Case Study — Wastewater Epidemiology
Deploying Wastewater Epidemiology Before the Federal System Existed
May 13, 2020 — four months before the CDC's National Wastewater Surveillance System — Purcellville became one of the first municipalities in the U.S. to track COVID-19 through sewage.
The Problem: Clinical Testing Was Not Enough
When COVID-19 arrived in Virginia in March 2020, Purcellville faced the same public health challenge confronting every U.S. municipality: understanding the actual prevalence of the disease in the community when clinical testing capacity was severely constrained. In the spring of 2020, Virginia's clinical testing was limited, turnaround times were long, and the tested population was biased toward symptomatic individuals — meaning confirmed case counts significantly undercounted actual infection levels.
Mayor Kwasi Fraser's administration identified wastewater-based epidemiology as a mechanism to fill the surveillance gap that clinical testing could not fill. The challenge: no federal infrastructure existed for this approach. The CDC's National Wastewater Surveillance System would not launch until September 2020, four months later. Any program Purcellville deployed would need to be assembled independently from academic and commercial partners who were themselves still developing the methodology.
The Partners
Biobot Analytics
Founded by MIT researchers specializing in wastewater epidemiology — provided the laboratory analysis infrastructure and mathematical modeling required to translate raw wastewater samples into estimated active case counts. In spring 2020, one of the few organizations in the United States with the capacity to perform this analysis at scale.
MIT and Harvard University
Academic institutions whose researchers had developed and validated the wastewater epidemiology methodology — provided scientific oversight and peer validation of the approach.
Brigham and Women's Hospital
A Harvard Medical School teaching hospital — provided clinical expertise in interpreting the relationship between wastewater viral load and actual community infection prevalence.
The Deployment
The first wastewater sample was collected in Purcellville on May 13, 2020 — eight weeks after the national emergency declaration and two months into the pandemic's first wave. The sample was drawn from a strategic point in the town's sewer network designed to capture representative flow from the residential and commercial population. Purcellville's sewer system serves approximately 9,000 residents — near the lower bound of sewer systems for which wastewater epidemiology was being validated. The deployment in a small system contributed to the national evidence base that the field needed: proof that WBE was viable below the scale of major urban systems.
What the Data Showed
The May 13 sample estimated approximately 50 active COVID-19 cases in Purcellville — a figure town officials could use for local decision-making at a time when clinical case counts were lagging and undercounting actual prevalence.
Subsequent samples tracked a surge: estimated active cases climbed to approximately 320 at a mid-May peak — before clinical case confirmation caught up with the wastewater signal by several days. The predictive gap between the wastewater estimate and the eventual clinical confirmation validated the core premise: wastewater provides a leading indicator that clinical testing cannot.
The Broader Significance
For Purcellville
The program gave the Fraser administration timely, actionable public health intelligence during a period when municipal governments were making consequential decisions — about facilities, events, and emergency operations — with limited data. A mayor who can see a surge coming 3–7 days before clinical testing confirms it is a mayor who can make earlier, better-calibrated decisions.
For the field
The Biobot Analytics / MIT / Harvard partnership that Purcellville participated in was building the national evidence base for wastewater epidemiology as a routine public health tool. Purcellville's small-system deployment demonstrated that the approach worked at 9,000 residents — a scale the academic literature had not yet validated. The CDC's NWSS eventually scaled to hundreds of monitoring sites nationally, building on the methodological validation that early deployments like Purcellville's contributed.
For municipal officials
The program demonstrated that a town of 9,000 residents — with a motivated mayor, a willing utility operations team, and access to the right partners — could deploy a novel public health surveillance tool in a crisis without waiting for federal infrastructure. The implementation complexity was manageable for a municipality of Purcellville's size.
Key Facts at a Glance
- Problem: clinical testing too limited to measure actual COVID-19 prevalence
- Solution: wastewater-based epidemiology with commercial/academic partners
- Partners: Biobot Analytics; MIT; Harvard University; Brigham and Women's Hospital
- First sample: May 13, 2020 (eight weeks post-national emergency)
- Initial estimate: ~50 active cases
- Mid-May peak estimate: ~320 active cases
- Predictive lead time: several days ahead of clinical confirmation
- System size: ~9,000 residents (near lower bound for WBE at the time)
- Federal context: CDC NWSS launched September 2020 — four months later