Health care workers screen a patient for COVID-19 at a drive-through coronavirus testing site on March 18, 2020 in Arlington, Virginia. Fewer patients are relying on PCR tests, preferring at-home antigen tests whose results aren’t typically reported back to the state. (Photo by Drew Angerer/Getty Images)
Earlier this week, the Virginia Department of Health updated its public reporting on breakthrough COVID-19 cases to “more accurately represent the impact of vaccinations on infection rates in the commonwealth,” according to a release from the agency.
As the Mercury reported, the data illustrates the higher risk for infection, hospitalization and death among unvaccinated Virginians. But the way it’s presented makes it tough to calculate the prevalence of breakthrough infections in the weeks since the Delta variant began to account for the vast majority of cases across the country.
In other words, VDH isn’t reporting the percentage of breakthrough cases out of all total known infections week-by-week. That makes it hard to know exactly how many of those cases are occurring among immunized Virginians amid the rise of a highly infectious variant that’s been shown to reduce the effectiveness of available vaccines.
Even without that data, though, recent state reporting has seemed to suggest that Virginia’s breakthrough infection numbers are curiously lower than other states — and even localities within the commonwealth. Before VDH changed its dashboard, the department was reporting that less than two percent of all reported COVID-19 cases between January and early August were among vaccinated Virginians. At the same time, some local health departments were reporting numbers that would ostensibly drive up that total.
Prince William, for example, announced that between 25 to 30 percent of its recorded cases throughout the first half of July were breakthroughs.
“And I don’t think Prince William is an outlier,” District Epidemiologist Sean Morris told the Mercury. But even national reporting has described Virginia as strangely immune to infections among vaccinated residents. Last week, The New York Times reported that data from seven states — including Virginia — reflected a rise in breakthrough cases. Virginia was the “outlier” of the bunch with 6.4 percent of its recorded infections among the fully vaccinated. In the other six states, they accounted for 18 to 28 percent of total cases in recent weeks.
It turns out the discrepancy has less to do with Virginia’s good luck and more to do with a likely undercount of breakthrough cases. And VDH officials say that determining the actual prevalence of breakthrough cases from week to week is “discouraged.”
‘HARDER TO VERIFY’
Prior to Aug. 19, VDH relied on laborious manual reporting to confirm an infection occurred in a fully vaccinated person. According to Dr. Julia Murphy, VDH’s state public health veterinarian (who, in the midst of the pandemic, is also assisting with the state’s overall epidemiological response), multiple things had to happen before a breakthrough case could be identified.
“First, the person needs to be tested for COVID-19 and test positive,” Murphy wrote in an email. Once a positive case is registered with VDH, it’s assigned to a local case investigator. Before last week, that investigation was an important — and sometimes critical — part of verifying a breakthrough infection. The investigators are tasked with asking more than two dozen questions to help determine the severity of a case and potential exposures — including whether the patient was vaccinated.
The details of a patient’s vaccination status are important. For an infection to count as a breakthrough, there needs to be proof the person completed a full vaccine series (two doses in the case of the Pfizer and Moderna vaccines and one in the case of Johnson & Johnson) in the time frame recommended by federal health officials. There also needs to be confirmation that the patient completed the series at least 14 days before the positive test.
In Virginia, public reporting includes infections among partially vaccinated people, as well. Confirming those cases requires verification that the patients hadn’t completed a full vaccine series, completed it outside the recommended time frame, or got sick less than 14 days after they received their full vaccine course, the time it takes for someone to be considered “fully vaccinated.”
Positive COVID-19 test results are reported in VEDSS, the Virginia Electronic Disease Surveillance System. Before Aug. 19, though, there was no integration between VEDSS and VIIS, or the Virginia Immunization Information System, where vaccinations are recorded.
“The process of comparing data in these systems used to be performed manually by a COVID-19 case investigator,” Murphy wrote. If those investigators couldn’t reach a patient to confirm vaccination status — which happened around 35 percent of the time in the last week of July, according to VDH data, and roughly 53 percent of the time from Aug. 13 to 19 — they would have to find that person in the state’s vaccination registry and manually transcribe the information into its disease surveillance system.
And while some local health districts took that step, it didn’t always happen.
“If a case investigator could not contact a case, that case’s vaccination status information was harder to verify and, therefore, may not have been captured as a breakthrough,” Murphy wrote.
For much of July and August, there were other factors complicating the reporting. The lack of automation made vetting fully vaccinated cases more challenging at the state level. Before the two systems were integrated, VDH’s central office in Richmond manually reviewed every identified breakthrough infection before reporting it on the public dashboard. Murphy said that contributed to the data lag and the apparent discrepancies between state and local data.
“Now that cases are being automatically cross-referenced between VIIS and VEDSS, more cases will be reported that have not been individually reviewed by Central Office,” she wrote. “The data will more closely match what local health departments are reporting.”
‘THERE MAY BE LESS OF A MARGIN FOR ERROR’
But even now, it’s likely breakthrough cases are still being underreported. The majority of those infections are mild or asymptomatic, making it less likely a patient will seek out testing. And at this point in the pandemic, the market is inundated with at-home testing options, which generally aren’t passed on to VDH unless patients disclose their results to a medical provider or local health department.
It’s nearly impossible, then, to confirm exactly how many breakthrough cases occurred in Virginia over the last month and a half, as Delta became the dominant variant across the country. From the available data, it’s clear rates of known infections among the fully vaccinated have nearly quintupled between early April — when vaccines became widely available — and Aug. 14, the state’s most recently available data. VDH’s new public dashboard also includes the total number of known breakthrough cases, but it’s not clear how they compare to all new COVID-19 infections from week to week.
Trying to calculate that prevalence is “discouraged,” Murphy wrote, “for several reasons.” As the Mercury reported earlier this week, state health officials have been concerned that a growing focus on breakthrough cases will minimize the success and importance of vaccines. Dr. John Swartzberg, a clinical professor emeritus of infectious diseases and vaccinology at the University of California, Berkeley School of Public Health, pointed out that an “infection” doesn’t correlate with actually getting sick.
GET THE MORNING HEADLINES DELIVERED TO YOUR INBOX
“Infection just means you have the virus that’s replicating in you,” he said. “It doesn’t say anything about whether you have symptoms or not.” Vaccines that prevent the development of severe disease, including hospitalization and death, are a public health success. And currently available vaccines are still so successful at preventing those outcomes that some public health experts argue booster shots aren’t really necessary.
As more and more people get immunized, and cases continue to rise, Murphy pointed out that a higher number of breakthrough infections is an expected outcome. In an interview earlier this month, state epidemiologist Dr. Lilian Peake gave the example of a 2006 mumps outbreak at the University of Virginia. The college required students to be immunized, so 100 percent of the cases were in a fully vaccinated population. But the high level of vaccination still meant very few people contracted the virus.
“It was a highly effective vaccine, but once you have a high proportion of the population vaccinated, the data is skewed,” she said. Still, some public health experts have criticized the overall lack of data on breakthrough infections, especially when it comes to mild cases. For researchers, knowing the true number of infections is important for parsing out the reasons why they’re occurring and whether some people are more vulnerable than others. But obscuring those numbers can also minimize the risk of transmission for vaccinated people.
“I think what we’re seeing is that viral loads can get very high with Delta,” Dr. Michael Mina, a Harvard epidemiologist, told New York Magazine earlier this month. “The idea that we’re going to vaccinate our way to true herd immunity — that idea has to be put to bed for a moment.”
And while Virginia’s data highlights that the relative risk of infection, hospitalization and death is much greater for unvaccinated residents, prevalence can help underscore the importance of continued mitigation measures like mask-wearing, social distancing and continuing to exercise caution about potential exposures.
“We know that if you get vaccinated, you prevent yourself from the worst outcomes,” Morris said. “But there may be less of a margin of error than with less transmissible versions, which means you really have to be on top of it and make sure you’re doing what’s recommended.”
SUPPORT NEWS YOU TRUST.
Our stories may be republished online or in print under Creative Commons license CC BY-NC-ND 4.0. We ask that you edit only for style or to shorten, provide proper attribution and link to our web site. Please see our republishing guidelines for use of photos and graphics.