Five years ago, an outbreak of Legionnaires’ disease killed 12 people and sickened dozens more at the Illinois Veterans Home in Quincy.
Most of the cases occurred in the summer of 2015 at the state-run retirement facility — and investigations later revealed Legionella bacteria had thrived in its poorly maintained, aging plumbing system. But that same summer, four Quincy residents with no known connection to the veterans home contracted the disease in less than a month, puzzling investigators.
These unexplained illnesses caught the attention of researchers at Virginia Tech, who spent a year collecting and analyzing public records related to the outbreak.
The records they gathered show Quincy made major changes to its water treatment processes in the months leading up to the Legionnaires’ disease outbreak. Though the changes never violated state or federal law, the amount of disinfectant in Quincy’s tap water dropped by half as a result, possibly allowing Legionella bacteria to multiply throughout the water system.
Virginia Tech civil engineer William Rhoads had studied another Legionnaires' outbreak in Flint, Michigan, a now-infamous case in which municipal water treatment decisions fueled a public health crisis. And he wondered whether something similar might have happened in Quincy.
Treatment plants often add a combination of disinfectants to tap water to kill bacteria and viruses. Because a portion of disinfectant naturally degrades over time, utilities carefully calibrate their procedures to be sure enough is left to sanitize the water.
As Rhoads combed through reports from the initial investigation of the Quincy outbreak, he noticed a surprising detail: There was no disinfectant in tap water samples collected from the veterans home.
“At the time, they attributed that to the fact that the facility had been on water restrictions for six days,” said Rhoads, now a research scientist at the Swiss Federal Institute of Aquatic Science and Technology. “But I found it kind of curious that they hadn't investigated or reported on what was going on at the distribution system level.”
Digging into the data
Rhoads and his colleagues requested reams of public records, including routine water quality data collected at the Quincy treatment plant — which revealed the city substantially changed how it treated its water in the months before the Legionnaires' outbreak.
In December 2014, seven months before the first reported case, the Quincy plant temporarily switched its disinfectant from chlorine to chloramine, a less potent but more stable chemical. Within a month, the concentration of disinfectant in city water plummeted to 50% of what it had been in the same period the previous year.
Then a record storm in Quincy flooded the system with murky water on July 13, just one week before the Legionnaires' outbreak.
“As it rained, the flow rate of the Mississippi River increased and suspended a whole bunch of gunk into the water,” Rhoads said, adding that previous research has shown a spike in Legionnaires’ cases after severe rainfall events. “It really makes the water more difficult to treat.”
These “sudden and major changes” in water quality could have allowed Legionella bacteria to grow throughout the city’s water system and, in combination with the many documented issues in the veterans home's plumbing system, were probably “the last straw that pushed the facility over the edge,” he said.
Still, the connection between disinfectant levels in municipal water and Legionnaires’ outbreaks isn’t new. Previous research from another high-profile outbreak in Flint, which sickened at least 87 people, showed that the chlorine levels in city water fluctuated wildly during the outbreak.
“Some regions of the city had very low chlorine, and others had screechingly high levels,” said Michele Swanson, a professor of microbiology at the University of Michigan who has studied Legionella bacteria for 25 years. “You don’t have to have expertise in water management to recognize that something was wrong.”
Swanson and her colleagues found that for every 1 milligram per liter drop in chlorine concentration, the chances of seeing a case of Legionnaires’ disease increased by 80%.
‘A mini version’ of Flint
But the similarities between Quincy and Flint don’t end there.
Like Flint, the Quincy water treatment plant stopped adding a critical phosphate-based chemical used to prevent pipes from corroding in May 2015, roughly six months after switching disinfectants. Without it, pipes can rapidly dissolve and release heavy metals, like iron and lead, into the water.
The month after the treatment plant made the change, the blood lead levels of children in Quincy spiked, according to data from the Illinois Department of Public Health included in the Virginia Tech analysis. Of 65 children, 10.7% had high concentrations of lead in their blood — compared to an average of 2.2% over the previous decade.
Marc Edwards, a professor of civil engineering at Virginia Tech and study co-author, called the findings “very worrisome.”
“In every way, shape and form, what happened in Quincy is a mini version of what happened in Flint, in terms of the lead and Legionella and interrupted corrosion control,” Edwards said. “It's really hard for me to understand how this problem was not openly acknowledged and investigated.”
St. Louis Public Radio submitted multiple requests for comment to the Illinois Department of Public Health, but the agency did not respond.
Because Quincy had reported very low water lead levels in previous years, under the federal Lead and Copper Rule, the city qualified for reduced monitoring requirements. As such, the municipal water system did not collect any samples for lead testing when the treatment processes were changed.
“It's somewhat horrifying to think that corrosion control could be interrupted and you could have years of potentially very high lead in water,” Edwards said. “Not only is no one told about it, there's not any data collected to show what really happened.”
Unlike Flint, Quincy did not violate any laws when it changed its treatment processes. The water treatment plant switched back to using chlorine, the original disinfectant, in November 2015 and resumed phosphate corrosion control in February 2016.
It’s unclear why the treatment plant changed how it treated city water, though Rhoads speculates it may be connected to equipment upgrades that occurred during the time period.
For him, the situation demonstrates the shortcomings of current water regulations.
“It is completely normal for utilities to do what they did,” Rhoads said. “It really points out this gap in our regulatory structure that we can have these major changes at the water treatment facility, and not require any additional monitoring or communication with the community.”
Follow Shahla on Twitter: @shahlafarzan