Fact check: Did North Carolina voting sites have any clusters of COVID-19 cases?
The issue:
Karen Brinson Bell, executive director of the North Carolina State Board of Elections, said Nov. 24 that no spread of COVID-19 had been linked to voting in North Carolina. The NCSBE also said in a news release that there have been zero clusters of COVID-19 connected to voting sites in North Carolina. Are these claims accurate, and what counts as a “cluster”? Does that mean there weren’t any coronavirus cases connected to voting at all?
Why we’re checking this:
During the coronavirus pandemic, people have been worried about voting safety. It’s the reason a historic number of absentee mail-in ballots were cast this year in North Carolina — 1 million, the NCSBE reports. It’s also the reason officials took precautions, like providing masks and hand sanitizer at polling sites for those who did not bring their own.
It’s time to see if those precautions worked.
“Election workers pulled off a successful election with the highest voter turnout in North Carolina history and a staggering increase in absentee-by-mail voting,” according to a release from the elections board. “They did so while protecting themselves and millions of voters from a deadly virus, ensuring that no COVID-19 clusters were tied to voting sites in North Carolina.”
But we’ve seen a rise in coronavirus cases in recent months, so we wondered if voting could have been a factor in that rise.
What you need to know:
The NCSBE stating there haven’t been any clusters doesn’t mean that there have been zero coronavirus cases connected to voting sites. There have been at least three cases, according to county health departments.
In Rockingham County, a Reidsville polling place shut down for a few hours Oct. 26 and reopened the next day, but was then completely shut down the following three days because of the coronavirus. Three poll workers tested positive, said Janet Odell, the interim Rockingham County Board of Elections director. Twenty other poll workers were also exposed to those three people and were told to quarantine for 14 days.
In May, the North Carolina Department of Health and Human Services defined “cluster” in COVID-19 terms, according to a memo. In community, occupational and educational settings, a cluster is defined as:
- “A minimum of five cases with illness onsets or initial positive results within a 14-day period AND
- Plausible epidemiologic linkage between cases.”
DHHS goes on to explain that having a “plausible epidemiologic linkage between cases” means that “local health departments should verify that cases were present in the same setting during the same time period (e.g., same shift, same classroom, same physical work area), that the timing fits with likely timing of exposure, and that there is no other more likely source of exposure for identified cases (e.g., household or close contact to a confirmed case in another setting).“
Polling places would fall under community settings. But Rockingham’s COVID-19 cases at the Reidsville polling place wouldn’t be considered a “cluster” because only three people tested positive.
With that said, the definition of a cluster varies.
In October, the Centers for Disease Control and Prevention stated that for workers in a non-healthcare setting, a cluster is defined as “two or more confirmed or probable cases among workers within a 14-day period.” Under that definition, the Rockingham cases would be included.
But it’s likely the North Carolina elections board was following the state’s health department definition.
Patrick Gannon, the NCSBE spokesperson, referred The News & Observer to the state Department of Health and Human Services when asked to elaborate on the board’s statements regarding COVID-19.
In an email, Conner wrote that the N.C. health department’s weekly report — which NCSBE used — “includes clusters identified by local health departments but under-represents the full scope of clusters and associated cases occurring across the state. In part, this is due to the fact that many COVID-19 cases are never diagnosed or cannot be linked to a specific setting or specific cluster.”
“While congregate living settings, schools and child care facilities are required to report clusters or outbreaks to their (local health departments), other settings are not required to report in the same manner,” Conner continued. “Thus, the cluster data provided in this report is limited to clusters that have been voluntarily reported to LHDs or identified through case investigation and contact tracing efforts.”
Julie Swann, an N.C. State professor of industrial and systems engineering and former CDC adviser, considers the election board’s statements to be accurate based on the definitions provided by state public health officials. She said the key word in Bell’s statement, “There has been no COVID-19 spread linked to voting in North Carolina” might be “voting.”
Swann said the use of that word could suggest the board is referring to voters and the voting process, rather than poll workers.
“I would take that statement to mean that there’s no evidence that poll workers transmitted the virus to voters in North Carolina, or that voters transmitted the virus to each other, because that is the ‘voting process,’” Swann says. “It doesn’t necessarily mean there was no spread, it just didn’t meet their definition for reporting it.”
Swann also said it’s not necessarily clear that there was spread within the voting site in Reidsville. If the poll workers socialized outside of work and spread it there, Swann would not consider that to be a spread linked to voting.
Beyond what happened in Reidsville, The News & Observer was not able to determine if there were other cases of COVID-19 at polling sites. A DHHS spokesperson said the agency isn’t aware of any, but that only schools, child care and congregate living settings are required to report cases to health departments.
BEHIND THE STORY
MOREOur process
We check claims that are widely shared or published; are about a topic of concern to many of our readers; can be proven or disproven through facts; and could cause people to act or vote in a certain way. This topic met all the criteria.
Send comments and suggested fact-checks using this form.
Find out more about our process here. And as always, we abide by our newsroom’s ethical guidelines.
Our sources. Here’s where we found information and research on this topic:
Email from DHHS spokesperson Kelly Haight
CDC definition of a COVID-19 cluster in a non-health care worksite
CDC definitions of Epidemic, Cluster, and Outbreak
NCSBE press release claiming zero clusters
NC Department of Health and Human Services: Outbreaks in Congregate Living Settings
CDC Guidelines for Investigating Clusters of Health Events
News & Observer article on NCSBE executive director speaking about Covid
News & Observer article on Rockingham County polling place cases
This story was produced by The News & Observer Fact-Checking Project, which shares fact-checks with newsrooms statewide. It was edited by Jordan Schrader and Jessica Banov. Submit a suggestion for what we should check, or a comment or suggestion about our fact-checking, at bit.ly/nandofactcheck.
This story was originally published December 4, 2020 at 11:23 AM with the headline "Fact check: Did North Carolina voting sites have any clusters of COVID-19 cases?."