Last week Allegheny County’s health department reported fewer than 100 new cases of COVID-19 a day. But the growing use of at-home testing coupled with an overall drop in people getting tested for COVID-19 makes it hard to know if this is truly a milestone.
A 100-foot deep well of raw sewage might hold the answers.
Skimming data
The sewage well is where wastewater for Pittsburgh and 82 other municipalities enter the Allegheny County Sanitary Authority, located on the banks of the Ohio River.
“There’s been no screening, filtering, no disinfectant,” explained Doug Jackson, ALCOSAN’s director of operations and maintenance. “This is as raw as the wastewater gets, coming into the treatment plant.”
COVID-19 is excreted in feces, and since this fall, the Allegheny County Health Department says it’s been analyzing ALCOSAN sewage for the virus.
Sample collection occurs in an adjacent building to the well. Wastewater flows into a container roughly the size of an extra-large cereal box. Every 10- to 15-minutes, a small amount of water is transferred to a 2.5-liter plastic bottle. Then the remaining liquid is emptied, and the process repeats so that the bottle slowly fills over 12 hours.
A commercial lab currently performs the actual analysis in Florida. The county health department’s Dr. Robert Wadowsky said it takes at least a week to get results on how much COVID-19 is in the water supply, then roughly another week to determine which variants are present.
Waiting this long diminishes the information’s value to act as an early warning system of potential surges. For example, the county only learned omicron was present in its wastewater a day or two before the first people tested positive for the variant, which is now dominant in the county.
Additionally, current methods cannot differentiate between subvariants and therefore can’t give data on omicron’s sister strain, which is driving up cases in other parts of the world. While individual Allegheny County residents have tested positive for the subvariant, both wastewater data and raw case numbers show that local transmission continues to decline.
Still, wastewater monitoring is a powerful tool, which is why the health department wants to start doing this analysis in-house.
“We have the technique ready to go, just about … but it is so difficult for us to hire staff to work in the laboratory,” said Wadowsky, who is the director of the department’s public health laboratory.
The county is collaborating with researchers at Carnegie Mellon University to perform that analysis and create an online dashboard for the public to view the data, which Wadowsky predicts will launch later this spring.
What’s reflected in the wastewater?
According to data from the county health department, in early March, when the county was reporting about 100 cases a day, the presence of COVID-19 in the wastewater was slightly below where it was around Halloween. At the time, the daily case counts exceeded 300.
While testing numbers and wastewater data don’t perfectly align, the analysis follows the same general pattern of hospitalizations in the county with a significant spike that peaked in early January, followed by a steep and rapid decline. This is significant because while only a fraction of people who catch the coronavirus will become seriously ill, the percentage of those who develop severe COVID-19 is relatively stable.
Sampling sewage for public health surveillance isn’t new – it's been done for opioids, HIV, and antibiotics, to name a few. But it became more common during the pandemic, and last month the Centers for Disease Control and Prevention launched a dashboard tracking COVID-19 in wastewater at more than 400 sites. There are so many applications because it captures populations other methods miss.
“People that don’t have access to care, that don’t have insurance and therefore don’t go to the doctor, and therefore can’t be counted through traditional hospital-based or clinic-based surveillance, they still flush the toilet,” said epidemiologist Matt Ferrari, director of Penn State’s Center for Infectious Disease Dynamics.
All those bathroom breaks create lots of data, which means greater sensitivity to subtle changes in transmission. For example, in areas where polio has been nearly eliminated, wastewater analysis can tell public health workers if there are cases they don’t know about.
“And that then alerts them to, ‘OK, we’ve seen it in wastewater,’” said Ferrari. “Now we’ve got to go look harder in people.”
Beyond COVID-19
Wadowsky admits that he was initially skeptical that the wastewater analysis would be effective. That changed when he saw the consistency of the data.
“I thought for sure that there was going to be so much fluctuation,” he said. “So, really, I was quite amazed.”
For that reason, he sees other eventual applications, such as monitoring for an enterovirus that causes acute flaccid myelitis – a rare but serious polio-like illness most commonly seen in young children. The CDC reports there have only been 679 confirmed cases since August 2014. Just between 20-30% of patients make full recoveries.
In the meantime, the county will continue watching for signs of COVID-19 surges in the sewage. For the moment, the wastewater looks calm.
