Wastewater Surveillance Provides Spatiotemporal SARS-CoV-2 Infection Dynamics

Xiawan Zheng , Kathy Leung , Xiaoqing Xu , Yu Deng , Yulin Zhang , Xi Chen , Chung In Yau , Kenny W.K. Hui , Eddie Pak , Ho-Kwong Chui , Ron Yang , Hein Min Tun , Gabriel Matthew Leung , Joseph Tsz Kei Wu , Malik Peiris , Leo Lit Man Poon , Tong Zhang

Engineering ›› 2024, Vol. 40 ›› Issue (9) : 77 -85.

PDF (2130KB)
Engineering ›› 2024, Vol. 40 ›› Issue (9) :77 -85. DOI: 10.1016/j.eng.2024.01.016
Research
Article
Wastewater Surveillance Provides Spatiotemporal SARS-CoV-2 Infection Dynamics
Author information +
History +
PDF (2130KB)

Abstract

Wastewater surveillance (WWS) can leverage its wide coverage, population-based sampling, and high monitoring frequency to capture citywide pandemic trends independent of clinical surveillance. Here we conducted a nine months daily WWS for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) from 12 wastewater treatment plants (WWTPs), covering approximately 80% of the population, to monitor infection dynamics in Hong Kong, China. We found that the SARS-CoV-2 virus concentration in wastewater was correlated with the daily number of reported cases and reached two pandemic peaks three days earlier during the study period. In addition, two different methods were established to estimate the prevalence/incidence rates from wastewater measurements. The estimated results from wastewater were consistent with findings from two independent citywide clinical surveillance programmes (rapid antigen test (RAT) surveillance and serology surveillance), but higher than the cases number reported by the Centre for Health Protection (CHP) of Hong Kong, China. Moreover, the effective reproductive number (Rt) was estimated from wastewater measurements to reflect both citywide and regional transmission dynamics. Our findings demonstrate that large-scale intensive WWS from WWTPs provides cost-effective and timely public health information, especially when the clinical surveillance is inadequate and costly. This approach also provides insights into pandemic dynamics at higher spatiotemporal resolutions, facilitating the formulation of effective control policies and targeted resource allocation.

Graphical abstract

Keywords

SARS-CoV-2 / Wastewater surveillance / Prevalence / Effective reproductive number

Highlight

• Large-scale intensive wastewater surveillance was conducted to obtain citywide spatiotemporal trends.

• Wastewater surveillance captured the pandemic trend three days earlier than reported cases.

• Two methods were established to estimate prevalence from wastewater datasets.

• Wastewater surveillance revealed that the reported cases were underestimated during the pandemic outbreak.

• The wastewater-based effective reproductive number (Rww) was comparable to the incidence-based Rcc.

Cite this article

Download citation ▾
Xiawan Zheng, Kathy Leung, Xiaoqing Xu, Yu Deng, Yulin Zhang, Xi Chen, Chung In Yau, Kenny W.K. Hui, Eddie Pak, Ho-Kwong Chui, Ron Yang, Hein Min Tun, Gabriel Matthew Leung, Joseph Tsz Kei Wu, Malik Peiris, Leo Lit Man Poon, Tong Zhang. Wastewater Surveillance Provides Spatiotemporal SARS-CoV-2 Infection Dynamics. Engineering, 2024, 40(9): 77-85 DOI:10.1016/j.eng.2024.01.016

登录浏览全文

4963

注册一个新账户 忘记密码

1. Introduction

Timely and accurate surveillance of the coronavirus disease 2019 (COVID-19) pandemic is essential for evaluating virus transmission patterns and implementing effective public health interventions. However, current clinical surveillance using individual reverse transcription quantitative polymerase chain reaction (RT-qPCR) is affected and biased by multiple factors, such as clinical testing capacity, individual willingness to be tested, and reporting policies. Wastewater surveillance (WWS) for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) provides an effective and efficient tool for large-scale surveillance and has been applied in more than 70 countries with 166 available online dashboards [1], [2]. In Hong Kong, China, the WWS studies were initiated since the start of the pandemic in early 2020, and have been used to inform public health interventions [3], [4], [5], [6]. As an emerging and evolving surveillance tool, no standardized measures have been developed to convert wastewater datasets into usable public health information. Hong Kong, China implemented intensive testing, case detection, and contact tracing together with strict public health and social measures (PHSMs), including mask-wearing and social distancing, from 2020 to 2022. As a densely populated city with a well-established wastewater system, Hong Kong, China provides a great opportunity for us to explore the potential use of wastewater data to reflect disease transmission kinetics.

Most studies have observed that the SARS-CoV-2 virus concentration in wastewater is significantly correlated with the clinical cases data [7], [8], [9]. However, the utilization of wastewater measurements to infer COVID-19 infections remains challenging and lacks cross-validation with concurrent clinical surveillance findings. A back-calculation method uses the division of total viral load at the catchment area and individual virus shedding load to estimate the number of infections [4], [10]. However, the results are influenced by various parameters, such as individual virus shedding loads and sewer hydraulic retention time [11]. To address these limitations, data-driven methods have been applied to estimate COVID-19 infections, such as linear regression models [12], [13], artificial neural network model [11], and so forth.

To facilitate the direct use of wastewater data for public health decision-making, recent studies have focused on public health metrics derived from wastewater measurements. For example, effective reproductive number (Rt) has been firstly estimated from wastewater (Rww) at two wastewater treatment plants (WWTPs) in Switzerland, yielding similar results as Rt from clinical case data (Rcc) [14]. In this study, we aimed to evaluate the effectiveness of WWS to reveal higher spatiotemporal resolution Rww at both citywide and regional levels and to compare these findings with Rcc from reported cases.

Intensive large-scale WWS studies are important for validating the utility of wastewater datasets for public health interventions; however, the long-term spatiotemporal datasets were limited so far. To address the current research gaps, this study used citywide daily WWS data from 12 WWTPs to reflect the fifth pandemic wave in Hong Kong, China, when the capacity for clinical testing was overwhelmed from February 2022 [15]. Additionally, we used the wastewater measurements to estimate the prevalence/incidence rates of SARS-CoV-2 infections and other epidemiological parameters, which can be applied to inform public health decision-making.

Our results emphasize the role of WWS in providing unbiased and timely results regarding pandemic trends. This highly multidisciplinary study contributes to a better understanding of the spread of COVID-19 in communities by using a combination of multiple independent clinical surveillance tools. We show that multidimensional datasets can provide a more accurate and comprehensive snapshot of the pandemic trends for informing effective public health interventions. The methodology and observation data in this study provides a framework and comparative dataset for other cities to translate WWS measurement into useful public health information, including the ratio between wastewater viral copy number and clinical cases (WC ratio), estimated prevalence/incidence rates and Rt. This study adds new knowledge to the interpretation of wastewater measurements as an informative public health tool. Also, this study facilitates the integration of WWS into local and global public health surveillance systems to monitor the COVID-19 infections in the community. These findings are essential to prevent emergence/re-emergence of other pandemic outbreaks in the future.

2. Materials and methods

2.1. Sample collections

Daily untreated influent samples were collected from 12 WWTPs for approximately nine months (261 d) from February 14, 2022 to November 1, 2022, serving approximately 80% of the population in Hong Kong, China (Fig. S1 in Appendix A). These wastewater samples were 24 h composite samples consisting of subsamples collected by autosamplers at 15 or 30 min sampling time interval. Further details were presented in Table S1 in Appendix A.

2.2. Sample processing and analysis

Heat-inactivation was performed at 60 °C for 30 min before samples processing. Wastewater samples were concentrated by the polyethylene glycol (PEG) precipitation method and then extracted using QIAamp Viral RNA Mini Kit (Qiagen, Germany) on the QIAcube Connect (Qiagen) to obtain 50 μL total RNA as described in Zheng et al. [16]. The SARS-CoV-2 virus concentration in wastewater was quantified by one-step RT-qPCR as described in Deng et al. [6]. The wastewater samples collected from February 14, 2022 to April 26, 2022 were detected using the N1 assay from the United States Centers for Disease Control, and the samples collected from April 27, 2022 to November 1, 2022 were detected using a newly designed assays (HKU-05) targeting the N gene of the SARS-CoV-2 genome to obtain more sensitive results in response to the emergence of new variants. The SARS-CoV-2 virus concentration in wastewater was quantified by using a standard curve in each 96 well plate, which was generated from a dynamic ranges of 10 to 107 copies per reaction. During this quantification step, we considered the amount of processed wastewater (40 mL), the RNA elution volume in each nucleic acid extraction (50 μL), and the template volume in a polymerase chain reaction (PCR) reaction (4 μL) and assumed the 100% recovery efficiency of the virus particles and no PCR inhibition for the entire analytical procedures. Further details were provided in Table S2 in Appendix A.

2.3. Quality assurance and control

The limit of detection (LoD) of the two assays were determined by the detection of SARS-CoV-2 RNA from the 2.5 to 10.0 copies·μL−1 for ten replicates. The LoD values of the quantitative PCR (qPCR) assays were 10.00 and 2.65 copies·μL−1 for the N1 and HKU-05 assays, respectively. Correspondingly, the LoD values of WWS data were 3125 and 828 copies·L−1 of wastewater for the N1 and HKU-05 assays, respectively. For the total 2887 collected wastewater samples, 808 samples (28.0%) were analyzed using the N1 assay, and 2079 samples (72.0%) were analyzed using the HKU-05 assay. The detection rates were 28.0% (808/2887) and 71.3% (2059/2887) for the N1 and HKU-05 assays, respectively.

Forty wastewater samples from the Northwest Kowloon Preliminary Treatment Works (NWK) were conducted in duplicate to investigate the reproducibility of the entire testing protocol. As shown in Fig. S2 in Appendix A, the SARS-CoV-2 virus concentration in two duplicates showed a slope of 0.90 (R2 = 0.68) and an insignificant difference was observed by paired T-test (p > 0.05).

2.4. Epidemiological data

The daily number of reported COVID-19 cases, including local cases and imported cases, was obtained from the official website of the Centre for Health Protection (CHP) of Hong Kong, China†. From February 26, 2022, the Government started to report the positive cases based on PCR tests and rapid antigen test (RAT). Therefore, from this timepoint onwards, the daily number of reported cases was defined as the total number of PCR and RAT positive tests. The PCR testing results were obtained from the laboratory tests of the collected respiratory specimens from sentinel sites, and the RAT testing results were collected by individuals reporting to the designated website after self-testing. Additional information on the multiple data sources was provided in Table S3 in Appendix A.

2.5. Calculation of SARS-CoV-2 virus concentration in wastewater of Hong Kong, China

The daily total viral load in Hong Kong, China was calculated using Eq. (1):

$ L=\sum_{i=1}^{n}\left(C_{i} \times Q_{i}\right) / \frac{\sum_{i=1}^{n} Q_{i}}{Q}$

where i is the order of each sampled WWTP; L is daily total viral load in a city (copies·d−1); Ci is daily SARS-CoV-2 virus concentration in wastewater from each WWTP (copies·L−1); Qi is daily flow rate of each sampled WWTP (103 L·d−1); n is number of the sampled WWTPs.

The citywide SARS-CoV-2 virus concentration in wastewater of Hong Kong, China was calculated using Eq. (2):

$ C=\frac{L}{Q}$

where C is daily citywide SARS-CoV-2 virus concentration in wastewater (copies·L−1); Q is daily total flow rate in a city, using a value of 2.8 million m3·d−1 in Hong Kong, China.

2.6. Calculation of WC ratio

The WC ratio has been proven to be an effective index for reflecting clinical testing capacity and policy changes during the pandemic [17], [18], and it was calculated as follows.

$WC=\frac{C}{N}$

where N is daily total number of reported cases by CHP.

2.7. Estimation of prevalence/incidence rates from wastewater datasets

2.7.1. Method 1: Based on virus shedding load

The viral load/virus concentration in wastewater represents the prevalence rate of infections within the sampled wastewater catchment area, including newly infections on the sampling day (defined as incidence rate) and existing infections that have not yet recovered. The prevalence/incidence rates were estimated using Eqs. (4), (5):

$\mathrm{PR}_{\text {load }}=\frac{L}{S \times W \times P} / \theta_{1}$
$\mathrm{IR}_{\text {load }}=\frac{\mathrm{PR}_{\text {load }}}{D_{1}}$

where PRload is daily prevalence rate estimated from wastewater datasets based on virus shedding load; S is virus shedding rate in fecal samples per infected person, assuming 7.30 log10 copies·g−1 of feces [19], The virus shedding rate in fecal samples varies from 2 to 8 log10 copies·mL−1 as reviewed by Zheng et al. [20]; W is wet weight of fecal samples, assuming 164 g·d−1, average values of 128 [21], [22] and 200 g·d−1 [10]; P is population size in Hong Kong, China, 7 413 100 people; θ1 is method sensitivity of WWS, assuming 81.8% based on the positivity rate in fecal samples [23] is the positivity rate in fecal samples ranged from 15.3% to 100% as reviewed by Guo et al. [24]. IRload is daily incidence rate estimated from wastewater datasets based on virus shedding load; D1 is period of individual shedding virus into wastewater, assuming 7 d [25].

2.7.2. Method 2: Based on reported cases in low infection rate period

This method assumed that the ascertainment ratio (here, defined as the ratio of the total number of reported cases to the total number of new infections on one day) was closer to 100% in the low infection rate period, given that there was sufficient testing capacity for clinical surveillance to identify all infections. The training dataset was selected from April 18, 2022 to June 2, 2022 because fewer than 1000 cases per day were reported during this period. Totally, 498 wastewater samples from 12 WWTPs and 46 daily prevalence rates from reported cases by CHP were used as training datasets. Next, a linear regression model was fitted and applied to estimate the prevalence rate at other time periods in Hong Kong, China and the regional prevalence rates covered by the 12 WWTPs. The method sensitivity was assumed to be 81.8% as Method 1. The daily incidence rate was estimated by the daily prevalence rate dividing the virus shedding period length D1 as Method 1.

2.8. Comparison with prevalence/incidence rates from clinical surveillance datasets

2.8.1. Method 3: Based on reported cases from CHP

This method assumed that the daily prevalence rate was the sum of the daily incidence rate for that day and the previous 6 d (7 d in total). The daily incidence rate was calculated using the daily number of reported cases from CHP dividing the population size in Hong Kong, China.

2.8.2. Method 4: Based on RAT surveillance

Starting from March 3, 2022, a daily RAT surveillance programme was conducted by delivering 10 000 RAT to a cohort of participants across 18 districts to obtain the daily point prevalence rate in Hong Kong, China [26]. Next, the daily point prevalence rate was converted to the daily incidence rate as follows [25].

$\mathrm{IR}_{\mathrm{RAT}}=\frac{\mathrm{PR}_{\mathrm{RAT}}}{D_{2}} / \theta_{2}$

where PRRAT is daily point prevalence rate by RAT surveillance from an online dashboard†. IRRAT is daily incidence rate based on RAT surveillance; D2 is period of RAT positive results, assuming 7 d [25]; θ2 is method sensitivity of RAT tests, assuming 75% [25].

2.8.3. Method 5: Based on serology surveillance

From January 1, 2022 to July 31, 2022, a community-wide serology surveillance was conducted with 5310 participants in Hong Kong, China and a cumulative infection attack rate (IAR) of 45% (confidence interval (CI) value: 41%-48%) was estimated during the entire monitoring period [27]. The IAR was defined as the proportion of the people in contacts with the disease. We assumed that the IAR was equivalent to the actual incidence rate in the community, and that there was a consistent conversion factor between the daily IAR by serology surveillance and the daily incidence rate of reported cases from CHP on the same day. From January 1, 2022 to July 31, 2022, the cumulative IAR was 45.0% and the cumulative incidence rate of reported cases was 18.1%; thus, the conversion factor was 2.5. Correspondingly, from March 3, 2022 to November 1, 2022, the cumulative incidence rate of reported cases was 20.4%, and the cumulative IAR during was estimated to be 50.9% based on the conversion factor from the above serology surveillance.

2.9. Calculation of Rt

2.9.1. Rcc

The calculation of incidence-based Rt used the number of reported cases from CHP as the input datasets for the R package EpiNow. The generation time of the disease was estimated with a mean value of 4.6 d and a standard deviation (SD) of 3.1 d [28]. The incubation time was described by a gamma distribution with the mean ± SD values of (3.5 ± 2.6) d [29]. The delay time between the infections and symptoms was estimated by using a gamma distribution with the shape of 1.83 and a rate of 0.43 [30].

2.9.2. Rww

The calculation of wastewater-based Rt used incidence rates estimated from Methods 1 or 2 as the input dataset into the EpiNow package. The generation time and incubation time were the same as the calculation of Rcc, but the delay time between infections and reported results was assumed to be zero, because the wastewater measurements were considered as a near real-time surveillance tool to obtain the prevalence rate within the catchment areas.

2.10. Statistical analysis

The trends of the wastewater and clinical datasets were normalized using the 7 d moving average values. Non-parametric Kendall's Tau-b coefficient was used to investigate the correlation between SARS-CoV-2 virus concentration in wastewater and reported cases numbers. The mean absolute error (MAE) values for the comparison of prevalence rates estimated from wastewater and clinical surveillance datasets were calculated using the R package “Metrics.” All the statistical analyses were performed using R version 4.0.3.

3. Results

3.1. Monitor pandemic trends

During the sampling period, the SARS-CoV-2 virus concentration in wastewater at most regional WWTPs changed in a similar pattern, with a sharp peak in early-February and a smaller peak in early-September (Fig. 1). From February 14, 2022 to November 1, 2022, the 7 d moving average virus concentration in wastewater showed a significant correlation with the daily number of reported cases from CHP in Hong Kong, China (Tau-b value = 0.74, p-value < 2.2 × 10−16, Fig. 2(a)). The wastewater data indicated two pandemic peaks on March 1, 2022 and September 6, 2022, and both of them were 3 d earlier than the peak of reported cases from CHP. The 3 d time window was observed based on the difference between the sample collection date of WWS and the reporting date of the clinical surveillance. These results indicated the effectiveness of WWS as an early warning system for monitoring the overall pandemic trends in a city and in different regions of a city.

The SARS-CoV-2 virus concentration in wastewater increased sharply from February 14, 2022, whereas the daily number of reported cases from CHP only increased slightly (likely due to reaching maximum clinically RT-qPCR testing capacity) until positive RAT results were included for reporting on February 26, 2022. The high WC ratio during this period (February 14 to 25) suggested an undercounting of infection cases, because the PCR testing capacity did not keep up with the exponentially rising infections during the rapidly increasing phase of the outbreak (Fig. 2(b)). Later, the WC ratio remained relatively stable for the second period, from February 26, 2022 to June 2, 2022, with a mean ± SD WC ratio of (10.70 ± 4.60). The WC ratios doubled (21.97 ± 8.70) from June 3, 2022 to October 16, 2022, probably caused by a decrease in self-testing and reporting willingness as the COVID-19 anxiety had abated. Additionally, increased WC ratio may be attributed to the introductions of other Omicron subvariants (e.g., BA.2.12.2 and BA.5) during this period, which resulted in an upsurge of wastewater viral load.

3.2. Estimate prevalence/incidence rates

Prevalence rates were estimated from wastewater datasets using two different methods, such as, based on virus shedding load and based on reported cases in low infection rate period (Methods 1 and 2), and these two methods presented similar results to community RAT surveillance estimates (Method 4). All these three methods were higher than the reported cases from CHP (Method 3) and captured the trend peaks earlier (Fig. 3). Additionally, MAE analysis indicated that the two wastewater estimates were more similar to RAT surveillance estimates than the estimates derived from the reported cases from CHP, owing to lower MAE values for the comparison of prevalence rates (Table S4 in Appendix A). These results imply that both WWS and RAT surveillance yield more effective and accurate prevalence rates rather than those of reported cases from CHP. In practice, the daily citywide pandemic situation was monitored using 12 WWTPs samples per day by WWS, whereas the RAT surveillance used 10 000 RATs per day.

From March 3, 2022 to November 1, 2022, the cumulative IAR was estimated to be 48.8% and 44.8% (CI: 29.6%-60.0%) from wastewater based on virus shedding load and based on reported cases in low infection rate period methods, respectively. The cumulative IAR from WWS was comparable to RAT surveillance estimates (46.8%) and serology surveillance estimates (50.9%; details in the Methods 4 and 5). These independent estimates with similar results provide evidence to support the applicability of WWS at WWTPs for obtaining timely and cost-effective public health information.

Additionally, the daily prevalence rates in each WWTP were estimated from wastewater at a higher spatial resolution (Fig. S2 in Appendix A). The regional prevalence rates in most WWTPs presented similar values and pandemic trends, except for those in NWK in late-February and Sai Kung in early-September which exhibited higher prevalence rates (approximately 50%). These results suggested that WWS provided geographical pandemic differences and near real-time regional infection dynamics at the regional WWTPs, which can help policymakers to obtain timely and accurate information for targeted resource allocation.

3.3. Estimate Rt

The epidemic parameter of Rt was calculated and compared using wastewater measurement (Rww) and reported cases (Rcc; Fig. 4). Both Rww and Rcc increased to more than 1 in mid-May, which was correlated with the beginning of the second pandemic peak. Rww changed in parallel with Rcc from June to August; however, Rww appeared to be two weeks ahead of Rcc and showed a steeper increase than Rcc. Starting from the early-August, both Rww and Rcc showed a similar increase above 1 and reached the peak after two weeks. Similarly, the Rww values from the regional WWTPs were estimated to present regional transmission dynamics (Fig. S4 in Appendix A). These results indicated a concordance between Rww and Rcc, where Rww provided trends earlier than Rcc. Both two parameters could reflect transmission dynamics.

4. Discussion

The WWS has been considered as an accurate, near real-time, cost-effective, and informative tool for obtaining data to inform public health interventions. Well-validated case studies are important to reveal its applicability in this regard. To achieve this purpose, we addressed the technical issues associated with large-scale long-term WWS monitoring. Firstly, this study demonstrated the feasibility of using the 24 h composite samples from WWTPs to monitor the pandemic dynamics in Hong Kong, China. Using Hong Kong as an example, compared to sampling in upstream stationary sites with smaller catchment areas, this sampling strategy is a cost-effective method because it covers larger catchment areas for each WWTP and results in less required sample number for sampling and detection. Also, longer sampling period suggests that it is more representative than 3 h composite samples in stationary sites. In the post pandemic period, with increasing “pandemic fatigue” in the population, conducting WWS at existing WWTPs infrastructures in most cities may be an efficient and effective method for monitoring pandemic trends to inform timely preventive and control measures, without the need to implement comprehensive compulsory testing of individuals. Secondly, we considered the reproducibility and quality control of sample processing needed for monitoring longitudinal trends. In this study, we processed duplicates of NWK wastewater samples for each batch experiment and found good correspondences (Fig. S3). Currently, most laboratories have developed their own protocols and quality control procedures, such as the recovery efficiency of spiked viruses [31], [32], or relative values of internal virus concentration (e.g., Pepper Mild Mottle Virus (PMMoV) and cross-assembly phage (CrAssPhage)) [14], [33]. Further evaluation of different quality control methods is needed to obtain a standardized protocol for the wide application of WWS and model establishment in the future.

Additionally, WWS is considered as an early warning indicator of impending COVID-19 outbreaks [3], [34]. This study demonstrates that WWS can provide situational awareness and early warning of the ebb and flow of an ongoing outbreak, provide signals of rapidly increasing Rt, and also indicate epidemic peaks 3 d earlier than those observed in clinically reported cases. All these advantages are useful to implement public health interventions especially when healthcare systems and testing capacities are overloaded. The provision of WWS data unbiased by confounders associated with case-testing and reporting are benefits that allow a better and more reliable data for public health decision making (e.g., distribution of medical supplies, allocation of hospital beds, and relocation of testing stations). However, the early warning efficacy may be affected by the practical arrangement of WWS, such as wastewater sample collection, sample transportation, virus pretreatment, and qPCR assays detection, The turnaround time of wastewater results ranged from several hours to longer, depending on the available resources (instrument, labor, cost, and so forth).

We found that the wastewater trends coincided with epidemic trends by clinical surveillance during the fourth [4] and fifth (this study) pandemic waves in Hong Kong, China. Wastewater trends further reflected the impact of changes on PHSMs (i.e., relaxation of travelling and social distancing measures) [35], [36]. Factors such as the emergence of the new variants of concern (i.e., BA.5 in mid-August) and increasing population immunity induced by vaccination and/or re-infections impacts on virus circulation, which may lead to differences in the proportions of mild and asymptomatic infections. While asymptomatic or mild infections may be missed by conventional case ascertainment methods, they are well reflected in WWS trends. Pandemic fatigue in the population may affect healthcare-seeking behaviors and individual testing willingness but these will not affect WWS signals. Thus, WWS may serve as a timely indicator of the epidemic or pandemic trends, more akin to weather forecasts.

In this study, we attempted to estimate prevalence/incidence rates from WWS data using two different methods, which yielded similar results to those from community RAT surveillance and serology surveillance. However, our methods for estimating infection case numbers from WWS data may have some limitations. Firstly, wastewater viral load cannot distinguish between new infections, re-infections, and convalescent patients with long term shedding. Infection, re-infection, and prolonged virus shedding patients are all included and may be double-counted in the estimates of prevalence/incidence rates from WWS data. Since asymptomatic infections, milder infections, and re-infections were more common during Omicron outbreaks than during the Alpha and Delta outbreaks [37], the cumulative IAR from WWS during Omicron outbreaks may differ from case-reporting data. Additionally, prolonged virus shedding patients may confound estimates of prevalence/incidence rates and cumulative IAR from WWS data. Secondly, the infection case numbers were estimated from WWS data assuming that the “average virus shedding load per infected person per day” was unchanged and the virus shedding period length was the same for each case. In reality, individual viral loads and duration of viral shedding may vary from different Omicron lineages [38]. Thirdly, other factors may lead to uncertainty in wastewater signals, such as the decay of virus signals during transportation in sewer systems, selection of flow-weighted or time-weighted composite wastewater samples, effect of virus inactivation before processing, method recovery efficiency of virus signals during pretreatment, RNA extraction efficiency, inhibitory effect of RT-qPCR detection, fecal weight of different populations, and so forth. The method based on reported cases in low infection rate period may also underestimate of infection cases number because the case ascertainment ratio in CHP data is surely not 100% (the number of reported cases did not include asymptomatic or very mild infections). Fourthly, establishing real-time wastewater-based prevalence/incidence rates depends on timely clinical data that reflects the true number of infections (i.e., based on serology surveillance, point-prevalence survey, and empirical virus load of infections), which need to be regularly updated with the emergence of new variants of concern and widespread vaccination programmes. Fifthly, in this study, regional prevalence/incidence rates were estimated from each WWTP to present the geographical difference, but the impact factors contributing to these differences are still uncertain due to the lack of enough regional epidemiological information for verification.

5. Conclusions

In summary, our data shows that WWS at WWTPs serves as an effective, near real-time, accurate, and cost-effective strategy to provide data on the transmission dynamics of COVID-19 with a higher spatiotemporal resolution, which can be used to inform public health actions at both the citywide and regional levels. The WWS has several advantages over passive case ascertainment surveillance because it is not biased by limitations in diagnostic testing capacity or patient’s healthcare-seeking behaviors. Additionally, WWS captures signals from mild and asymptomatic infections that are relevant for ongoing transmission but are often missed by conventional case ascertainment surveillance. The collected datasets and practical experiences from this study will provide guidance for the future use of WWS to monitor COVID-19 pandemic and other diseases (i.e., influenza, poliovirus, monkeypox virus, and emerging unknown diseases) in other countries and regions.

Acknowledgments

This study was financially supported by the Health and Medical Research Fund (COVID1903015), the Food and Health Bureau, the Government of the Hong Kong Special Administrative Region (SAR), China. This research was supported by the AIR@InnoHK (KL, GML, and JTW) and Health@InnoHK (MP and LLMP) administered by the Innovation and Technology Commission of the Government of the Hong Kong SAR. We appreciated the help of the Environmental Protection Department and Drainage Services Department of the Government of the Hong Kong SAR for the wastewater sample collection and delivery. Xiawan Zheng, Xiaoqing Xu, and Yulin Zhang would like to thank the University of Hong Kong for the postdoctoral fellowship. Xi Chen thanks the University of Hong Kong for the Postgraduate Studentship. We also express our gratitude to Vicky Fung, Hiuchung Lam, and Sijue Wang for their technical support and assistance with sample logistics.

Compliance with ethics guidelines

Xiawan Zheng, Kathy Leung, Xiaoqing Xu, Yu Deng, Yulin Zhang, Xi Chen, Chung In Yau, Kenny WK Hui, Eddie Pak, Ho-Kwong Chui, Ron Yang, Hein Min Tun, Gabriel Matthew Leung, Joseph Tsz Kei Wu, Malik Peiris, Leo Lit Man Poon, and Tong Zhang declare that they have no conflict of interest or financial conflicts to disclose.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.eng.2024.01.016.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]

Naughton CC, Roman FA, Alvarado AGF, Tariqi AQ, Deeming MA, Bibby K, et al. Show us the data: global COVID-19 wastewater monitoring efforts, equity, and gaps. FEMS microbes 2023; 4:xtad003.
[2]

Keshaviah A, Diamond MB, Wade MJ, Scarpino SV, Ahmed W, Amman F, et al. Wastewater monitoring can anchor global disease surveillance systems. Lancet Glob Health 2023; 11(6):e976-81.
[3]

Xu X, Zheng X, Li S, Lam NS, Wang Y, Chu DKW, et al. The first case study of wastewater-based epidemiology of COVID-19 in Hong Kong. Sci Total Environ 2021; 790:148000.
[4]

Zheng X, Li S, Deng Y, Xu X, Ding J, Lau FTK, et al. Quantification of SARS-CoV-2 RNA in wastewater treatment plants mirrors the pandemic trend in Hong Kong. Sci Total Environ 2022; 844:157121.
[5]

Deng Y, Zheng X, Xu X, Chui HK, Lai WK, Li S, et al. Use of sewage surveillance for COVID-19: large-scale evidence-based program in Hong Kong. Environ Health Perspect 2022; 130(5):057008.
[6]

Deng Y, Xu X, Zheng X, Ding J, Li S, Chui HK, et al. Use of sewage surveillance for COVID-19 to guide public health response: a case study in Hong Kong. Sci Total Environ 2022; 821:153250.
[7]

Hillary LS, Farkas K, Maher KH, Lucaci A, Thorpe J, Distaso MA, et al. Monitoring SARS-CoV-2 in municipal wastewater to evaluate the success of lockdown measures for controlling COVID-19 in the UK. Water Res 2021; 200:117214.
[8]

Wu F, Xiao A, Zhang J, Moniz K, Endo N, Armas F, et al. Wastewater surveillance of SARS-CoV-2 across 40 U.S. states from February to June 2020. Water Res 2021; 202:117400.
[9]

Graham KE, Loeb SK, Wolfe MK, Catoe D, Sinnott-Armstrong N, Kim S, et al. SARS-CoV-2 RNA in wastewater settled solids is associated with COVID-19 cases in a large urban sewershed. Environ Sci Technol 2021; 55(1):488-98.
[10]

Wu F, Zhang J, Xiao A, Gu X, Lee WL, Armas F, et al. SARS-CoV-2 titers in wastewater are higher than expected from clinically confirmed cases. mSystems 2020; 5(4):e00614-20.
[11]

Jiang G, Wu J, Weidhaas J, Li X, Chen Y, Mueller J, et al. Artificial neural network-based estimation of COVID-19 case numbers and effective reproduction rate using wastewater-based epidemiology. Water Res 2022; 218:118451.
[12]

Schoen ME, Wolfe MK, Li L, Duong D, White BJ, Hughes B, et al. SARS-CoV-2 RNA wastewater settled solids surveillance frequency and impact on predicted COVID-19 incidence using a distributed lag model. ACS EST Water 2022; 2 11):2167-74.
[13]

Zulli A, Pan A, Bart SM, Crawford FW, Kaplan EH, Cartter M, et al. Predicting daily COVID-19 case rates from SARS-CoV-2 RNA concentrations across a diversity of wastewater catchments. FEMS Microbes 2021; 2:xtab022.
[14]

Huisman JS, Scire J, Caduff L, Fernandez-Cassi X, Ganesanandamoorthy P, Kull A, et al. Wastewater-based estimation of the effective reproductive number of SARS-CoV-2. Environ Health Perspect 2022; 130(5):057011.
[15]

Chow WK, Chow CL, Cheng CH. Estimation of the total number of infected cases in the 5th wave of COVID-19 in Hong Kong. 2022. medRxiv:22278708.
[16]

Zheng X, Wang M, Deng Y, Xu X, Lin D, Zhang Y, et al. A rapid, high-throughput, and sensitive PEG-precipitation method for SARS-CoV-2 wastewater surveillance. Water Res 2023; 230:119560.
[17]

Xiao A, Wu F, Bushman M, Zhang J, Imakaev M, Chai PR, et al. Metrics to relate COVID-19 wastewater data to clinical testing dynamics. Water Res 2022; 212:118070.
[18]

D’Aoust PM, Tian X, Towhid ST, Xiao A, Mercier E, Hegazy N, et al. Wastewater to clinical case (WC) ratio of COVID-19 identifies insufficient clinical testing, onset of new variants of concern and population immunity in urban communities. Sci Total Environ 2022; 853:158547.
[19]

Schmitz BW, Innes GK, Prasek SM, Betancourt WQ, Stark ER, Foster AR, et al. Enumerating asymptomatic COVID-19 cases and estimating SARS-CoV-2 fecal shedding rates via wastewater-based epidemiology. Sci Total Environ 2021; 801:149794.
[20]

Zheng S, Fan J, Yu F, Feng B, Lou B, Zou Q, et al. Viral load dynamics and disease severity in patients infected with SARS-CoV-2 in Zhejiang Province, China, January-March 2020: retrospective cohort study. BMJ 2020; 369:m1443.
[21]

Penn R, Ward BJ, Strande L, Maurer M. Review of synthetic human faeces and faecal sludge for sanitation and wastewater research. Water Res 2018; 132:222-40.
[22]

McMahan CS, Self S, Rennert L, Kalbaugh C, Kriebel D, Graves D, et al. COVID-19 wastewater epidemiology: a model to estimate infected populations. Lancet Planet Health 2021; 5(12):e874-81.
[23]

Ling Y, Xu SB, Lin YX, Tian D, Zhu ZQ, Dai FH, et al. Persistence and clearance of viral RNA in 2019 novel coronavirus disease rehabilitation patients. Chin Med J 2020; 133(9):1039-43.
[24]

Guo M, Tao W, Flavell RA, Zhu S. Potential intestinal infection and faecal-oral transmission of SARS-CoV-2. Nat Rev Gastroenterol Hepatol 2021; 18 (4):269-83.
[25]

Li Ka Shing Faculty of Medicine of The University of Hong Kong. Clarification on the interpretation of daily point prevalence of the HKU real-time community surveillance initiative [Internet]. Hong Kong: The University of Hong Kong; 2023 Jan 03 [cited 2023 May 25]. Available from: https://www.med.hku.hk/en/news/press/20230103-real-time-community-surveillanceinitiative.

[26]

Tsang NNY, So HC, Cowling BJ, Leung GM, Ip DKM. Effectiveness of BNT162b2 and CoronaVac COVID-19 vaccination against asymptomatic and symptomatic infection of SARS-CoV-2 omicron BA.2 in Hong Kong: a prospective cohort study. Lancet Infect Dis 2022; 23(4):421-34.
[27]

Lau JJ, Cheng SMS, Leung K, Lee CK, Hachim A, Tsang LCH, et al. Real-world COVID-19 vaccine effectiveness against the Omicron BA.2 variant in a SARSCoV- 2 infection-naive population. Nat Med 2023; 29(2):348-57.
[28]

Elliott P, Bodinier B, Eales O, Wang H, Haw D, Elliott J, et al. Rapid increase in Omicron infections in England during December 2021: REACT-1 study. Science 2022; 375(6587):1406-11.
[29]

Leung K, Lau EHY, Wong CKH, Leung GM, Wu JT. Estimating the transmission dynamics of SARS-CoV-2 Omicron BF.7 in Beijing after adjustment of the zero-COVID policy in November-December 2022. Nat Med 2023; 29(3):579-82.
[30]

Lin Y, Yang B, Cobey S, Lau EHY, Adam DC, Wong JY, et al. Incorporating temporal distribution of population-level viral load enables real-time estimation of COVID-19 transmission. Nat Commun 2022; 13(1):1155.
[31]

Carcereny A, Martinez-Velazquez A, Bosch A, Allende A, Truchado P, Cascales J, et al. Monitoring emergence of the SARS-CoV-2 B.1.1.7 variant through the Spanish national SARS-CoV-2 wastewater surveillance system (VATar COVID- 19). Environ Sci Technol 2021; 55(17):11756-66.
[32]

Ho J, Stange C, Suhrborg R, Wurzbacher C, Drewes JE, Tiehm A. SARS-CoV-2 wastewater surveillance in Germany: long-term RT-digital droplet PCR monitoring, suitability of primer/probe combinations and biomarker stability. Water Res 2022; 210:117977.
[33]

Agrawal S, Orschler L, Schubert S, Zachmann K, Heijnen L, Tavazzi S, et al. Prevalence and circulation patterns of SARS-CoV-2 variants in European sewage mirror clinical data of 54 European cities. Water Res 2022; 214:118162.
[34]

Medema G, Heijnen L, Elsinga G, Italiaander R, Brouwer A. Presence of SARScoronavirus- 2 RNA in sewage and correlation with reported COVID-19 prevalence in the early stage of the epidemic in the Netherlands. Environ Sci Technol Lett 2020; 7(7):511-6.
[35]

The Government of the Hong Kong Special Administrative Region. Government announces adjustments to quarantine arrangements for inbound persons [internet]. Hong Kong: The Government of the Hong Kong Special Administrative Region; 2022 Aug 08 [cited 2023 May 25]. Available from: https://www.info.gov.hk/gia/general/202208/08/P2022080800803.htm.

[36]

The Government of the Hong Kong Special Administrative Region. Government announces lifting of compulsory quarantine requirement on arrival at Hong Kong [internet]. Hong Kong: The Government of the Hong Kong Special Administrative Region; 2022 Sep 24 [cited 2023 May 25]. Available from: https://www.info.gov.hk/gia/general/202209/24/P2022092400048.htm.

[37]

Flacco ME, Acuti Martellucci C, Baccolini V, De Vito C, Renzi E, Villari P, et al. Risk of reinfection and disease after SARS-CoV-2 primary infection: metaanalysis. Eur J Clin Invest 2022; 52(10):e13845.
[38]

Puhach O, Meyer B, Eckerle I. SARS-CoV-2 viral load and shedding kinetics. Nat Rev Microbiol 2023; 21:147-61.

RIGHTS & PERMISSIONS

THE AUTHOR

PDF (2130KB)

5554

Accesses

0

Citation

Detail
Sections
Recommended

/

〈 〉