Where is the Minnesota serology survey?

When it comes to knowing how many of us have gotten COVID-19, the state still faces a glaring blind spot. Why so slow?

Photo: Amornthep Srena/Pexels
We are part of The Trust Project.

ROCHESTER, Minn. — It seems like the first question a state health department would want answered: Just how many of us have had the coronavirus, anyway?

Called a seroprevalence survey, it measures how many Minnesotans have gotten the virus and developed antibodies, and a person could be forgiven for asking why it's taking so long.

Think about it like this: here at the start of July, we know a lot more about COVID-19 than we did back in February. We know about its strange mix of symptoms: fever, low oxygen, dry cough and losing your ability to taste.

We know how COVID-19 spreads fastest: extended, loud conversation and singing at bars, parties, choir practices, funerals and other sustained indoor contact.

We know what prevents the spread of COVID-19 best (masks, handwashing and distancing).


We know COVID-19 sometimes has no symptoms, that the aymptomatic can shed the most virus of anyone, and who it hits the hardest: the elderly, those with underlying health problems and possibly the healthy who have contracted high doses of the virus.

Rural workers have learned that living in wide-open spaces won't protect you if you work in close quarters, and urban dwellers have learned they can put on a mask, protest in the open air and come away fine.

But we haven't learned how many of us have already had it.

This is strange, because the answer would tell us a lot. Moreover, it's not a complicated question to answer, we have the tools to do it and we have had them for several months.

In theory, a basic COVID-19 seroprevalence study should take about a week, two at most. You recruit a representative sample of 500-1000 statewide residents. You send phlebotomy technicians out to their homes. You draw vials of blood, pay them for their time, and test it all for the virus.

But you wouldn't know this by remarks last Monday, June 29, as Gov. Tim Walz was joined by representatives of Mayo Clinic and the University of Minnesota labs to celebrate the state having reached its target capacity to process 20,000 diagnostics tests a day.

Asked about serology testing, state Commisioner of Health Jan Malcolm deferred to Mayo and University of Minnesota, adding that "they have been doing that serology testing under this partnership."

First up was Dr. Timothy Schacker, vice dean for research at the University of Minnesota Medical School.


"At this point in time, a lot of the antibody testing that's been done in Minnesota has been done on select populations," said Schacker. "We don't have a lot of data to say what the seroprevalence is across the entire state."

"I can tell that you in our serology testing we have focused on health care workers and we have focused on just general seroprevalence in the Twin Cities," Schacker added, "and it's come out between four and five percent. And that's about what you'd expect given the number of positive tests in the state."

Few would characterize studies of health care workers and metro-area residents as meaningful, however, for gauging the spread of COVID-19 statewide. Mayo Clinic has a ready-made representive seroepidemiology database in the Olmsted Epidemiology Project, but seemingly even less information about the spread of COVID-19 at this time.

"I do think it's (still) the very early days with serological testing to really draw conclusions," said Dr. William Morice. "It will be part of the discussion as we go forward, how will serological testing be used to help manage the pandemic, now that we are in the management phase."

In May, a group of researchers from Stanford conducted serosurveys in northern and southern California, and found the virus was far more widespread than believed, suggesting it lethality was overstated. The former was subjected to critcism for analytical reasons, but few have offered to produce counter-surveys.

For the next 12 months, an ongoing CDC seroprevalence study is making monthly checks of donor blood for COVID-19 antibodies, sampling from 1,000 donors in each of 25 metro areas. That investigation will be broadened to 25,000 donors at the 18-month mark. Minneapolis is one of these cities.

MDH running, preparing 5 serosurveys

MDH spokeman Doug Schultz says that five serology studies utilizing finger-prick blood testing are now being funded with $2 million from the state’s COVID response fund and from the CDC. Schultz said the state prioritized blood-prick technology in order to broaden their sample size, but that waiting for this technology to come online had slowed the start time.

Schultz said the MDH received serosurvey proposals from the U of M in late April, however, and began planning the studies in early May while waiting for state COVID relief funds to be approved.


With the state money now in hand, these have included a bloodbank study, and a so-called CASPER field methodology study (Community Assessment of Public Health Emergency Response). Both are to be conducted by MDH personnel who will be reassigned from other positions.

Schultz said the bloodbank survey is the simplest, because it uses leftover blood from donors. The samples are de-identified — the state will receive some demographics with them, but they are not linked to cases. "The bloodbank is already pulling specimens," Schultz said. "We will look at 1,000 samples now, 1000 in fall and 600 of these from metro, with 400 from the northland. We will repeat that in fall."

During a press call Wednesday, July 1, state epidemiologist Dr. Ruth Lynfield reported that the June specimens collected showed that 1.2% of those in the blood bank had antibodies for COVID-19. "I might have expected it to be a little bit more," Lynfield said, "but 1.2% for a group of healthy adults is plausible."

"As we study other communities who may have had more exposure to the virus, I think we will see an increase in that number. I think it is great there have been people who have been uninfected, but it also means we have a large population of people who are still at risk."

Bloodbank sample studies however are limited in usefullness as they are not representative of the broader population, which is why the CASPER data will likely prove more useful.

"CASPER is labor-intensive," Schultz says. "It will involve six regional studies, 30 sites within each region, 7 households in each site," for 210 households in total. Schultz said that to do each of the six regions will require 15 teams of two people.

As for the work now being undertaken by the University of Minnesota, Schultz cited the aforementioned study of health care workers, as well as studies of grocery store workers, adding that the U of M is taking part in a national survey through the University of Chicago. He knew of no Mayo seroprevalence studies using state funds.

"Just because it takes some time to get projects like this going," Schultz said, "does not mean they are not a high priority. In fact, the opposite is true: We want these studies to yield good, useful results, so they need to be done right. More time spent on the front end assures fewer problems, and better quality on the back end."

"Doing this well, so that the results will be useful and meaningful, it just takes more time than running out and bleeding 500 random people over a weekend. We don’t want to repeat the mistakes made by the Stanford study."

Determining whether COVID-19 has infected 5%, 10% or even 30% of the state would tell us how many people can get it and recover, and how deadly the disease really is. On top of that, if it turns out that COVID-19 antibodies confer long-lasting immunity — a big if — a seroprevalence study would also tell us how far we are on our way to herd immunity.

"To have herd immunity we like to see it more at 60%-70%," Lynfield said, "so we have a long way to go to get to herd immunity." Topping that off, "we don't know enough about how long immunity lasts with this virus," Lynfield said. "We think it is not going to be as long-lasting as something like measles, where you have lifetime immunity."

"Based on similar coronaviruses, the immunity for COVID-19 may last for a period of months to a couple of years. ...This virus is going to be with us for a while. I know everybody is sick and tired of prevention measures, and want to get bak to normal life, but it's going to be with us for a while."

Paul John Scott is the health correspondent for NewsMD and the Forum News Service. He is a novelist and was an award winning magazine journalist for 15 years prior to joining the FNS in 2019.
What to read next
When those first baby teeth appear, it's time to start teaching little ones about good dental health. In this episode of NewsMD's "Health Fusion," Viv Williams consults a pediatric dentist about when kids should have their first dental appointment and she shares tips on brushing.
Long road trips provide ample time for both reflection and rumination — the good and the bad of hours and hours spent behind the wheel. In this Health Fusion column, Viv Williams shares stories of a recent drive to Colorado and how a pit stop at a botanical garden's butterfly house made a faulty air conditioner tolerable and brought meaning to the buzz word "mindfulness."
The new report, released this morning, showed a rise among the state's ninth-graders battling long-term mental and emotional problems
When you sprain your ankle or have an infection inflammation helps to heal tissues. But when inflammation is chronic, or long term, it can contribute to conditions such as heart disease and autoimmune diseases. Researchers have found a link between chronic inflammation and low levels of vitamin D. Viv Williams has details in this episode of NewsMD's "Health Fusion."