption were found in Auckland WWTP-A and Auckland WWTP-B sites and only on a single day in the selected sampled week, preventing any comparisons to Australia sites (Table 1). A higher level of codeine and methadone consumption was found in all Australian WWTPs compared to Auckland.

Table 1: Estimated population-normalised consumption of methamphetamine, MDMA and cocaine, codeine and methadone (mg/day/1,000 people aged 15–64 years) for Auckland, Queensland, New South Wales, ACT and Victoria urban sites, 2014 and 2015.

‍

‍

Figure 1: Estimated methamphetamine consumption (mg/day/1,000 people aged 15–64 years) for combined Auckland urban WWTPs compared to Queensland, New South Wales, Australian Capital Territory and Victoria urban WWTP, 2014 and 2015.

Discussion

The comparison of WBE results presented in this paper confirms some important qualitative differences in the types of drugs used in New Zealand compared to Australia. Cocaine was not detected at all in the two Auckland WWTPs during the sampled weeks. In contrast, cocaine was detected on every day at all eight Australian WWTPs over the week selected. Over the whole two and a half months of sampling at Auckland WWTPs, cocaine was only detected at the Auckland WWTP-A and only on six occasions.21 The low level of cocaine use in Auckland is consistent with previous New Zealand population drug surveying and studies of ‘at risk’ populations, such as frequent drug users, who report low prevalence of use and poor availability of cocaine.7,8 Similarly, while MDMA was detected on only one day of the sampled week in each of the Auckland WWTPs, it was detected on every day at the eight Australian WWTPs over the week selected. It is important to note that as described above, sampling was not conducted at Auckland WWTP-B on Fridays and Saturdays, representing Thursday and Friday respectively, and if these days were sampled more MDMA may have been found. It should also be noted that the WBE analysis specifically detects the compound MDMA, rather than the array of MDMA analogues which are commonly sold as “ecstasy” in New Zealand. ESR (Institute of Environmental Science and Research)analysis of “ecstasy” tablets seized in New Zealand has found they contain a range of compounds other than MDMA, including methylone and MDPV.24 Many of these substitute ecstasy compounds are associated with more serious adverse effects and hospitalisations than MDMA.24

Our paper also confirms that methamphetamine use is a problem in many areas of both New Zealand and Australia with methamphetamine detected on every day in all 10 WWTPs. As a point of comparison, comparable wastewater studies in 15 of 17 European countries found levels of methamphetamine consumption of less than 200mg per 1,000 people per day.20 This reflects different drug availability between countries with high availability of methamphetamine in Oceania and Asia, while amphetamine and cocaine is more available and preferred in Western Europe. Yet even when a broader comparison is made of all stimulants (ie, methamphetamine, amphetamine, cocaine and MDMA), Australia still ranked second compared to 17 European countries.20 These findings are consistent with the record seizures of methamphetamine made at the border in both Australia25 and New Zealand in recent years,7 and with recent findings from drug monitoring studies of police arrestees and frequent drug users which found increasing methamphetamine use and availability.7,8 The United Nations Office of Drugs and Crime (UNODC) reported the quantity of methamphetamine seized in East and South-East Asia “almost quadrupled” from 2009 to 2014.26 Australian WBE analysis has shown rising use of methamphetamine from 2009 to 2016 in Queensland and South Australia.18,20

Limitations

It is important to note the WBE findings in this paper are from individual WWTP and consequently represent the drug use in a particular local catchment rather than the entire region, state or country. The more extensive WBE in Australia highlights the high level of variability in drug use between WWTP sites, detail which is often lost in the national prevalence findings from national social surveys of drug use. For example, in some Australian states the difference in mass loads of methamphetamine was more than threefold (ie, South Australia, Tasmania, Victoria).20 WBE cannot provide data on the demographics of substance users, the distribution of consumption among users, extent of poly-drug use, routes of administration and the effects of drug use on health and social functioning.15 As a result, WBE is generally advocated as complementary rather than a replacement for existing drug monitoring methods. The populations contributing to wastewater in each WWTP catchment are based on the most recent census, and this may have changed over time.16 The analytical limitations of wastewater analysis have been discussed in detail elsewhere.11,27 WBE calculations of drug consumption assume the use of a single substance and an “average” metabolism time for each drug under investigation (chronic users may have different metabolisation from occasional users).11

‍