Our Pandemic Summer
The fight against the coronavirus won’t be over when the U.S. reopens. Here’s how the nation must prepare itself.
Updated at 11:50 a.m. ET on April 15, 2020.
What a difference a few months can make.
In January, the United States watched as the new coronavirus blazed through China and reached American shores. In February, hindered by an unexpected failure to roll out diagnostic tests and an administration that had denuded itself of scientific expertise, the nation sat largely idle while the pandemic spread within its borders. In March, as the virus launched several simultaneous assaults on a perilously stretched-thin health-care system, America finally sputtered into action, frantically closing offices, schools, and public spaces in a bid to cut off chains of transmission. Now, in April, as viral fevers surge through American hospitals and cabin fever grows in American homes, the U.S. has cemented itself as the new center of the pandemic—the country that should have been more prepared than any other, but that now has the worst COVID-19 outbreak in the world.
What will May bring? Or June? What happens as this seemingly interminable spring rolls into a precarious summer? When will things go back to normal?
The options are limited. Early inaction left the U.S. with too many new cases, and just one recourse: Press a societal pause button to buy enough time for beleaguered hospitals to steel themselves for a sharp influx in patients. This physical-distancing strategy is working, but at such an economic cost that it can’t be sustained indefinitely. When restrictions relax, as they are set to do on April 30, the coronavirus will likely surge back, as it is now doing in Singapore, China, and other Asian states that had briefly restrained it.*
As I wrote last month, the only viable endgame is to play whack-a-mole with the coronavirus, suppressing it until a vaccine can be produced. With luck, that will take 18 to 24 months. During that time, new outbreaks will probably arise. Much about that period is unclear, but the dozens of experts whom I have interviewed agree that life as most people knew it cannot fully return. “I think people haven’t understood that this isn’t about the next couple of weeks,” said Michael Osterholm, an infectious-disease epidemiologist at the University of Minnesota. “This is about the next two years.”
The pandemic is not a hurricane or a wildfire. It is not comparable to Pearl Harbor or 9/11. Such disasters are confined in time and space. The SARS-CoV-2 virus will linger through the year and across the world. “Everyone wants to know when this will end,” said Devi Sridhar, a public-health expert at the University of Edinburgh. “That’s not the right question. The right question is: How do we continue?”
A lockdown is a measure of last resort, to be used only when a virus is spreading so rapidly that it cannot be controlled through other means. Having deployed that measure, albeit unevenly, the U.S. has now bought itself some time. It can use that time to address its lack of tests and medical supplies, and find less economically devastating ways of controlling COVID-19. When sufficiently braced, states could begin lifting their sweeping restrictions and focus on finding and helping people who are actually infected. But the conditions for making that switch are not clear. “We’ve never faced a pandemic like this before in modern times, so we’re going to have to be flexible,” said Caitlin Rivers of the Johns Hopkins Center for Health Security. “There’s no real playbook.”
The White House seems to be relying heavily on one mathematical model from the University of Washington that, in its first incarnation, was criticized for failing to accurately predict death rates even a day or two out. (While traditional models simulate how a disease spreads through virtual communities, this one simply made predictions based on death numbers in the U.S. and death rates in other countries.) “When we make weather predictions or plan for hurricanes, we don’t just plan based on one thing,” said Natalie Dean, a statistician at the University of Florida. Alternative models are being built, but in the meantime, it’s hard to accurately forecast how long the current phase of the pandemic will last.
Absent any certainty, one group of health experts, led by Ezekiel Emanuel, an oncologist and former adviser to the Obama administration, argues that stay-at-home orders must continue until at least May 20. Another team, led by Scott Gottlieb, a former FDA commissioner, suggests that states should relax their restrictions only after new case counts have fallen for 14 consecutive days. Currently, the U.S. is still averaging about 30,000 new confirmed cases every day. “My sense is that we haven’t turned the corner,” said Rivers, who is a co-author on Gottlieb’s road map.
Both groups agree that before relaxing the guidelines on social distancing the U.S. urgently needs to expand its ability to test for the virus, and to shore up hospitals with sufficient supplies. These recommendations are sensible, but they hinge on the expectation that the U.S. can recover the ground it lost due to its early inaction. It might not be able to.
For example, with help from private companies and academic institutions, the U.S. is certainly testing more people than it was before. Over the past week, about 145,000 people have been tested every day, according to the COVID Tracking Project, a volunteer collaboration spearheaded by The Atlantic. Those numbers look to rise even further, thanks to a new, rapid test from Abbott Laboratories that can deliver results in less than 15 minutes. But as testing capacity has grown, so has the pandemic. As my colleagues Robinson Meyer and Alexis Madrigal have reported, private labs have taken on more orders than they can fulfill, and are experiencing huge backlogs. Demand for tests has ballooned, fueled by a rise in actual infections and the fact that Donald Trump keeps wrongly assuring the public that testing is no longer a problem. “The net gain just hasn’t been there,” said Kelly Wroblewski of the Association of Public Health Laboratories.
Compounding these issues, testing companies all rely on the same chemical ingredients, which were already running low and are becoming even scarcer as the pandemic spreads. “Trying to get these materials is not about dollars and cents; it’s about physics,” Osterholm said. “You can’t just buy these things. It takes time to make them.” To be clear, there is wide consensus that testing is crucial for working out where the virus is and adequately preparing for it. In an ideal reality, testing should be done extensively. But in this reality, Osterholm noted, it might be several months before manufacturers can meet global demand. For the summer, the U.S. might have to abandon the dream that “everyone who gets sick is going to get a test and we’ll be able to count every case with a lab diagnosis,” Wroblewski said. “That might not be realistic.”
Crucial medical drugs are also running out. According to a University of Minnesota analysis, about 40 percent of the 156 drugs that are essential parts of critical care are becoming limited. Many of these depend on supply chains that involve China (where the pandemic began), Italy (the hardest-hit region in Europe), or India (which halted several exports). These chains have been discharging their contents like a sputtering garden hose that has now begun to run dry. “The medium term is going to be particularly perilous,” said Nada Sanders, a professor of supply-chain management at Northeastern University. “Global demand is so high, and supply is so far behind, that it’s very hard to envision enough of a ramp-up.”
Albuterol, the drug used in asthma inhalers, is scarce. Antibiotics, which control the secondary bacterial infections that afflict COVID-19 patients, are being depleted. Basic painkillers and sedatives, which are needed to keep patients on ventilators, are being exhausted. Hydroxychloroquine, the drug that Trump has repeatedly touted as a COVID-19 treatment despite a lack of good evidence, is running out, to the detriment of people with lupus and arthritis who depend on it. “It’s like everything we give to patients, we’re in short supply of,” said Esther Choo, an emergency physician at Oregon Health and Science University. “We’re now scrambling to find the backup medications, and we’ll run out of those too.”
Meanwhile, hospitals are still struggling. There aren’t enough masks and gowns to adequately protect staff, ventilators to deliver oxygen to patients who can’t breathe, or respiratory technicians and nurses to operate those ventilators. Overwhelmed and underprotected, doctors and nurses are falling sick. In Michigan, more than 700 employees at one hospital have tested positive for the coronavirus; at another, 1,500 are out of work with consistent symptoms. Hospitals are now bringing back retired physicians, graduating medical students early, and re-tasking orthopedists and dermatologists to emergency rooms to help with the coronavirus surge.
Even if cases abate in the summer, exhausted health-care workers will have to catch up on elective surgeries that were postponed because of the pandemic (and that are important to a hospital’s financial survival), while also treating people who delayed getting help for heart problems, cancers, and other emergencies. “Patients are putting off health concerns that really need to be seen, and they’re coming in sick,” Choo said. If the pandemic rebounds in the fall, health-care workers may have to greet it without having had a summertime respite. Hospital disaster plans “are all structured around a single discrete event that is over pretty quickly,” Choo said. “We don’t expect our diseases to be stuttering.”
These problems—the continuing testing debacle, the drying supply chains, the relentless pressure on hospitals—should temper any impatience about reopening the country. There won’t be an obvious moment when everything is under control and regular life can safely resume. Even after case counts and death rates fall, the pandemic’s challenges will continue, and will not automatically subside on their own. After all, despite ample warning, the U.S. failed to anticipate what would happen when the coronavirus knocked on its door. It cannot afford to make that mistake again. Before the spring is over, it needs a plan for the summer and fall.
There’s good reason to reopen the U.S. slowly and methodically. When the pandemic first hit, a wide range of social-distancing measures—closing offices, shutting schools, banning mass gatherings, implementing stay-at-home orders—were rolled out in a sudden panic. The rushed deployment was necessary, but made it hard to know which measures actually mattered. The next few months offer opportunities to find out. Communities could relax restrictions gradually, and see if the virus remains at a simmer or returns to a boil. When the virus returns, political leaders should be able to make more informed decisions about which levers to flip. “We should absolutely be throwing everything we can to figure that out,” said Jessica Metcalf, an infectious-disease ecologist at Princeton.
The most crucial piece of missing information, and the one that most dramatically shapes the nation’s options, is what proportion of people have actually been infected. So far, the U.S. has more than 576,000 confirmed cases. But between the lack of testing and the unknown proportion of people who experience mild or nonexistent symptoms, the true number of cases is anyone’s guess.
There’s an easy way to find out. When someone is infected, their immune system produces defensive molecules called antibodies, which recognize (and, with hope, neutralize) the virus. These antibodies endure after the infection is cleared, and should theoretically provide some degree of lasting immunity. By detecting such antibodies through what are known as serological tests, scientists can deduce how many people have been infected in these past months. (Serological tests differ from diagnostic tests, which search for pieces of the coronavirus’s genetic material to work out who is infected right now.) Such “serosurveys” are ongoing, and while preliminary data have emerged, they don’t paint a clear picture yet. In a German town that became a hot spot of the country’s outbreak in February, about 15 percent of people have been infected; in a Colorado county, the figure stands between 0.4 and 1.5 percent.
If it turns out that, say, 20 percent of the U.S. has been infected, that would mean the coronavirus is more transmissible but less deadly than scientists think. It would also mean that a reasonable proportion of the country has some immunity. If that proportion could be slowly and safely raised to the level necessary for herd immunity—60 to 80 percent, depending on the virus’s transmissibility—the U.S. might not need to wait for a vaccine. However, if just 1 to 5 percent of the population has been infected—the range that many researchers think is likelier—that would mean “this is a truly devastating virus, and we have built up no real population immunity,” said Michael Mina, an epidemiologist and immunologist at Harvard. “Then we’re in dire straits in terms of how to move forward.”
Even in the optimistic scenario, a quick and complete return to normalcy would be ill-advised. And even in the pessimistic scenario, controlling future outbreaks should still be possible, but only through an immense public-health effort. Epidemiologists would need to run diagnostic tests on anyone with COVID-19–like symptoms, quarantine infected people, trace everyone those people had contact with in the previous week or so, and either quarantine those contacts or test them too. These are the standard pillars of public health, but they’re complicated by the coronavirus’s ability to spread for days before causing symptoms. Every infected person has a lot of potential contacts, and may have unknowingly infected many of them.
Tracking such a pathogen requires a lot of people, but due to chronic underfunding, local U.S. health departments lost more than 55,000 workers from 2008 to 2017. In their absence, a corps of volunteers could be quickly trained in the basics of contact tracing, as Massachusetts Governor Charlie Baker is planning to do. “It might be an opportunity to bring in people who are recently unemployed—a wartime effort where people aren’t doing their normal jobs,” said Crystal Watson of the Johns Hopkins Center for Health Security.
If there aren’t enough tests, as seems likely in the near term, health officials could focus their attention on looking for spikes in flu-like symptoms, or for less orthodox indicators such as crowded hospital parking lots and restaurant cancellations. More controversially, they could quickly track an infected person’s contacts by comparing their cellphone activity with that of others in the same area. As South Korea and Singapore demonstrated, such methods clearly work. They undoubtedly raise privacy concerns, but as my colleague Derek Thompson argues, “Compared with our present nightmare, strategically sacrificing our privacy might be the best way to protect other freedoms.”
Once the U.S. gets better at tracking the coronavirus, it could use social distancing more flexibly and precisely. COVID-19 counts could feature on the nightly weather report, said Stephen Kissler, an infectious-disease modeler at Harvard. “In the same way that cities issue evacuation orders a few days before a hurricane hits, they could issue distancing orders a few days before we reach the threshold that would threaten to overwhelm our ICUs,” he added.
There’s a risk in trying to be too clever, though. Dylan Morris of Princeton, who also models infectious diseases, notes that exponentially growing epidemics are not just harder, but riskier, to control. Slight delays in action can have huge consequences, as the United Kingdom learned last month. Relying too heavily on models, the British government believed that it could precisely control the spread of COVID-19 by rolling out social restrictions at carefully chosen moments. Its hubris led to a substantial spike in cases. Now that the U.S. is slowing the pandemic, gently easing back on social distancing would be safer, Morris argues, than snapping back to business as usual when small missteps could be catastrophic. “If we’re judicious about how we lift restrictions, we might never have to go back into lockdown,” he said.
Stay-at-home orders might lift first, allowing friends and family to reunite. Small businesses could reopen with limitations: Offices might run on shifts and still rely heavily on teleworking, while restaurants and bars could create more space between tables. Schools could restart once researchers determine if children actually spread the virus.
This process might take several weeks to unfold, and even at the end of it, none of the experts I spoke with was comfortable with the return of crowded public spaces. Gottlieb’s road map, for example, recommends that until a vaccine or an effective treatment is produced, social gatherings should be limited to 50 people or fewer. That will be especially challenging in large cities: An average Manhattan street or subway car is the equivalent of a mass gathering. Elsewhere, concerts, conferences, summer camps, political rallies, large weddings, and major sporting events may all have to be suspended for at least this year. “It’s hard for me to imagine anyone going to Fenway Park and sitting with 30,000 fans—that will almost surely be a bad idea,” said Ashish Jha, an internist and public-health expert at Harvard. “This isn’t going to look like a normal summer in America.”
During the long wait for a vaccine, other countermeasures could conceivably dampen the threat of COVID-19. The simplest of these is the face mask. Medical masks are still in short supply, and must be reserved for health-care workers. But homemade alternatives might help slow the spread of the coronavirus, less by preventing healthy wearers from getting infected and more by stopping unknowingly sick wearers from infecting others. As I wrote earlier this month, masks are symbols as well as shields. In East Asian countries, where they are widely worn, they signify civic-mindedness and conscientiousness. As their use grows in Western nations, they could send the message that society is collectively acting against a serious threat.
Effective treatments for the new coronavirus could also blunt the sting of future outbreaks. As my colleague Sarah Zhang reports, “More than 100 existing and experimental drugs are being tested against COVID-19.” But it’s unlikely that any of these would be an outright cure in the way that antibiotics can be for bacterial infections. (Antibiotics do not work against viruses.) That’s partly because viruses are simpler than bacteria, with fewer vulnerabilities to exploit. Viruses are also more likely than bacteria to actually destroy our cells; by the time symptoms appear, viruses have caused a lot of tissue damage that isn’t easily reversed. And they are more likely to trigger cytokine storms—massive overreactions from the immune system that cause more damage than the infections they are trying to clear. “It’s likely that a therapeutic would only provide incremental benefit over the backbone of supportive medical care,” said Luciana Borio, a physician who served in the National Security Council’s pandemic-preparedness office, which was largely disbanded in 2018. Tamiflu, for example, shortens flu infections by just a day or two, and works best if taken before symptoms appear or shortly after. A similar drug would not obviate the need for a vaccine, or fully negate the coronavirus threat.
More realistically, treatments might give critically ill patients a better chance of survival, or prevent some people with early symptoms from ever needing critical care at all. Either would be an important win. “If people are feeling cruddy at home, and [an anti–COVID-19 drug] can reduce their risk of progressing to hospitalization from 20 percent to 10 percent, that would feel like a massive home run,” Jha said.
Even without antivirals, many COVID-19 patients will recover on their own. If they retain antibodies that confer lasting immunity against the coronavirus, they could conceivably be free to work, support health-care workers, or care for the elderly and other vulnerable groups. Several countries, including the U.S., are now hoping to identify immune individuals with serology tests and affirm their status with “immunity passports,” akin to the yellow card that’s issued following a yellow-fever vaccination. But such a system faces many substantial problems.
First, antibodies aren’t always effective at neutralizing viruses. If you picture the coronavirus as a car, an antibody might slash its tires, or just gum up its wipers—and simple serology tests can’t tell which. Second, even if the antibodies are the right kind, no one knows what concentration you’d need to confer immunity. “Even for diseases we’ve been studying for over 100 years, like whooping cough, we still don’t know what level of antibody would indicate that you’d be protected if you got reexposed,” said Sam Scarpino of Northeastern University, who studies infectious-disease dynamics. The only way to find out is through long studies.
Third, serological tests for the new coronavirus could be deeply misleading for individuals. Consider the test produced by Cellex—the only one thus far with emergency use authorization from the FDA. The test has a 93.8 percent chance of correctly identifying people with antibodies against the new coronavirus, and a 95.6 percent chance of correctly identifying people who lack those antibodies. Those numbers sound great, but if only a minority of Americans have been infected, the test would return far more false positives than true ones. Put it this way: If you have a positive result, the odds that you actually have any relevant antibodies are roughly one in two if 5 percent of the U.S. has been infected, and just one in six if only 1 percent has been infected. Scientists can correct for these errors if they use serological tests to assess immunity in a population, but it’s much harder on a person-by-person basis.
Finally, any certificate that affords special status, like the ability to work while others are quarantined, will create incentives for people to deliberately infect themselves or game the system with counterfeits. Immunity passports would unfairly “favor individuals who didn’t comply with social distancing and got sick early on,” said Alexandra Phelan of Georgetown University, who works on legal and policy issues related to infectious diseases. And “the idea that there would be a midpoint where some people could resume the right to be citizens and others could not is effectively an apartheid system,” said Sharon Abramowitz, a consultant at UNICEF who studies community responses to pandemics. “It might serve specific public-health ends, but in this society would be very problematic.” History affirms that concern: When yellow fever hit the American South in the 19th century, “immunoprivilege” worsened existing forms of discrimination while creating new ones.
Passports and antibody tests aside, it’s reasonable to assume that someone who recovers from COVID-19 would be less likely to get infected again (for some duration that is still unclear). “But we can’t assume that you won’t bring that virus to someone else,” said Mina, the Harvard immunologist. “I worry that a lot of employers are just assuming that having antibodies or having been infected means you’re good to get back to work.” A false sense of security could quicken the spread of the virus during future surges, especially if people ignore social-distancing orders on the mistaken belief that they are immune.
These problems might be surmountable. The U.S. is still a scientific and biomedical powerhouse. To marshal that power, it needs a massive, coordinated, government-led initiative to find the cleverest ways of controlling COVID-19—a modern-day Apollo program. No such program is afoot. Former Trump- and Obama-era officials have published detailed plans. Elizabeth Warren is on her third iteration. But the White House either has no strategy or has chosen not to disclose it.
Without a unifying vision, governors and mayors have been forced to handle the pandemic themselves. Ludicrously, states are bidding against one another—and the federal government—for precious supplies. Six states still haven’t issued any kind of stay-at-home order, while those that moved late, such as Florida, may have seeded infections in the rest of the country. “A patchwork approach to fighting a pandemic is very dangerous,” said Jeremy Konyndyk of the Center for Global Development. “It’s a recipe for a response that’s less than the sum of its parts.” While several states have created their own coordinated groups, Konyndyk’s worry, shared by others, is that there are limits to what even the most capable state leaders can do without federal coordination. “We almost need to devise a public-health government in exile which can take on the responsibility of national coordination,” said Osterholm, the University of Minnesota epidemiologist.
That responsibility would typically fall to the Centers for Disease Control and Prevention, but it has been strangely silent. For almost a month, the CDC’s leaders have been absent from press briefings, and its supporters fear that its reputation has been tarnished. Local leaders are making hundreds of public-health decisions—should we close this park, or open that beach?—without consulting the nation’s top public-health agency. Jha said he regularly fields calls from mayors and governors who would normally ring the CDC. “People don’t know where to turn to for expertise,” he said.
During the Vietnam War, Vice Admiral James Stockdale spent seven years being tortured in a Hanoi prison. When asked about his experience, he noted that optimistic prison-mates eventually broke, as they passed one imagined deadline for release after another. Stockdale’s strategy, instead, was to meld hope with realism—“the need for absolute, unwavering faith that you can prevail,” as he put it, with “the discipline to begin by confronting the brutal facts, whatever they are.”
The U.S. needs to learn that lesson, but Trump is still behaving as if he’s engaged in a brief skirmish rather than a protracted siege. On April 8, before even the first pandemic peak had subsided, he tweeted: “Once we OPEN UP OUR GREAT COUNTRY, and it will be sooner rather than later, the horror of the Invisible Enemy, except for those that sadly lost a family member or friend, must be quickly forgotten.” The enemy isn’t going anywhere. To forget it would be to beget further horror.
In 2018, when writing about whether the U.S. was ready for the next pandemic, I noted that the country was trapped in a cycle of panic and neglect. It rises to meet each new disease, but then settles into complacency once the threat is over. With COVID-19, I fear that the U.S. might enter the neglect phase before the panic part is even finished. If the current shutdown succeeds in flattening the curve, sparing the health-care system and minimizing deaths, it will feel like an overreaction. Contrarians will use the diminished body count to argue that the panic was needless and that the public was misled. Some are already saying that.
Others will divert responsibility from the country’s leaders and its institutions. “There’s a lot of victim-blaming in the U.S., which comes from that neoliberal perspective where it’s your fault if you have bad circumstances,” said Emily Brunson, an anthropologist at Texas State University. The virus is disproportionately killing people in low-income jobs who don’t have the privilege of working from home, but who will nonetheless be shamed for not distancing themselves. The virus is disproportionately killing black people, whose health had already been impoverished through centuries of structural racism, but who will nonetheless be personally blamed for their fate. The virus is disproportionately killing elderly people, who had already been shunted to the fringes of society, but who will nonetheless be told to endure further loneliness so that everyone else can be freer.
If that happens, the panic-neglect cycle will inevitably continue. The U.S. will miss the chance to reexamine how systemic failures of its health-care system left so many citizens vulnerable, or to put in place measures that might forestall another resurgence. It will get hammered by the same damn virus again, and be driven into more severe lockdowns. People will tire. Compliance will fall. The nation would lose its single most effective weapon against the pandemic—the willingness of its citizens to make individual sacrifices for the sake of all. It could forcibly quarantine uncooperative people or impose criminal penalties, but “we know from past outbreaks that criminalizing public-health responses makes people go underground, perpetuates stigma, and fractures society,” Phelan said. “For the marathon we’re facing, we need solidarity.”
Over the coming months, we need “to normalize COVID in the public psyche, and reinforce that this will be a part of our day-to-day lives,” said Kissler, the Harvard disease modeler. “Many people I've spoken with are aghast at the thought. We thirst for a swift and decisive 'victory.' But I'm reminded of images from World War II as people in London walked to work, briefcases in hand, against a backdrop of bombed-out buildings. I think we are in store for a similar period in history, as we learn to make greater peace with the world's chaos and our own mortality.”
This kind of psychological resilience already exists within large groups of people who have been marginalized during periods of normalcy, and who are now the most at-risk from the pandemic. Elderly people, for example, have been most frequently cast as a vulnerable group in need of protection. But “older people have been through a lot—the civil-rights movement, the women’s movement, the Cold War,” said Monica Schoch-Spana, a medical anthropologist at the Johns Hopkins Center for Health Security. “They are culture keepers, full of stories of how you get through major tragedies and upheavals.”
The disability community has also noted that, at a time when their health is in jeopardy and their value is in question, abled people are struggling with a new normal that is their old normal—spatial confinement, unpredictable futures, social distance. “We know how to do community from afar, and how to organize from bed,” said Ashley Shew of Virginia Tech, who studies the intersection between technology and disability. “Instead of feeling this great vacuum, our social life hasn’t radically changed.” Disability scholars have written about “crip time”—a flexible attitude toward timekeeping that comes from uncertainty. “Everything I enter in my calendar has an asterisk in my mind,” Shew said. “Maybe it’ll happen, maybe it won’t, depending on my next cancer scan or what’s happening in my body. I already live in this world when I’m measuring in shorter increments, when my future has always been planned differently.”
As the rest of the U.S. comes to terms with the same restless impermanence, it must abandon the question When do we go back to normal? That outlook ignores the immense disparities in what different Americans experience as normal. It wastes the rare opportunity to reimagine what a fairer and less vulnerable society might look like. It glosses over the ongoing nature of the coronavirus threat. There is no going back. The only way out is through—past a turbulent spring, across an unusual summer, and into an unsettled year beyond.
Listen to Ed Yong discuss this story on an episode of Social Distance, The Atlantic’s guide to the pandemic:
* This article previously stated that Hong Kong and Taiwan were experiencing a resurgence of COVID-19 cases. In fact, both outbreaks appear to be stabilizing or declining, though it is too early to tell if this trend will hold.