Downtown Detroit is buzzing with new businesses, lofts, and entertainment, but the city’s neighborhoods continue to struggle.
You may have woken up Thursday to the good news that Detroit’s population is rising for the first time since 1957, a time when white people began flocking to the suburbs.
Between July 1, 2022, and July 1, 2023, Detroit gained 1,852 residents, putting the city’s population at 633,366, according to U.S. Census estimates released Monday morning.
Detroit is now ranked as the 26th most populated city in the U.S., leapfrogging Memphis, Louisville, and Portland.
While population gains are a positive sign for any city, the growth in Detroit is far more nuanced and complicated than a single estimate can reveal.
Between 2000 and 2020, Detroit lost about 295,000 Black residents, or 37.4% of its African American population. No other city has lost more Black residents.
Meanwhile, Detroit’s white population grew by more than 5,100 between 2010 and 2020.
Black people now account for 77.2% of the city’s overall population, compared to 82.2% in 2010, when Detroit had the highest percentage of Black residents in the country.
You can see the growth in the pricey lofts and condos that are cropping up in Midtown, downtown, Corktown, Brush Park, the Cass Corridor and the riverfront.
At the same time, a disproportionate number of Black residents are living in neighborhoods dominated by blight, abandonment, and crime. The number of middle-class neighborhoods in Detroit shrunk from 22 in 2010 to 11 in 2020, leaving longtime residents with fewer options to find a decent place to live.
While the latest census information doesn’t break down data by race, it’s difficult to imagine that the Black population suddenly began to rise.
As part of a series Metro Times published last year about the growing racial and economic disparities in Detroit, we talked to Black residents who fled the city and asked them why they left. Overwhelmingly, they said they couldn’t find decent-paying jobs in the city. By contrast, white newcomers are disproportionately getting employed by high-paying businesses.
Over the last decade, the median income of white Detroiters rose 60%. For Black Detroiters, the increase was 8%, according to Detroit Future City, a think tank that develops strategies for a more equitable city.
The average income of a white Detroiter is $46,650, compared to $32,290 for a Black resident. The unemployment rate for Black Detroiters is 1.5 times higher than white residents.
In a recent report, Detroit Future City found that metro Detroit’s fastest-growing, well-paying jobs are disproportionately going to white workers. About 16% of Black workers in the region are in so-called growth occupations, compared to 26% of white workers.
Jobs are considered growth occupations if they are growing at the same or higher rate than the region as a whole, pay at least a middle-class salary, have increased wages between 2014 and 2019, and employ at least 300 people. Most of the jobs pay more than $73,000 a year.
“What we’re seeing pretty consistently unfortunately is that the highest growth for Detoiters in terms of workforce is lower-wage jobs, which means the jobs that you would think of as middle wage or higher wage are not being occupied by Detroiters,” Anika Goss, CEO of Detroit Future City, told Metro Times in May 2023. “The jobs are either going to people who are moving here from other places or suburbanites. They are not Detroiters.”
Black Detroiters are also more likely to be denied mortgages, regardless of their income level. Higher-income Black residents, for example, were denied a loan at a higher rate than moderate-income white applicants.
“We have known for some time that Detroit’s population has been growing, but this is the first time the U.S. Census Bureau has confirmed it in its official estimate,” Duggan said Thursday. “This day is for the Detroiters who stayed and for everyone who has put in the hard work to make Detroit a great place to live.”
Despite the good news about Detroit’s overall population growth, much work still needs to be done to address a future for Black residents.
As a result of the inequities, many Black children are facing long odds of succeeding later in life. More than half of the city’s Black children live in poverty. About 20% of young adults who grow up in poverty end up poor in their 20s, according to the National Center for Children in Poverty.
Detroit’s Black population grew exponentially in the early and mid-1900s, lured by the bustling auto industry. But those fleeing Jim Crow laws in the U.S. south found themselves in similar situations in Detroit, largely relegated to substandard homes in segregated, overpopulated neighborhoods.
In the 1950s, when Detroit’s population peaked at nearly 2 million, Mayor Albert Cobo campaigned on a platform of “Negro removal” — a pledge to force Black people out of predominantly white neighborhoods and deny federal funding for Black housing projects.
In the mid-1950s, the construction of highways decimated the city’s historic Black communities, Black Bottom and Paradise Valley.
By the time federal civil rights laws banned racial discrimination in the 1960s, white people were fleeing the city for the suburbs, and the jobs followed, leaving behind a majority-Black population that lacked the resources to thrive.
Now that white flight is reversing, it’s up to city leaders and wealthy landowners to ensure that Black residents have a fair shake this time.
Home buyers still active. (Photo by Tim Boyle/Getty Images)
Getty Images
Housing has become increasingly expensive. According to the National Association of Realtors,mortgage rates on average have risen by more than a full percentage point over the last 12 months. The average price of a home has inched up as well. Taken together and measured against household incomes, the statisticians at the Association estimate that affordability of home ownership for the average American has plummeted nearly 10% over the past year and today sits at its lowest level since 2011. Economics 101 would tell us that demand should slack off. Yet, home sales continue to rise. The Census Bureau reports that sales of privately owned houses fell off a bit in October, the most recent period for which data are available, but remain some 18% above year-ago levels. Sales have held up in defiance of standard price theory because Americans are still very concerned about inflation.
Indeed, the behavior of the housing market, more than any other economic gauge, announces that Americans, though aware of easing rates of inflation recently, are concerned that the economy is far from out of the woods on this matter. They fear a rising cost of living and exhibit that fear by flocking to the best inflation hedge available to them — home ownership — and secure it even if it means stretching their household budget to the limit. Few homeowners can quote the numbers, but the history of the last great inflation guides their decisions. From the mid-1970s to the mid-1980s, the crushing burden of 6.2% inflation a year tracked by the Bureau of Labor Statistics was still bested by an 8.7% rise in residential real estate values recorded by the Census Bureau. The 2.5 percentage point difference more than made up for the burden of paying mortgage rates that rose to double digits during that time.
For others, the logic of ownership is compelling in yet a different way, even if it means paying high mortgage rates and stretching the household budget to do so. Once the house is secure, whether financed with a cash purchase or a fixed rate mortgage even a high-rate one, the family has fixed the price of a major budget item – shelter – a great comfort when people fear that all other prices will rise unpredictably. For those who remain wary of inflation – and that is most people outside the White House – the peace of mind purchased this way is well worth the budget strains. Affordability might prevent a purchase as large or in as desirable location as hoped, but these benefits justify sliding down the pricing distribution. And this kind of buying has held up demand, despite rising costs.
Pricing might have given way despite this support for demand were it not that supply has also declined. It seems that existing owners, especially those who purchased at the very low mortgage rates that prevailed until last year, have no desire to walk away from such advantages. If for some reason, they need to change residences, they cling to the original mortgage and the house to which it is attached and rent the property, encouraged further by the 11% rise in national rents recorded between 2021 and 2022. They then rent in their new location until conditions for a new purchase are more favorable. Then they sell the old house to buy a new one. At the same time, homebuilders, the Census Bureau reports, have cut back on the construction of single-family houses, some 4.4% over the last year, and some, noting the earlier rise in rents, have turned to the construction of rental properties. Together, this shift by builders and the relative slowdown in the supply of owner-occupied dwellings available for sale has held up prices in this area of the market while causing a sudden halt this year in what was a powerful uptrend in rents.
As is so often the case, the matter is more complex than simple supply-demand-price considerations, especially in a product like housing that lasts for a lot longer than a haircut. If and when inflation fears fade and the Federal Reserve begins to lower interest rates, matters will seem equally perplexing as this confluence of motivations works in reverse.
Charlie McCone has been struggling with the symptoms of long COVID since he was first infected, in March 2020. Most of the time, he is stuck on his couch or in his bed, unable to stand for more than 10 minutes without fatigue, shortness of breath, and other symptoms flaring up. But when I spoke with him on the phone, he seemed cogent and lively. “I can appear completely fine for two hours a day,” he said. No one sees him in the other 22. He can leave the house to go to medical appointments, but normally struggles to walk around the block. He can work at his computer for an hour a day. “It’s hell, but I have no choice,” he said. Like many long-haulers, McCone is duct-taping himself together to live a life—and few see the tape.
McCone knows 12 people in his pre-pandemic circles who now also have long COVID, most of whom confided in him only because “I’ve posted about this for three years, multiple times a week, on Instagram, and they’ve seen me as a resource,” he said. Some are unwilling to go public, because they fear the stigma and disbelief that have dogged long COVID. “People see very little benefit in talking about this condition publicly,” he told me. “They’ll try to hide it for as long as possible.”
I’ve heard similar sentiments from many of the dozens of long-haulers I’ve talked with, and the hundreds more I’ve heard from, since first reporting on long COVID in June 2020. Almost every aspect of long COVID serves to mask its reality from public view. Its bewilderingly diverse symptoms are hard to see and measure. At its worst, it can leave people bed- or housebound, disconnected from the world. And although milder cases allow patients to appear normal on some days, they extract their price later, in private. For these reasons, many people don’t realize just how sick millions of Americans are—and the invisibility created by long COVID’s symptoms is being quickly compounded by our attitude toward them.
Most Americans simply aren’t thinking about COVID with the same acuity they once did; the White House long ago zeroed in on hospitalizations and deaths as the measures to worry most about. And what was once outright denial of long COVID’s existence has morphed into something subtler: a creeping conviction, seeded by academics and journalists and now common on social media, that long COVID is less common and severe than it has been portrayed—a tragedy for a small group of very sick people, but not a cause for societal concern. This line of thinking points to the absence of disability claims, the inconsistency of biochemical signatures, and the relatively small proportion of severe cases as evidence that long COVID has been overblown. “There’s a shift from ‘Is it real?’ to ‘It is real, but …,’” Lekshmi Santhosh, the medical director of a long-COVID clinic at UC San Francisco, told me.
Yet long COVID is a substantial and ongoing crisis—one that affects millions of people. However inconvenient that fact might be to the current “mission accomplished” rhetoric, the accumulated evidence, alongside the experience of long haulers, makes it clear that the coronavirus is still exacting a heavy societal toll.
As it stands, 11 percent of adults who’ve had COVID are currently experiencing symptoms that have lasted for at least three months, according to data collected by the Census Bureau and the CDC through the national Household Pulse Survey. That equates to more than 15 million long-haulers, or 6 percent of the U.S. adult population. And yet, “I run into people daily who say, ‘I don’t know anyone with long COVID,’” says Priya Duggal, an epidemiologist and a co-lead of the Johns Hopkins COVID Long Study. The implication is that the large survey numbers cannot be correct; given how many people have had COVID, we’d surely know if one in 10 of our contacts was persistently unwell.
But many factors make that unlikely. Information about COVID’s acute symptoms was plastered across our public spaces, but there was never an equivalent emphasis that even mild infections can lead to lasting and mercurial symptoms; as such, some people who have long COVID don’t even know what they have. This may be especially true for the low-income, rural, and minority groups that have borne the greatest risks of infection. Lisa McCorkell, a long-hauler who is part of the Patient-Led Research Collaborative, recently attended a virtual meeting of Bay Area community leaders, and “when I described what it is, some people in the chat said, ‘I just realized I might have it.’”
Admitting that you could have a life-altering and long-lasting condition, even to yourself, involves a seismic shift in identity, which some people are understandably loath to make. “Everyone I know got Omicron and got over it, so I really didn’t want to concede that I didn’t survive this successfully,” Jennifer Senior, a friend and fellow staff writer at The Atlantic, who has written about her experience with long COVID, told me. Duggal mentioned an acquaintance who, after a COVID reinfection, can no longer walk the quarter mile to pick her kids up from school, or cook them dinner. But she has turned down Duggal’s offer of an appointment; instead, she is moving across the country for a fresh start. “That is common: I won’t call it ‘long COVID’; I’ll just change everything in my life,” Duggal told me. People who accept the condition privately may still be silent about it publicly. “Disability is often a secret we keep,” Laura Mauldin, a sociologist who studies disability, told me. One in four Americans has a disability; one in 10 has diabetes; two in five have at least two chronic diseases. In a society where health issues are treated with intense privacy, these prevalence statistics, like the one-in-10 figure for long COVID, might also intuitively feel like overestimates.
Some long-haulers are scared to disclose their condition. They might feel ashamed for still being sick, or wary about hearing from yet another loved one or medical professional that there’s nothing wrong with them. Many long-haulers worry that they’ll be perceived as weak or needy, that their friends will stop seeing them, or that employers will treat them unfairly. Such fears are well founded: A British survey of almost 1,000 long-haulers found that 63 percent experienced overt discrimination because of their illness at least “sometimes,” and 34 percent sometimes regretted telling people that they have long COVID. “So many people in my life have reached out and said, ‘I’m experiencing this,’ but they’re not telling the rest of our friends,” McCorkell said.
Imagine that you interact with 50 people on a regular basis, all of whom got COVID. If 10 percent are long-haulers, that’s five people who are persistently sick. Some might not know what long COVID is or might be unwilling to confront it. The others might have every reason to hide their story. “Numbers like 10 percent are not going to naturally present themselves in front of you,” McCone told me. Instead, “you’ll hear from 45 people that they are completely fine.”
Illustration by Paul Spella / The Atlantic; Getty
The same factors that stop people from being public about their condition—ignorance, denial, or concerns about stigma—also make them less likely to file for disability benefits. And that process is, to put it mildly, not easy. Applicants need thorough medical documentation; many long-haulers struggle to find doctors who believe their symptoms are real. Even with the right documents, applicants must hack their way through bureaucratic overgrowth, likely while fighting fatigue or brain fog. For these reasons, attempting to measure long COVID through disability claims is a profoundly flawed exercise. Even if people manage to apply, they face an average wait time of seven months and a two-in-three denial rate. McCone took six weeks to put an application together, and, despite having a lawyer and extensive medical documentation, was denied after one day. McCorkell knows many first-wavers—people who’ve had long COVID since March 2020—“who are just getting their approvals now.”
An alternative source of data comes from the Census Bureau’s Current Population Survey, which simply asks working-age Americans if they have any of six forms of disability. Using that data, Richard Deitz, an economics-research adviser at the Federal Reserve Bank of New York, calculated that about 1.7 million more people now say they do than in mid-2020, reversing a years-long decline. These numbers are lower than expected if one in 10 people who gets COVID really does become a long-hauler, but the survey doesn’t directly capture many of the condition’s most common symptoms, such as fatigue, neurological problems beyond brain fog, and post-exertional malaise, where a patient’s symptoms get dramatically worse after physical or mental exertion. About 900,000 of the newly disabled people are also still working. David Putrino, who leads a long-COVID rehabilitation clinic at Mount Sinai, told me that many of his patients are refused the accommodations required under the Americans With Disabilities Act. Their employers won’t allow them to work remotely or reduce their hours, because, he said, “you look at them and don’t see an obvious disability.”
Long COVID can also seem bafflingly invisible when people look at it with the wrong tools. For example, a 2022 study by National Institutes of Health researchers compared 104 long-haulers with 85 short-term COVID patients and 120 healthy people and found no differences in measures of heart or lung capacities, cognitive tests, or levels of common biomarkers—bloodstream chemicals that might indicate health problems. This study has been repeatedly used as evidence that long COVID might be fictitious or psychosomatic, but in an accompanying editorial, Aluko Hope, the medical director of Oregon Health and Science University’s long-COVID program, noted that the study exactly mirrors what long-haulers commonly experience: They undergo extensive testing that turns up little and are told, “Everything is normal and nothing is wrong.”
The better explanation, Putrino told me, is that “cookie-cutter testing” doesn’t work—a problem that long COVID shares with other neglected complex illnesses, such as myalgic encephalomyelitis/chronic-fatigue syndrome and dysautonomia. For example, the NIH study didn’t consider post-exertional malaise, a cardinal symptom of both ME/CFS and long COVID; measuring it requires performing cardiopulmonary tests on two successive days. Most long-haulers also show spiking heart rates when asked to simply stand against a wall for 10 minutes—a sign of problems with their autonomic nervous system. “These things are there if you know where to look,” Putrino told me. “You need to listen to your patients, hear where the virus is affecting them, and test accordingly.”
Contrary to popular belief, researchers have learned a huge amount about the biochemical basis of long COVID, and have identified several potential biomarkers for the disease. But because long COVID is likely a cluster of overlapping conditions, there might never be a singular blood test that “will tell you if you have long COVID 100 percent of the time,” Putrino said. The best way to grasp the scale of the condition, then, is still to ask people about their symptoms.
Large attempts to do this have been relatively consistent in their findings: The U.S. Household Pulse Survey estimates that one in 10 people who’ve had COVID currently have long COVID; a large Dutch study put that figure at one in eight. The former study also estimated that 6 percent of American adults are long-haulers; a similar British survey by the Office for National Statistics estimated that 3 percent of the general population is. These cases vary widely in severity, and about one in five long-haulers is barely affected by their symptoms—but the remaining majority very much is. Another one in four long-haulers (or 4 million Americans) has symptoms that severely limit their daily activities. The others might, at best, wake every day feeling as if they haven’t had any rest, or feel trapped in an endless hangover. They might work or socialize when their tidal symptoms ebb, but only by making big compromises: “If I work a full day, I can’t also then make dinner or parent without significant suffering,” JD Davids, who has both long COVID and ME/CFS, told me.
Some people do recover. A widely cited Israeli study of 1.9 million people used electronic medical records to show that most lingering COVID symptoms “are resolved within a year from diagnosis,” but such data fail to capture the many long-haulers who give up on the medical system precisely because they aren’t getting better or are done with being disbelieved. Other studies that track groups of long-haulers over time have found less rosy results. A French one found that 85 percent of people who had symptoms two months after their infection were still symptomatic after a year. A Scottish team found that 42 percent of its patients had only partially recovered at 18 months, and 6 percent had not recovered at all. The United Kingdom’s national survey shows that 69 percent of people with long COVID have been dealing with symptoms for at least a year, and 41 percent for at least two.
The most recent data from the U.S. and the U.K. show that the total number of long-haulers has decreased over the past six months, which certainly suggests that people recover in appreciable numbers. But there’s a catch: In the U.K., the number of people who have been sick for more than a year, or who are severely limited by their illness, has gone up. A persistent pool of people is still being pummeled by symptoms—and new long-haulers are still joining the pool. This influx should be slower than ever, because Omicron variantsseem tocarry alower risk of triggering long COVID, while vaccines and the drug Paxlovid can lower that risk even further. But though the odds against getting long COVID are now better, more people are taking a gamble, because preventive precautions have been all but abandoned.
Even if prevalence estimates were a tenth as big, that would still mean more than 1 million Americans are dealing with a chronic illness that they didn’t have three years ago. “When long COVID first came on the scene, everyone told us that once we have the prevalence numbers, we can do something about it,” McCorkell told me. “We got those numbers. Now people say, ‘Well, we don’t believe them. Try again.’”
To a degree, I sympathize with some of the skepticism regarding long COVID, because the condition challenges our typical sense of what counts as solid evidence. Blood tests, electronic medical records, and disability claims all feel like rigorous lines of objective data. Their limitations become obvious only when you consider what the average long-hauler goes through—and those details are often cast aside because they are “anecdotal” and, by implication, unreliable. This attitude is backwards: The patients’ stories are the ground truth against which all other data must be understood. Gaps between the data and the stories don’t immediately invalidate the latter; they just as likely show the holes in the former.
Laura Mauldin, the disability sociologist, argues that the U.S. is primed to discount those experiences because the country’s values—exceptionalism, strength, self-reliance—have created what she calls the myth of the able-bodied public. “We cannot accept that our bodies are fallible, or that disability is utterly ordinary and expected,” she told me. “We go to great pains to pretend as though that is not the case.” If we believe that a disabling illness like long COVID is rare or mild, “we protect ourselves from having to look at it.” And looking away is that much easier because chronic illnesses like long COVID are more likely to affect women—“who are more likely to have their symptoms attributed to psychological problems,” Mauldin said—and because the American emphasis on work ethic devalues people who can’t work as much or as hard as their peers.
Other aspects of long COVID make it hard to grasp. Like other similar, neglected chronic illnesses, it defies a simplistic model of infectious disease in which a pathogen causes a predictable set of easily defined symptoms that alleviate when the bug is destroyed. It challenges our belief in our institutions, because truly contending with what long-haulers go through means acknowledging how poorly the health-care system treats chronically ill patients, how inaccessible social support is to them, and how many callous indignities they suffer at the hands of even those closest to them. Long COVID is a mirror on our society, and the image it reflects is deeply unflattering.
Most of all, long COVID is a huge impediment to the normalization of COVID. It’s an insistent indicator that the pandemic is not actually over; that policies allowing the coronavirus to spread freely still carry a cost; that improvements such as better indoor ventilation are still wanting; that the public emergency may have been lifted but an emergency still exists; and that millions cannot return to pre-pandemic life. “Everyone wants to say goodbye to COVID,” Duggal told me, “and if long COVID keeps existing and people keep talking about it, COVID doesn’t go away.” The people who still live with COVID are being ignored so that everyone else can live with ignoring it.
This article originally misstated the name of the bank where Richard Deitz works.
It is a truth universally acknowledged among health experts that official COVID-19 data are a mess right now. Since the Omicron surge last winter, case counts from public-health agencies have become less reliable. PCR tests have become harder to access and at-home tests are typically not counted.
Official case numbers now represent “the tip of the iceberg” of actual infections, Denis Nash, an epidemiologist at the City University of New York, told me. Although case rates may seem low now, true infections may be up to 20 times higher. And even those case numbers are no longer available on a daily basis in many places, as the CDC and most state agencieshave switched to updating their data once a week instead of every day.
How, then, is anyone supposed to actually keep track of the COVID-19 risk in their area—especially when cases are expected to increase this fall and winter? Using newer data sources, such as wastewater surveillance and population surveys, experts have already noticed potential signals of a fall surge: Official case counts are trending down across the U.S., but Northeast cities such as Boston are seeing more coronavirus in their wastewater, and the CDC reports that this region is a hotspot for further-mutated versions of the Omicron variant. Even if you’re not an expert, you can still get a clearer picture of how COVID-19 is hitting your community in the weeks ahead. You’ll simply need to understand how to interpret these alternate data sources.
The problem with case data goes right to the source. Investment in COVID-19 tracking at the state and local levels has been in free fall, says Sam Scarpino, a surveillance expert at the Rockefeller Foundation’s Pandemic Prevention Initiative. “More recently, we’ve started to see lots of states sunsetting their reporting,” Scarpino told me. Since the Pandemic Prevention Initiative and the Pandemic Tracking Collective started publishing a state-by-state scorecard of breakthrough-case reporting in December 2021, the number of states with a failing grade has doubled. Scarpino considers this trend a “harbinger of what’s coming” as departments continue to shift resources away from COVID-19 reporting.
Hospitalization data don’t suffer from the same reporting problems, because the federal government collects information directly from thousands of facilities across the country. But “hospitalizations often lag behind cases by a matter of weeks,” says Caroline Hugh, an epidemiologist and volunteer with the People’s CDC, an organization providing COVID-19 data and guidance while advocating for improved safety measures. Hospitalizations also don’t necessarily reflect transmission rates, which still matter if you want to stay safe. Some studies suggest, for example, that long COVID might now be more likely than hospitalization after an infection.
For a better sense of how much the coronavirus is circulating, many experts are turning to wastewater surveillance. Samples from our sewage can provide an advanced warning of increased COVID-19 spread because everyone in a public-sewer system contributes data; the biases that hinder PCR test results don’t apply. As a result, Hugh and her colleagues at the People’s CDC consider wastewater trends to be more “consistent” than constantly fluctuating case numbers.
When Omicron first began to wreak havoc in December 2021, “the wastewater data started to rise very steeply, almost two weeks before we saw the same rise” in case counts, Newsha Ghaeli, the president and a co-founder of the wastewater-surveillance company Biobot Analytics, told me. Biobot is now working with hundreds of sewage-sampling sites in all 50 states, Ghaeli said. The company’s national and regional dashboard incorporates data from every location in its network, but for more local data, you might need to go to a separate dashboard run by the CDC or by your state health department. Some states have wastewater surveillance in every county, while others have just a handful of sites. If your location is not represented, Ghaeli said, “the wastewater data from communities nearby is still very applicable.” And even if your county does have tracking, checking up on neighboring communities might be good practice. “A surge in a state next door … could very quickly turn into a surge locally,” Ghaeli explained.
Ghaeli recommends watching how coronavirus levels in wastewater shift over time, rather than homing in on individual data points. Look at both “directionality” and “magnitude”: Are viral levels increasing or decreasing, and how do these levels compare with earlier points in the pandemic? A 10 percent uptick when levels are low is less concerning than a 10 percent uptick when the virus is already spreading widely.
Researchers are still working to understand how wastewater data correlate with actual infections, because every community has unique waste patterns. For example, big cities differ from rural areas, and in some places, environmental factors such as rainfall or nearby agriculture may interfere with coronavirus tracking. Still, long-term-trend data are generally thought to be a good tool that can help sound the alarm on new surges.
Wastewater data can help you figure out how much COVID-19 is spreading in a community and can even track all the variants circulating locally, but they can’t tell you who’s getting sick. To answer the latter question, epidemiologists turn to what Nash calls “active surveillance”: Rather than relying on the COVID-19 test results that happen to get reported to a public-health agency, actively seek out and ask people whether they recently got sick or tested positive.
Nash and his team at CUNY have conducted population surveys in New York City and at the national level. The team’s most recent survey (which hasn’t yet been peer-reviewed), conducted from late June to early July, included questions about at-home test results and COVID-like symptoms. From a nationally representative survey of about 3,000 people, Nash and his team found that more than 17 percent of U.S. adults had COVID-19 during the two-week period—about 24 times higher than the CDC’s case counts at that time.
Studies like these “capture people who might not be counted by the health system,” Nash told me. His team found that Black and Hispanic Americans and those with low incomes were more likely to get sick during the survey period, compared with the national estimate. The CDC and Census Bureau take a similar approach through the ongoing Household Pulse Survey.
These surveys are “a goldmine of data,” though they need to be “carefully designed,” Maria Pyra, an epidemiologist and volunteer with the People’s CDC, told me. By showing the gap between true infections and officially reported cases, surveys like Nash’s can allow researchers to approximate how much COVID-19 is really spreading.
Survey results may be delayed by weeks or months, however, and are typically published in preprints or news reports rather than on a health agency’s dashboard. They might also be biased by who chooses to respond or how questions are worded. Scarpino suggested a more timely option: data collected from cellphone locations or social media. The Delphi Group at Carnegie Mellon University, for example, provides data on how many people are Googling coldlike symptoms or seeking COVID-related doctor visits. While such trends aren’t a perfect proxy for case rates, they can be a helpful warning that transmission patterns are changing.
Readers seeking to monitor COVID-19 this fall should “look as local as you can,” Scarpino recommended. That means examining county- or zip-code-level data, depending on what’s available for you. Nash suggested checking multiple data sources and attempting to “triangulate” between them. For example, if case data suggest that transmission is down, do wastewater data say the same thing? And how do the data match with local behavior? If a popular community event or holiday happened recently, low case numbers might need to be taken with a grain of salt.
“We’re heading into a period where it’s going to be increasingly harder to know what’s going on with the virus,” Nash told me. Case numbers will continue to be undercounted, and dashboards may be updated less frequently. Pundits on Twitter are turning to Yankee Candle reviews for signs of surges. Helpful sources still exist, but piecing together the disparate data can be exhausting—after all, data reporting and interpretation should be a job for our public-health agencies, not for concerned individuals.
Rather than accept this fragmented data status quo, experts would like to see improved public-health systems for COVID-19 and other diseases, such as monkeypox and polio. “If we get better at collecting and making available local, relevant infectious-disease data for decision making, we’re going to lead healthier, happier lives,” Scarpino said.
