Abstract
To explore the utility of wastewater-based epidemiology (WBE) as a public health tool for characterizing the opioid epidemic in the United States, the Water Research Foundation (WRF) funded a three-month study in three U.S. communities. WBE involves using wastewater surveillance and complementary data to measure a community's health, habits, and consumptions-such as illicit drug use. WBE studies have assessed a range of illicit and licit drugs, including stimulants and opioids. While wastewater testing to estimate drug use is slowly gaining traction across the country, more studies are needed to determine how the information can be used for public health decision-making. Eastern Research Group (ERG) and its partners conducted three pilot studies using WBE to monitor the use of pharmaceutical opioids and other drugs in and around Knox County, Tennessee; Memphis, Tennessee; and Madison, Wisconsin. A sampling and monitoring protocol was developed and piloted in each community. The goal of this project was to design a successful WBE study that generates insights about population-level drug use, and to identify challenges around sampling, testing, and reporting that may need to be carefully considered. In measuring opioids and other drugs in wastewater, we sought to identify optimal target biomarkers for monitoring and for estimating community drug use while assessing the generalizability of findings across the sites. Measurements were also evaluated for temporal and spatial trends and compared to public health surveillance data (e.g., opioid-related overdoses, emergency department visits, and prescription data). Insights, lessons-learned, and recommendations for developing a WBE monitoring program for drugs will be presented. Methods The research team reviewed recent literature on WBE for pharmaceutical opioids in wastewater and developed a Sampling and Monitoring Protocol (SMP) to ensure sample integrity and maintain quality control. The SMP was pilot tested at three utilities: First Utility District of Knox County (FUD Knox), Tennessee; Memphis Public Works Division, Tennessee; and Madison Metropolitan Sewerage District, Wisconsin. Staff at each utility collected 24-hour flow-proportional composite wastewater samples at the influent of their wastewater treatment plant (WWTP), and in some cases, at additional upstream locations throughout their collection area (such as pumping stations). A total of 268 composite samples were collected over a period of up to three months. All wastewater samples were shipped to the Laboratory for Aquatic Environmental Microbiology and Chemistry at the University of Wisconsin–Milwaukee for analysis of 27 drugs or drug metabolites by tandem liquid chromatography-mass spectrometry (LC-MS/MS). The 27 target drugs and metabolites are classified as either opioids, opioid antagonists, and stimulants and were selected to include compounds that have been detected previously in wastewater studies of opioids and illicit drugs. Three major research objectives informed our analysis of opioids and other drugs in wastewater. First, to assess the similarity and dissimilarity of wastewater opioid levels across the three sites, we present summary statistics (detection frequency, mean, standard deviation, median, range) of opioid metabolite levels in wastewater samples and compare these statistics to data obtained from other studies. The ERG team also back-calculated average drug consumption rates (i.e., mass of compound per person per day) for each community using correction factors that account for drug metabolism and excretion. Second, to characterize the spatial and temporal variability of opioids in wastewater, we present time series plots and choropleth maps. Following established methods, we examined how flow and population normalization methods would affect the trends of different opioid levels across sites. Third, to better understand the observed temporal trends in wastewater, results from the wastewater analyses for each utility were compared to available local public health monitoring data related to opioid use (e.g., drug overdoses, emergency department visits, and prescriptions). We also present data on how the COVID-19 pandemic may have influenced drug use behaviors in the community. Key Findings The relative detection frequencies of target drugs and their metabolites provide insight into which analytes are better suited for wastewater monitoring. Among other considerations, a WBE study should prioritize the analysis of the more sensitive analytes-that is a WBE monitoring program should target drugs that will likely be detected in wastewater. Across the three pilot studies the following observations can be made: Eight target opioids or opioid-metabolites (codeine, fentanyl, methadone, morphine, norfentanyl, oxycodone, oxymorphone, and tramadol), and four target stimulants (amphetamine, benzoylecgonine, cocaine, and methamphetamine) were detected in wastewater samples collected from all three communities. Nine target opioids or metabolites were not detected in any of the samples collected as part of this study (4-ANPP, 6-AM, buprenorphine, carfentanil, 3-methylfentanyl, meperidine, naloxone, norcodeine, and sufentheretanil) despite likely use in the upstream community. In some cases, the metabolites of drugs were detected more frequently than the parent compound, and in other cases, the parent compound was detected more frequently. Dose consumption rates (DCR) within each pilot study's collection area were estimated for frequently detected drugs. DCRs are the average amount of drug consumed per person per day. In the two larger communities of Memphis, TN and Madison, WI, cocaine was estimated to be the most consumed drug over the sampling period and had comparable DCRs ranging from 3.1 to 4.0 mg/person/day. In contrast, in the smaller community served by FUD Knox, the estimated DCR for cocaine ranged from 0.38 to 0.57 mg/person/day, and the most consumed drug was estimated to be codeine 1.20 mg/person/day (Figure 1). We also show that normalizing wastewater measurements by wastewater flow and population are critical for interpreting spatial and temporal trends of drug use in the upstream community. For example. Figure 2 shows a comparison of raw wastewater measurements and flow-adjusted measurements at the Memphis PWD WWTP for four stimulants. An increasing trend can be seen in the flow-adjusted values over time that is not observed in the raw data. Figure 3 shows a comparison of raw wastewater measurements and population and flow-adjusted dose consumption rates for Methadone at the Madison MSD WWTP. A distinctly different spatial distribution can be seen between the two maps. Public health interpretations of wastewater measurements are not apparent when looking at raw wastewater concentrations alone. The temporal and spatial variability of wastewater concentrations observed in this study throughout a week, over multiple weeks, and across sampling points within a collection system can provide useful information to utilities regarding appropriate sample collection frequencies and sampling locations when developing a WBE surveillance program. For example, we show that some drugs (e.g., cocaine and its metabolite benzoylecgonine) were present at greater levels on weekends compared with weekdays. Administrative data sources on opioid misuse (e.g., OD deaths, ED visits) were not associated with wastewater measurements and may be too crude to describe detailed trends within a community. For example, in one pilot study, no OD deaths occurred within the community during the sampling period despite the presence of drugs in wastewater. However, prescription drug use data provided better insights into understanding the wastewater data than other public health data sources. For example, in Madison, population level dose consumption rates estimated from wastewater were comparable to prescription data for some drugs such as oxycodone and tramadol, but not for other drugs like methamphetamine, which suggests the latter has more illicit use. The results of these three pilot studies are useful for a wide range of community partners including: utilities looking to develop WBE protocols when implementing a surveillance program, public health officials and communities looking to use WBE as an additional public health tool for monitoring drug use and abuse, and laboratories looking to develop analytical methods for measuring a suite of drugs in wastewater. Estimating population drug use through wastewater testing is slowly gaining traction across the United States, but more pilot studies are needed to determine the policy value of wastewater testing- that is, how this tool can be operationalized for decision making. This study contributes to the growing body of evidence that wastewater is a valued resource for all communities.
The following conference paper was presented at the Public Health and Water Conference & Wastewater Disease Surveillance Summit in Cincinnati, OH, March 21-24, 2022.
Author(s)C. Dassuncao1; R. DeVries2; C. Hu3; A. Keshaviah4; T. Miller5
SourceProceedings of the Water Environment Federation
Document typeConference Paper
Print publication date Mar, 2022
DOI10.2175/193864718825158292
Volume / Issue
Content sourcePublic Health and Water Conference
Copyright2022
Word count24