Georgia was the first state to start reopening its economy after shutting down due to the coronavirus pandemic. When the shelter-in-place order expired on April 30, a lot of experts and much of the public worried about the worst: a sustained spike in Covid-19 cases that would overwhelm emergency rooms and lead to a surge in deaths.
Yet more than a month later, the worst hasn’t arrived.
Even as the state has increased its testing capacity, boosting its ability to pick up new coronavirus cases, the total of daily new cases has remained relatively flat, despite some ups and downs. The state isn’t getting significantly better — dozens of Covid-19 deaths are still reported each day in Georgia — but it doesn’t seem to be getting much worse either.
It’s possible this data doesn’t tell the full story — maybe the state is underreporting or even manipulating the data to look better, which Georgia has already been caught doing at times.
But if that were the case, coronavirus would still show up in hospitals — by way of sick patients. Based on data from the Centers for Disease Control and Prevention taken from hospitals, that doesn’t seem to be happening either: With about 6 percent of inpatient beds occupied by Covid-19 patients, Georgia is, again, not doing great. But it’s not in the 10 worst states nationally, with several states that started to reopen later ahead.
I turned to experts with one question: Why isn’t Georgia, at least so far, experiencing the worst-case scenario that some expected?
Most experts agreed that it’s important not to conclude too much from the current data. It’s still early in the process, given the coronavirus’s incubation period and how long it can take to spread. It’s still possible that things will get worse.
“What I and others were saying back at the end of April is we’re not going to really see anything until June,” Eleanor Murray, an epidemiologist at Boston University, told me. “I think that’s still the case.”
There’s also the real possibility of data manipulation. Georgia has not proven itself to be above this. Until recently, it was including a type of test into its test count that experts say shouldn’t be included. It has also, for example, published a chart that switched dates in a way that, deliberately or not, made the state’s data look more favorable.
“It is very difficult to sort out where we are in Georgia,” Melanie Thompson, a principal investigator at the AIDS Research Consortium of Atlanta, told me. “We have some confusing issues with our data and the way they are reported by the state of Georgia on their website.”
Aside from that, experts said other factors could play a role. Maybe people are staying home even though the shelter-in-place order is over. Perhaps masks and other practices people have adopted as a result of the pandemic, like washing our hands more thoroughly, are playing a big role in driving down transmission even as people go out more. It’s possible the warmer weather could help — or there could be some other factor we don’t know about, even just luck, that could play a driving role.
More than showing whether government-mandated lockdowns are necessary to suppress the coronavirus, Georgia’s experience so far seems to demonstrate how little we still know about this virus warping all of our lives. How things progress from here, though, could teach us a lot about Covid-19.
1) It’s too early
It’s been a month since Georgia’s shelter-in-place order. In theory, that gives enough time for the virus to spread and incubate in people.
But life doesn’t always work like theory, and there are several reasons new coronavirus cases could take longer than two weeks — maybe even months — to start going up after a government ends its shelter-in-place order.
People have to go out more. They have to get infected, and, typically, symptoms have to show. They have to get tested for the virus. That sample then has to be processed before it’s reported as a completed test. That test has to be reported to Georgia’s government. The state then has to report the test results.
Deaths are even more delayed, since they take place up to weeks after the initial infection. “If a whole lot more people are being infected on one day, and I’m monitoring the death data, I’m not going to see that uptick in cases for maybe a month or more,” Lauren Meyers, a mathematical biologist at the University of Texas at Austin, told me.
And even if cases do start to pick up at any particular point, a full outbreak can take time to get going. Daily reported cases in Georgia initially sat below 100 for much of March before quickly shooting up to the hundreds through late March and early April.
The result is we may not know how the reopening has affected Georgia for weeks or even months after it began.
2) There might be some data manipulation
At the least, one can say that the Georgia Department of Public Health has had some major glitches with its coronavirus data reporting. At the worst, there could be some manipulation going on.
As Willoughby Mariano and J. Scott Trubey reported for the Atlanta Journal-Constitution, the state in May posted a chart that seemed to show that “new confirmed cases in the counties with the most infections had dropped every single day for the past two weeks.” Upon closer inspection, the chart was extremely misleading — displaying two Sundays in one week and putting data from May 2 before data for April 26. It seemed like an attempt to create a downward slope where there wasn’t one.
Georgia officials later apologized for the original version of the chart.
But this wasn’t the first or last time that Georgia got caught making a mess of its data. Mariano and Trubey noted in the Atlanta Journal-Constitution, “This unforced error — at least the third in as many weeks — is confounding observers who have noted sloppiness in case counts, death counts and other measures that are fundamental to tracking a disease outbreak.”
More recently, Georgia was caught including the wrong kind of test in its test count.
According to experts, the main test for the daily counts should be diagnostic tests. Those gauge whether a person has the virus in their system and is, therefore, sick right at the moment of the test. Antibody tests check if someone ever developed antibodies to the virus to see if they had ever been sick in the past. Since diagnostic tests give a more recent gauge of the level of infection, they’re seen as much more reliable for evaluating the current state of the Covid-19 outbreak in a state.
“We need to understand that there is a new case of a new disease happening in our community,” Pia MacDonald, an epidemiologist at the research institute RTI International, told me. “There are public health interventions that need to happen around that.”
But Georgia was among several states including antibody tests in its count. Only later in May did it start separating out antibody tests from the total — showing that the antibody tests were inflating the overall number by nearly 16 percent as of June 4.
When all of these events are taken together, it’s enough to make experts ask if Georgia is more interested in making the numbers look favorable than it is in reporting them truthfully. And that might create a warped view of how exactly the state is doing after reopening.
“The virus isn’t going to care whether they were manipulating the numbers or not in order to look more favorable; it’s going to continue to spread,” Crystal Watson, a senior scholar at the Johns Hopkins Center for Health Security, told me. “It’s better to really understand what’s going on and report that accurately.”
A Georgia Department of Public Health spokesperson said, “There have been several improvements to the COVID-19 Daily Status Report on the Georgia Department of Public Health’s (DPH) website. These changes are designed to make the dashboard more user-friendly while providing an accurate picture of COVID-19 in Georgia.” The spokesperson added, “DPH will regularly review and update features of the dashboard to improve data quality and accuracy.”
3) People are staying home anyway
Before governments told people to stay home, many people were staying home anyway.
Georgia Gov. Brian Kemp didn’t issue a shelter-in-place order until April 2 — making the state among the last to shut down. But OpenTable data indicates that dine-in seating fell by more than 90 percent by the end of March compared to the previous year. A mix of individual behavioral changes, local and other state government actions, federal advisories, and other factors led people to stay home, shelter-in-place order or not.
Similarly, after Georgia’s shelter-in-place order expired, dining in has remained suppressed: As of June 4, it’s down more than 75 percent compared to the year before. That’s more people out than several weeks prior, when dining in was down by 80 to 90 percent, but it’s still a lot of empty restaurant seats.
Thompson, who lives in Atlanta, vouched for this. When Georgia began to reopen, she said, “Traffic was still very low. … Businesses were still not open. Most people were not going to get haircuts.”
Data from Google, which tracks people’s movement through their Google accounts, tells a similar story: Georgians are starting to trickle out more, but they’re still going out less than they were before. As of May 29, retail and recreation outings are down 16 percent compared to a baseline based on activity before the coronavirus. Transit stations are down 38 percent. Workplaces are down 36 percent, and grocery stores and pharmacies are down 4 percent. Meanwhile, residential places — meaning, where people stay home — are up 13 percent.
Only parks, which have a relatively low risk of transmission, are up 44 percent.
Again, this is slowly changing; people are trickling out more, bit by bit. The past week of protests over police brutality have led to a lot of people showing up in large gatherings, including in Atlanta. This speaks to why it’s too early to judge Georgia’s experience: Things could change as more people get out more.
“If at some point people do decide this is over and opening up didn’t do anything [bad], and they go out to concerts and crowded parks and whatever, then we’ll probably start to see an increase in cases again,” Murray said.
4) Masks, good hygiene, and other behavioral changes may make a difference
Besides staying at home, other behavioral changes could be putting a dent in Covid-19 spread as well. Compared to just several months ago, Americans are more likely to wash their hands, avoid touching their faces, wear a mask, and keep at least 6 feet from each other. They’re also probably less likely to go out when they do feel sick.
We don’t know how much all of these things are helping, but experts suspect that these practices are reducing how much transmission happens when people do go out.
Consider masks. Just a few months ago, the idea that many Americans would go out with one was unthinkable. Today, polls show that the great majority of Americans are wearing masks sometimes if not always when they go out.
The research on masks isn’t great, but there’s some suggestive evidence: As long as people actually wear the masks and use them properly, studies indicate that they have some effect in reducing disease transmission overall. Some experts hypothesize — and preliminary research suggests — that masks have played a significant role in containing Covid-19 outbreaks in several Asian countries where their use is widespread, like South Korea and Japan.
Experts say they wouldn’t be surprised if masks play a bigger role than previously expected. As we’ve learned more about the coronavirus, we’ve gotten more and more evidence that it spreads through respiratory droplets. A mask quite literally stops those respiratory droplets.
And that’s only one of the things people have changed. From no longer shaking hands to appreciating the power of soap more, we have done a lot to make sure this virus doesn’t spread as easily as it did when we first learned of it.
This applies not just to individuals but to institutions too — workplaces, restaurants, public services, and so on. “Companies are also increasingly taking responsibility for making sure their workforce stays safe,” MacDonald said.
Whether it’s masks or something else, a large beneficial effect from behavioral changes while going out would be very good news. It would suggest that it’s safe to ease at least some social distancing as long as people follow certain practices. That might let Americans go out and about more — and stay safe — even without strict shelter-in-place orders in effect.
“It’s possible that those kinds of measures can slow transmission, can reduce transmission,” Meyers said. “I’m certainly hopeful that they do.” But, she added, “the jury is still out.”
5) Maybe luck, or something else we don’t fully understand, is playing a role
There’s another answer for why coronavirus cases aren’t increasing in Georgia: We don’t know.
The reality is we’re still learning a lot about the coronavirus, its spread, and how humans behave as a result of the virus. There likely are factors we haven’t anticipated that influence its spread.
It could even come down to sheer luck. Although there are myriad factors that affect the coronavirus — the virus’s characteristics, population density, public transportation use, how people behave, and more — whether all these factors come together to create an event that leads to explosive spread of the virus does come a bit down to chance.
Does that flight with a person who’s infected land in Atlanta or New York City? Does that person who’s, for whatever reason, more likely to shed the virus live in south Georgia or across the border in north Florida? Do several people who are sick end up at a concert in South Carolina or Georgia?
There are things places can do to mitigate these events — stop flights, discourage travel, prohibit large gatherings, encourage good hygiene and mask use, and so on. But chaos can play a major role. “There’s absolutely an element of chance and a lot of uncertainty about what’s going to happen in every single community and every single country,” Meyers said.
And especially because there’s so much about the coronavirus we don’t know, some of the steps we could be taking to prevent its spread are unknowable. That adds more chance to the equation, along with more uncertainty as to why one place suffered a big outbreak but not another.
Of course, luck and uncertainty aren’t something you want to rely on to keep the public safe. That’s why experts focus on how what we do know can help stop the spread of Covid-19. And they worry that by reopening too early — before they build up systems to test, trace, and isolate the coronavirus — states like Georgia may not be doing enough of what we do know helps. We’ll find out if the concerns are warranted in the coming weeks and months.
“We can do a lot to reduce our risk,” Murray said. “Then we can also hopefully have good chance outcomes as well.”