How do we know when to stop social distancing and reopen the world? It’s complicated. But there are four major goalposts we should meet before we start getting back to normal.
How do we know when to stop social distancing and reopen the world? It’s complicated. But there are four major goalposts we should meet before we start getting back to normal.
Can I buy stuff from China? Should I be worried about excess screen time for my kids? We’re answering your questions about COVID-19 again. Click on the time code in the video description to go straight to a question.
The following originally appeared on The Upshot (copyright 2020, The New York Times Company)
Many experts are beginning to coalesce around a set of benchmarks that could help determine when it might be safe to reopen parts of the country. But even though most areas of the United States are nowhere near achieving the goals necessary to do so, there has been a push to relax social distancing soon.
Even more alarming, some experts say even those still-not-yet-achieved goals aren’t close to enough. It appears that it may be time to think bigger.
“These are unprecedented times, and so we need to think on a scale that would previously be considered unimaginable,” said Natalie Dean, an assistant professor of biostatistics at the University of Florida.
The cautionary tale at the moment is Singapore. For weeks, public health officials have been enviously lauding its response to Covid-19.
Singapore officials have been screening and quarantining all travelers from outside the country since the beginning of the pandemic. Their contact tracing is second to none. Every time they identify an infection, they commit to determining its origin in two hours. They post online where identified infected people work, live and have spent time so that potential contacts can be identified. They enforce quarantines and isolation of such contacts, with criminal charges for those who violate orders.
And yet, in the last week, they’ve put the entire country into lockdown. All migrant workers are confined to their compounds for at least two weeks. Citizens may leave their homes, but only to buy food or medicine, or to exercise. Anyone who breaks the rules, including spending time with anyone not in their household, can be imprisoned, fined the equivalent of $7,000 U.S., or both.
What Singapore was doing (more on that below) dwarfs what most are discussing in the United States. Its present circumstances bode poorly for our ability to remain open for a long time.
“There’s just no way that we’re going to be able to keep most of the country open through the year,” said Ezekiel Emanuel, vice provost of Global Initiatives at the University of Pennsylvania. “If Singapore can’t do it, I don’t imagine how we think we can. As I have said, this is going to be a roller coaster with multiple waves of opening and partial re-closings necessary.”
Given the U.S. government’s limited and lagging response to date, the idea of a hugely ambitious project may seem implausible. But the cost of another future shutdown is so high that previously unfathomable ideas may be worth considering. Here are a few:
Paul Romer, a Nobel-winning economist and N.Y.U. professor, proposes that 7 percent of the population be tested each day. If put on a rotating schedule, that would mean everyone would be tested roughly once every two weeks.
He argues that even if there are plenty of false negatives, if we committed to isolating everyone with a positive test, we could keep the vast majority of Americans out and about in normal life. All told, that would mean 150 million tests a week.
Critics will argue that’s impossible. We cannot even seem to manage a million a day. They say we lack the materials, as well as the reagents for chemical analysis, the delivery infrastructure and the machines to run so many tests.
Mr. Romer is not dissuaded. “I’ve been focused on a single idea my whole career, that just because something is unfamiliar doesn’t mean it’s impossible,” he said. “Building interstate highways, scanning every book, going to the moon — these were all outrageous ideas at one time. But if we put enough resources and our minds behind it, we are able to make the impossible possible.”
His plan would rely less on contact tracing and isolation, since everyone would be tested regularly, and this might make infection control easier in many parts of the country. Contact tracing requires significant infrastructure and is hard to do well.
“We spend something like $700 billion a year to protect us against military threats,” he said. “We are at greater risk from a biological threat at the moment than any military threat. We should be prepared to spend at least a hundred billion a year not only to protect us against this virus, but any potential new viruses that could threaten us in the future.”
Other ambitious ideas can be found in a plan from the Center for American Progress, written by Dr. Emanuel and colleagues. Part of the proposal is an enormous information technology monitoring system. It would call for all Americans to download apps to their phones that would monitor where they go and whom they get near, which would allow contact tracing to be done instantaneously. Everyone could sign in electronically before using public transportation, entering large buildings or schools or gathering in groups above a certain number. They even propose requiring the app to be downloaded in order to receive test results. In an ideal situation, it would run in the background, regardless of whether users signed in.
“If we could do real-time contact tracing based on a person’s phone and GPS signals, and alert people that they have been exposed to a Covid-19 positive person,” that would greatly ease the containment strategy, he said.
Of course, such a system would be considered a large intrusion on privacy, and it’s not clear it is politically feasible — or even legal. Additionally, not every American has a smartphone.
Meredith Matone, scientific director of PolicyLab and an assistant professor of pediatrics at the University of Pennsylvania School of Medicine, says we may need to get away from testing to more grass-roots approaches.
“A more realistic and useful approach would focus much more on surveillance, monitoring communities more than individuals,” she said.
As detailed in a PolicyLab Policy Review, such surveillance could relax our need for active testing. It would be more reliant on passive systems, like monitoring electronic medical records or traditional infection monitoring systems to pick up signals for outbreaks, like increased visits to doctors or emergency rooms for respiratory illnesses. Surveillance could also involve a “participatory approach,” like asking patients to be tested before clinic visits, or to enter symptoms on web-based tracking platforms, or to regularly check their temperatures at home. Thermometers would be ubiquitous, and could even be linked to the internet for reminders and reporting.
If such systems work well, we don’t need to capture an entire population to detect a signal. We could identify hot spots, telling us where to do more focused testing.
Such testing could even be done by pooled samples. In such an approach, areas would have their individual samples combined together for testing, which saves resources. If it’s clear, everyone is safe. If an infection is found, then — again — more focused testing could be done on the individuals in the pool.
The PolicyLab review also highlights the benefits of improving workplace safety, especially in high-risk areas like child care, school and health care environments, to make infection control more robust and surveillance easier to accomplish.
Not everyone thinks we need to aim quite that high. Caitlin Rivers, an author of a recent American Enterprise Institute report on reopening the nation and an epidemiologist at the Johns Hopkins Center for Health Security, said: “While Singapore is adding in some community mitigation measures, they’ve been able to successfully keep levels of infection under control for months, and they’re still only seeing one to two hundred infections a day, which is far fewer than we are. A case-based approach is still the best way to move forward, and while it’s possible that some areas may have to revert to staying home, I don’t think that’s inevitable. Of course, we should still prepare for that with economic aid that can quickly snap into place if that needs to happen so that there’s much less disruption than this time.”
In a city-state of 5.7 million people in an area the size of Indianapolis, Singapore has had 140 people dedicated to contact tracing, working in conjunction with the police. A month ago, it could test 2,000 people each day. That’s the equivalent of testing about 115,000 people in the U.S. We were testing barely a tenth of that amount then.
Singapore has always provided free testing and medical care for all citizens; more recently, it distributed reusable face masks to everyone.
Officials were careful. While stores and restaurants were open, people were told to keep at a distance from one another, and gatherings of more than 10 people were considered inadvisable.
All of this is to say that people in Singapore have been operating in an environment that looks like what we might hope to create as we reopen — with safeguards beyond what we are probably going to achieve. And yet Singapore is in lockdown now.
It’s not clear how tolerant the United States would be of another national pause. If Americans failed to comply, the results could be disastrous. Preventing a second lockdown could even be considered a long-term investment.
“Our trajectory right now does not give me hope,” said Gregg Gonsalves, a professor of epidemiology and law at Yale. “Social distancing is happening in only a patchwork across the United States. The next phase needs a massive national mobilization not seen since World War II, with dramatic scale-up of the production of tests for the virus and its antibodies, the commodities we need to do these tests, from long-stemmed swabs to RNA extraction kits and the personal protective equipment to keep those conducting the tests safe. We also need a huge new cadre of people to do these tests, trained and deployed across the country.”
“And that’s the first step,” he added.
All this sounds expensive. But consider that the cost of a shutdown is trillions of dollars. We clearly don’t want to do this again. As Mr. Romer says, if it costs a couple of hundred billion to avoid it, that may still be a relatively low price to pay.
Aaron’s back answering your submitted questions about all things COVID-19. Time stamps for each question are in the video description, and if you’d like to submit a question for a future episode you can do so at healthcaretriage.info
The following originally appeared on The Upshot (copyright 2020, The New York Times Company)
Everyone wants to know when we are going to be able to leave our homes and reopen the United States. That’s the wrong way to frame it.
The better question is: “How will we know when to reopen the country?”
Any date that is currently being thrown around is just a guess. It’s pulled out of the air.
To this point, Americans have been reacting, often too late, and rarely with data. Most of us are engaging in social distancing because leaders have seen what’s happening in Europe or in New York; they want to avoid getting there; and we don’t have the testing available to know where coronavirus hot spots really are.
Since the virus appears to be everywhere, we have to shut everything down. That’s unlikely to be the way we’ll exit, though.
Some cities or states will recover sooner than others. It’s helpful to have criteria by which cities or states could determine they’re ready. A recent report by Scott Gottlieb, Caitlin Rivers, Mark B. McClellan, Lauren Silvis and Crystal Watson staked out some goal posts.
Other cities and states fear that they will approach New York City’s state of crisis. They’re trying to increase the number of available beds and ventilators — as well as doctors, nurses and other health care providers — to make sure they aren’t overwhelmed in their capacity to provide care to all those who need it.
This is the most immediate bar, and the focus of most public health officials’ attention. At the moment, there’s no reason to believe any area is over a surge of cases, and analysts’ models predict many places won’t peak for weeks to come.
Dr. Gottlieb and colleagues estimate that the nation would need to have the capacity to run 750,000 tests a week — this is after things have calmed down greatly. There are times we might need even more.
“The 750,000 number should be viewed as a reasonable expectation for when we haven’t been having any major pockets or regional outbreaks to manage,” said Mark McClellan, an author of the report and a professor of business, medicine and policy at Duke. “If more testing to help contain outbreaks and potential outbreaks is needed, which seems very plausible, especially early on, the number would need to be significantly larger. We’ll also have to do some surveillance of people without symptoms, especially in higher-risk settings.”
A national estimate means less in deciding whether a state can reopen than its local capabilities. A state would need to be sure it could test every single person who might be infected, and have the results in a timely manner. That would be the only way to achieve the next requirement.
A robust system of contact tracing and isolation is the only thing that can prevent an outbreak and a resulting lockdown from recurring. Every time an individual tests positive, the public health infrastructure needs to be able to determine whom that person has been in close contact with, find those people, and have them go into isolation or quarantine until it’s established they aren’t infected, too.
This will be a big challenge for most areas. Other countries have relied on cellphone tracking technology to determine whom people have been near. We don’t have anything like that ready, nor is it even clear we’d allow it. The United States also doesn’t have enough people working in public health in many areas to carry out this task.
Building that capacity will take significant time and money, and the country hasn’t even started.
Because it can take up to two weeks for symptoms to emerge, any infections that have already happened can take that long to appear. If the number of cases in an area is dropping steadily for that much time, however, public health officials can be reasonably comfortable that suppression has been achieved, defined by every infected person infecting fewer than one other.
In suppression, cases will dwindle at an exponential fashion, just as they rose. It’s not possible to set a benchmark number for every state because the number of infections that will be manageable in any area depends on the local population and the public health system’s ability to handle sporadic cases.
“We wanted to suggest criteria that would allow locations to safely and thoughtfully begin to reopen, but what that looks like exactly will vary from state to state,” said Caitlin Rivers, another author of the report and an epidemiologist at the Johns Hopkins Center for Health Security. “We therefore included some flexibility for jurisdictions to tailor these criteria to their local context.”
These four criteria are a baseline. Other experts think we will need to add serological testing, which is different from the viral detection going on now. This type of testing looks for antibodies in the blood that our bodies created to fight the infection, not the infection itself. These tests can be much cheaper and faster than the ones we’re currently using to detect the virus in sick people.
Testing for antibodies will tell us how many people in a community have already been infected, as opposed to currently infected, and may also provide information about future immunity.
Gregg Gonsalves, a professor of epidemiology and law at Yale, said: “I’d feel better if we had serological testing, and could preferentially allow those who are antibody positive and no longer infectious to return to work first. The point is, though, that we are nowhere even near accomplishing any of these criteria. Opening up before then will be met with a resurgence of the virus.”
He added, “That’s the thing that keeps me up every night.”
Until we get a vaccine or effective drug treatments, focusing on these major criteria, and directing efforts toward them, should help us determine how we are progressing locally, and how we might achieve each goal.
It would also prevent us from offering false hope about when America can start reopening. Instead of guesses, people could have clear answers about when they might be able to go back to a closer-to-normal way of life.
We’re answering 40 more COVID-19 questions from viewers! You can go to the timecode in the video description to see an answer to a specific question. Stay safe!
Most recent policies aimed at mitigating the spread of the novel coronavirus within the United States focus on preventing the transmission of the virus with methods like social distancing, school closures, and paid sick leave. But what drugs are available to treat people once they are infected with the virus? What about drugs that prevent transmission in the first place?
Given that the first cases of COVID-19 emerged less than four months ago, the FDA has not yet approved any drug to specifically treat the novel virus. Even globally, no drug has been proven to effectively treat COVID-19. Clinical trials to evaluate the safety and efficacy of new treatments can take years, though the FDA has mechanisms in place to speed up the process when necessary to “treat serious or life-threatening conditions.” President Trump has publicly called for the agency to do just that for two antiviral drugs, a generic antimalarial drug called hydroxychloroquine and Gilead Sciences’ Remdesivir, which was originally developed as a treatment for SARS and MERS.
The president’s claim at a press conference on Thursday that these two therapies were “essentially approved” to treat COVID-19 is misleading. Neither drug has been approved by the FDA to treat COVID-19, and neither will be approved to treat COVID-19 until months from now, if at all.
As of now, Remdesivir has not been approved by the FDA to treat any condition, let alone COVID-19. A randomized clinical trial of Remdesivir conducted by the National Institute of Allergy and Infectious Diseases is currently underway with patients with laboratory-confirmed novel coronavirus, but it’s still unclear whether the drug effectively fights the virus in human subjects. Some positive anecdotes have emerged, but as of today, there is no solid causal evidence on whether Remdesivir effectively treats COVID-19.
The president’s statement is a bit more accurate in the context of hydroxychloroquine, a generic drug that has been approved for use in the U.S. to treat malaria and other conditions for decades. Aside from some promising anecdotal evidence, though, hydroxychloroquine has not yet been proven definitively to treat COVID-19. Doctors are technically permitted to prescribe hydroxychloroquine off-label, meaning to treat a different condition or indication than the condition the FDA approved it to treat, which applies to patients with COVID-19. Reports of increases in demand for the drug seem to indicate that physicians are doing just that.
But this spike in prescriptions has led to shortages of the drug, causing dire consequences for patients that need the drug for other conditions such as lupus. In response, Teva Pharmaceutical Industries and Novartis have committed to donating millions of pills to hospitals and pharmacies across the country, and Mylan announced that it will ramp up its production of the drug.
It’s crucial to note that no peer-reviewed clinical evidence exists to support the effectiveness of hydroxychloroquine as a treatment for COVID-19. Physicians are either prescribing the drug to coronavirus patients based on hunches or stockpiling the drug in anticipation of forthcoming research. These off-label prescriptions, based only on doctors’ intuition, could prevent patients with lupus from accessing a drug that they need to survive.
Aside from hydroxychloroquine and Remdesivir, there are quite a few new drugs entering preclinical and Phase I testing to treat COVID-19, though these therapies have little to no documented evidence supporting their safety and efficacy as of now. Some companies are trying to engineer treatments for the virus using antibodies from recovered patients, though these drugs are also still in the early stages of development.
Antiviral treatments like hydroxychloroquine and Remdesivir are not vaccines, though, and some experts aren’t mincing words about the necessity of finding one. In a recent article in The Atlantic, Dr. Aaron Carroll and Dr. Ashish Jha stated that “we need to pour vast sums of intellectual and financial resources into developing a vaccine that would finally bring this nightmare to a close.” The National Institutes of Health announced on Monday that clinical trials had begun in Seattle to test a potential vaccine, a messenger RNA-based therapy that “has shown promise in animal models.” As even President Trump admits, though, “it’s still a long process.” And vaccine development absolutely needs to be thorough in order to prevent harmful complications and anti-vaccination sentiments.
One way to bypass the long process of conducting clinical trials to obtain FDA approval for a new drug, however, is to direct doctors to prescribe existing drugs for the off-label purpose of treating COVID-19. But a substantial body of evidence on the safety and effectiveness of a drug needs to be developed for off-label prescriptions to be advisable or even ethical. That’s why researchers at the Quantitative Biosciences Institute at the University of California are testing existing FDA-approved drugs to try to find an already on-market product that can double as a treatment for COVID-19. It’s also why some promising clinical trials are underway in China to test whether favipiravir, a flu drug, could double as a COVID-19 treatment.
Physicians should always keep the murky ethics of off-label prescribing in mind. But as this pandemic continues to devastate human lives and economies across the globe, though, it’s definitely worth considering whether some risks need to be taken to go on the offensive against the pandemic.
Aaron Carroll is answering listener questions about COVID-19, social distancing, and how to stay safe. You can ask questions for future Q&As at http://www.healthcaretriage.info.
This episode was originally released as a YouTube video on 3/18/2020.
Available wherever you get your podcasts! Including iTunes
Yesterday, I canvassed the Michigan governor’s emergency powers, which are surprisingly broad. But they do not cover the waterfront of emergency authority in the state. The director of the Michigan Department of Health and Human Services—currently Robert Gordon—has his own suite of emergency powers.
Once the M-DHHS director determines “that control of an epidemic is necessary to protect the public health,” the Michigan Public Health Code authorizes him to issue “emergency orders” that, among other things, to “prohibit the gathering of people for any purpose.” MCL 333.2253(1). Director Gordon could draw on this authority to adopt statewide or county-specific shelter-in-place rule along the lines of what we’ve seen so far in California.
Emergency orders can also “establish procedures to be followed during the epidemic to insure continuation of essential public health services and enforcement of health laws.” MCL 333.2253(1). The authority here is really broad: the Code clarifies that “emergency procedures shall not be limited to this code,” meaning that the director could adopt rules for areas that he doesn’t normally oversee. The director, for example, could probably draw on this power to adopt new rules for prisons or other state institutions if he feared that prison outbreaks could compromise “essential public health services.”
Violation of one of these emergency orders is a misdemeanor, with a maximum prison term of no more than six months. MCL 333.2261. The director can also create a schedule of fines, not to exceed $1,000 per day, for any violations.
In addition, when the director believes that a crisis “constitute[s] a menace to the public health,” he “may take full charge of the administration” of any laws or rules as necessary to address that menace. MCL 333.2251. As I read it, for example, the director could effectively direct the activities of county and local public health agencies, even if they normally fall outside the ambit of his authority.
Beyond emergency orders, the director can issue targeted orders to eliminate an “imminent danger” to public health. MCL 333.2251. Though the Code contemplates that these orders will be tailored to particular individuals or discrete groups—it says they “shall be delivered to a person authorized to avoid, correct, or remove the imminent danger or be posted at or near the imminent danger”—the orders may be useful if the pandemic moves into a phase in which we are testing widely for coronavirus and isolating positive cases.
Even more to the point, the director has specific authorities with respect to suspected “carriers” of infectious diseases. When such a person is either unwilling or unable to stop putting other people at risk, M-DHHS can serve a warning notice on them. If that notice is ignored, the agency can seek an emergency order from the court and have the carrier taken into custody. MCL 333.5205 & 333.5207; see also MCL 333.2453. The procedure is clunky, however, so it’s unlikely to come into play at this stage of the pandemic.
Finally, M-DHHS has broad authority to “offer free immunization treatments to the public in case of an epidemic or threatened epidemic.” If we get a vaccine, this authority will be key (though the legislature will need to appropriate the money necessary to secure an adequate supply of the vaccine). Conceivably, the director could also issue an emergency order requiring people to get vaccinated if doing so was necessary to protect the broader public health. But we are a long way from that eventuality.
The following originally appeared on The Upshot (copyright 2020, The New York Times Company)
Even as we take significant steps to distance ourselves from one another to “flatten the curve” of the coronavirus pandemic, one of the hardest decisions has been whether to close schools.
There are strong arguments on both sides.
The biggest concern of many experts is that if we get too many infections too fast, the number of sick people could overwhelm the system’s capacity to care for them. By slowing transmission in the population, we flatten the curve, and keep the number of people sick at any one time at a manageable number.
Although most children do not appear to suffer much when they contract the virus (many probably don’t even know they’re sick), they do contract it, and they can give it to others.
Adults can be given instructions on how to prevent person-to-person transmission, and can be relied upon to follow those instructions to varying degrees, but it’s almost impossible to get children, especially younger ones, to do so. If you have a child, you most likely rolled your eyes if you read my recent article about the importance of getting children to wash their hands rigorously, cough into their elbows only, and not touch their face.
Further, the school environment is well suited to spread disease. Students are often packed into small classrooms, where it’s impossible to sit six feet apart. They mingle and form other similar groups by changing rooms to go to different classes. They are put into one large room to eat together, sitting side by side.
It’s not just children that we need to worry about. Plenty of adults work in schools: teachers, janitors, food preparation workers and more. They’re all being put at risk by keeping schools open. Arguably they’re more at risk than many other workers at businesses that have already been shut down.
Closing schools can make a big difference in flattening the curve, evidence from past epidemics shows. A study in Nature in 2006 that modeled an influenza outbreak found that closing school during the peak of a pandemic could reduce the peak attack rate, or speed of spread, by 40 percent. Another study in 2016 in BMC Infectious Diseases found that, based on the H1N1 pandemic of 2009, closing schools could reduce the attack rate up to 25 percent and the peak weekly incidence, or rate of new cases, by more than 50 percent.
Even the Spanish flu pandemic of 1918-1919 provides some data. Comparing cities that took action with those that did not, researchers reported in a study in JAMA in 2007 that measures like school closings contributed to significant reductions in the peak death rate as well as overall deaths.
China and South Korea closed all their schools, and they’re seeing a significant flattening of the curve.
Children are a clear and present danger when it comes to influenza, and almost all of the research cited so far here is based on that. We don’t know whether the studies necessarily apply to coronavirus.
There is also the obvious downside of disruptions to education. Some schools can move to online learning, but not all are prepared. Not all students have access to the internet at home, let alone computers or devices with which they can actively participate in e-learning.
Missing half a semester, which is what many schools are looking at, is a significant hit to education. It will take many students a lot of time and effort to catch up.
There are also effects related to child care. Something like 1.5 million students are homeless in the U.S. For some, school is the only safe space. Many more can’t just stay at home alone. Unless parents can also work from home, and many cannot, children will either be left unsupervised or watched by others, perhaps grandparents. That is possibly the worst outcome because older people are at highest risk of serious illness and death.
We are also facing a potential health care worker shortage. If such workers are forced to skip work to take care of their children, that’s a problem.
Finally, there’s food. Almost 30 million children in America depend on the school lunch program. Almost 15 million depend on it for breakfast, too. If schools are closed with no steps taken to continue to feed them, they will go hungry.
All of these considerations made the decision to shut New York City’s schools highly contentious.
It’s not as if schools must close. They could change their routines. They could commit to increased physical distancing; more and regular hand washing; daily screening; and increased cleaning. Singapore didn’t close schools, and officials there are achieving remarkable success at limiting transmission.
More and more schools have chosen to close in the past few days, reflecting a growing consensus that the benefits of closings outweigh the harms, especially since many of the harms can be mitigated.
A transition to e-learning is possible. So is making sure that parents receive paid sick leave so that children aren’t left with grandparents. It’s possible to provide child care to health workers or those at risk (indeed, many places are doing this). And it’s possible to make sure that food can be delivered or picked up by families that need it.
Many of the changes to make school safer are harder to do, such as cutting class sizes. Short of major structural renovations, we can’t prevent transmission from child to child, and then to adults, if schools remain open. We also can’t currently test the way we’d need to in order to target students who need to be quarantined at home.
The immediate goal is to flatten the curve so that the peak infection rate stays manageable. With better testing and screening, it’s possible to imagine keeping schools open and still protecting families. Failing that, and we in the U.S. have been failing so far, school closures and significant physical distancing are starting to look like the best bet.