Wastewater-integrated pathogen surveillance dashboards enable real-time, transparent, and interpretable public health risk assessment and dissemination

Abstract

Timely pathogen surveillance and reporting is essential for effective public health guidance. Web dashboards have become a key tool for communicating public health information to stakeholders, health care workers, and the broader community. Over the SARS-CoV-2 pandemic, wastewater surveillance has increasingly been incorporated into public health workflows for outbreak monitoring and response, enabling community-representative and low-cost monitoring to supplement clinical surveillance. However, the methods used for visualization and dissemination of clinical and wastewater surveillance data differ across programs, and best practices are yet to be defined. In this work, we demonstrate data workflows and dashboards used to perform wastewater-based public health surveillance in tandem with clinical data across local and national scales, leveraging custom-built, reproducible, and open-source software. Using a centralized data aggregation and analysis hub approach, we establish multiple data pipelines for data storage, wrangling, and standardized analyses, and deploy custom-built web dashboards that allow for immediate public release. We find that our approach is effective across scales, computing architectures, and dissemination strategies, and provides an adaptable model to incorporate additional pathogens and epidemiological data.

Fig 1. Distributed data analysis and multi-source integration.

A. For the SEARCH dashboard, separate analysis pipelines are used for wastewater sequencing data and viral load measurements, but both are uploaded to a public data repository, triggering automated recalculation of lineage growth rates. Clinical case data are pulled from county data resources, integrated within plotting modules, and inserted individually into the SEARCH website. B. For the NICD dashboard, viral loads in wastewater samples are first quantified to determine if they can be sequenced. The results are added to a RedCap database. Sample sequences are analyzed with Freyja and uploaded to a public data repository alongside clinical case counts obtained through the NICD’s surveillance data warehouse. Data from the public data repository is pulled by the dashboard when loaded by the user.

Fig 2. SEARCH wastewater dashboard components.

A. Zip code-level mapping of catchment areas for each San Diego wastewater treatment plant. B. Combined visualization of wastewater viral loads and catchment specific-case counts. C. Lineage prevalence tracking for each catchment, with sub-lineages grouped with parent lineage. D. Relative growth advantage estimates for each lineage grouping, using circulating lineages as background. Zip code boundary map is public domain and available from San Diego County’s Open Data Portal[25], https://geo.sandag.org/server/rest/directories/downloads/Zipcodes.geojson.

Fig 3. NICD national wastewater dashboard core components.

A. Data cards provide information on the current epidemiological week as well as the clinical case counts and wastewater samples collected. B. SARS-CoV-2 viral load compared to the number of clinical cases reported each week, with interactive hover to see exact values in the chart. C. Lineage prevalence over time, with a toggle option to switch between smoothed and binned data.

Fig 4. NICD wastewater dashboard district-level components.

A. Users can select a district of interest (highlighted in green), and the corresponding WWTPs within that district are highlighted in red. B. The viral loads for each WWTP in the district, along with district-level case counts, are shown over time. C. Lineage prevalence tracking for each WWTP, resolved by date of sampling. Province shapefiles were obtained under an open CC-BY 4.0 license [26], https://simplemaps.com/static/svg/country/za/admin1/za.json.