This article first appeared on FREOPP.org. It is co-authored with Lanhee Chen, Mark Dornauer, Gregg Girvan, Dan Lips, Avik Roy, and Bob Wachter.
It’s imperative that we responsibly restore the economy while working to flatten the coronavirus curve.
A quasi-consensus has emerged among many policymakers and commentators that the U.S. should continue to close schools and non-essential businesses until coronavirus testing and immunity is widespread. But there is a significant possibility that we are many months, if not years, away from meeting these thresholds. Time is of the essence, given the severe human cost of a prolonged economic shutdown.
The good news is that policymakers have an opportunity to strategically reopen the economy, by taking into account a unique feature of COVID-19: its heavy skew toward bad outcomes in the elderly and the near-elderly who also have other chronic diseases. With the proper precautions, and the deployment of tools like contact tracing, self-quarantines, and telemedicine, we can continue to protect the most vulnerable, while returning as many Americans as possible to work.
(Authors’ note: This is a working paper that will be continuously updated with new evidence and policy ideas. Revisions will be noted at the end of the paper. Please check back frequently.)
Deaths and hospitalizations are continuing to rise in the United States from COVID-19, the disease caused by the novel coronavirus SARS-CoV-2, albeit more slowly than before. Many policymakers and members of the public have expressed the hope that life will be “back to normal” by May or June. But it won’t be. Indeed, life may never return to exactly the way it was before 2020.
And that is the optimistic scenario most widely articulated by public health commentators, in which the U.S. successfully “flattens the curve,” establishes nearly ubiquitous testing, and succeeds at rapidly developing treatments and vaccines for SARS-CoV-2. A recent paper published by former FDA Commissioner Scott Gottlieb and others at the American Enterprise Institute outlines this scenario.
This optimistic scenario leaves us concerned that the public, the media, the business community, and policymakers are largely unprepared for a pessimistic scenario in which accurate, near-ubiquitous testing is difficult to achieve; infected individuals’ antibodies do not lead to immunity; anti-viral treatments take longer to develop; and vaccines never arrive.
We are concerned that the public, the media, the business community, and policymakers are largely unprepared for a pessimistic scenario in which near-ubiquitous testing is difficult to achieve, infected individuals’ antibodies do not lead to immunity, anti-viral treatments take longer to develop, and vaccines never arrive.
Under the present policy paradigm, many commentators and policymakers argue that something resembling the pessimistic scenario will require a prolonged shutdown of six months or longer, or an indefinite series of on-and-off stay-at-home orders that could go on for years.
As Morgan Stanley biotechnology research analyst Matthew Harrison put it in an April 6 op-ed, “Hope that…the U.S. has not reached crisis levels…will be shortlived, as the reality sets in that the path to reopening the U.S. economy is going to be long, and marred by stops and starts. It will be fully resolved only when vaccines are widely available in spring 2021, at the earliest.”
But the U.S. economy cannot sustain a shutdown of that magnitude, nor the permanent closure of tens of thousands of businesses that will accompany it.
Several features of SARS-CoV-2 and the COVID-19 disease lead us to believe that there is at least a 30% chance that the pessimistic scenario unfolds instead of the optimistic one, or — more precisely — that the optimistic scenario will take too long to play out.
Hence, it is essential that we develop a strategy to reopen the economy for both the optimistic and pessimistic scenarios, as well as for a spectrum of intermediate scenarios: that is to say, even if we fail to achieve near-ubiquitous testing, effective anti-viral treatments, and a vaccine. In this working paper, we discuss a plan to do that.
Ubiquitous testing may be difficult to achieve in the near term
There are two principal kinds of tests for SARS-CoV-2: antibody-based or serology tests that detect if a patient has developed antibodies to the virus; and reverse transcriptase polymerase chain reaction (RT-PCR) tests, which measure viral RNA levels in a patient’s nasal secretions. Both forms of testing have significant technical limitations.
Statistically speaking, most antibody-based tests currently yield too many false negative cases, in which infected people go undetected, to be useful for population-wide surveillance.
For example, the latest IgG antibody test from Abbott Laboratories is best at detecting the presence of antibodies two weeks after a patient complains of COVID-19 symptoms, but can yield a high false negative rate, or low sensitivity, before then, dropping down to 50% three days after COVID-19 symptoms. On top of that, there is some evidence that many people infected with SARS-CoV-2 do not ever develop detectable levels of anti-coronavirus antibodies.
SARS-CoV-2 antibody tests will also need to reach a specificity, or true negative rate, above 99.5% to be useful on a mass scale. That is to say, for every 100,000 uninfected people, the test will correctly identify all but 500 of them.
RT-PCR, which measures viral RNA levels in a patient’s nasal secretions, has more accuracy, especially at detecting active infections. But RT-PCR is more difficult to scale up quickly, because RT-PCR is more labor-intensive to administer, and is much more expensive than a conventional blood test.
On March 27, Abbott Laboratories announced the emergency use authorization of a PCR-based SARS-CoV-2 test that it can manufacture at a rate of 50,000 per day. That would be a meaningful boost to U.S. testing levels—at the end of April, the U.S. was testing around 200,000 individuals per day, counting both serology and PCR. PCR tests, however, generally must be administered in clinical settings like a doctor’s office or an urgent care clinic. On April 21, LabCorp announced emergency use authorization for an at-home RT-PCR COVID-19 test kit, but due to scalability issues, LabCorp stated that they are only offering the kit to first responders and health care professionals for the time being.
And in order to reach South Korean levels of coronavirus testing, Matthew Harrison estimates that the U.S. will need to administer 1 million tests per day. In order to repeatedly screen the entire population, Nobel laureate economist Paul Romer estimates that the U.S. will need to reach 22 million tests per day.
Effective anti-viral treatments may take longer to develop
Historically, anti-viral medicines have proven much more difficult to develop than anti-bacterial medicines (a.k.a. antibiotics). While we now have effective medicines for HIV and hepatitis C, these took decades of research to commercialize.
According to the Milken Institute’s COVID-19 Treatment and Vaccine Tracker, there are over 150 treatments for COVID-19 in preclinical or clinical trials. While there is great reason to hope that some of these trials will eventually deliver us an effective treatment against the COVID-19 disease, there is no guarantee that such a treatment will arrive soon.
There is preliminary evidence that convalescent serum exchange could help some patients, but there is no definitive evidence at this time, and even if properly randomized trials show benefit, serum exchange will be costly and difficult to scale.
Kevzara, an anti-arthritis agent also known as sarilumab, has been studied in COVID-19 patients, with the hopes that the drug could mitigate the inflammatory process that seems to make COVID-19 worse in many patients. But Kevzara manufacturers Regeneron and Sanofi announced on April 27 that the drug failed to demonstrate a benefit to patients.
Direct anti-viral treatments also have their drawbacks. On April 16, Adam Feuerstein and Matthew Herper of STAT obtained a confidential recording from a videoconference among University of Chicago clinicians who are participating in a clinical trial of remdesivir, a potential anti-viral treatment for SARS-CoV-2 developed by Gilead Sciences. The physicians expressed hope that remdesivir was helping to rapidly resolve COVID-19 symptoms in their patients. But optimistic reports from physicians at a single site of a clinical trial did not translate into full-fledged success for the broader study.
Indeed, a fuller picture of the remdesivir trial showed no benefit for severe COVID-19 patients. A greater proportion of remdesivir patients died at 28 days (13.9 percent) than placebo patients (12.8 percent), though this was not a statistically significant finding. A National Institutes of Health study of less severe patients found a modest but statistically significant benefit in recovery from hospitalization, with remdesivir patients recovering in 11 days on average vs. 15 for placebo.
On the basis of this somewhat positive result, the FDA issued an Emergency Use Authorization for remdesivir, and Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases, declared remdesivir the “standard of care” for COVID-19 hospitalizations. Paradoxically, deploying remdesivir in all hospitalized COVID-19 patients will make it harder to conduct clinical trials of more effective agents that will no longer be compared to placebo.
Vaccines may never arrive
While there have been several attempts, a successful vaccine has never been developed against a deadly coronavirus, including SARS-CoV-2 predecessors MERS (2012) and SARS-CoV-1 (2003). And while the biopharmaceutical industry is placing an extraordinary effort into developing a SARS-CoV-2 vaccine, effort alone is no guarantee of success; witness our failure to develop vaccines for HIV or hepatitis C. Indeed, the fastest viral vaccine development program ever was for the Ebola vaccine, which took five years.
Many commentators are assuming that antibodies generated by infected patients will protect them from being re-infected with SARS-CoV-2. It is likely that antibodies are protective, but we do not know the extent to which this is true.
Based on experience with the 2003 SARS pandemic, many experts believe that anti-SARS-CoV-2 antibodies are likely to be at least partially protective. But an absence of complete protection would make it challenging to put workers into high-risk situations. A larger challenge is that, due to shelter-in-place measures, most Americans are unlikely to have been infected for many months; hence, over the near- to mid-term, most Americans will remain highly susceptible to this very contagious virus.
In fact, little is known about whether anti-SARS-CoV-2 antibodies are neutralizing; i.e., protective against subsequent infection. “Right now, we have no evidence that the use of a serological test can show that an individual has immunity or is protected from reinfection,” said Maria van Kerkhove, the COVID-19 technical lead for the World Health Organization, at an April 17 press conference.
There are few reports around the world of individuals suffering from COVID-19 on two separate occasions, but they are beginning to emerge. On August 24, 2020, researchers at the University of Hong Kong reported a confirmed case of reinfection, in which the genetic sequence of the virus in the second infection significantly differed from that in the first.
Reinfection is not unprecedented; individuals infected with HIV often possess large quantities of anti-HIV antibodies in their serum that have no effect on the HIV virus or the progression of AIDS. People often develop antibodies to the common cold, which is also a coronavirus, but those antibodies rarely confer immunity beyond one year, if at all.
Americans are optimistic by nature, and the public is well within its rights to hope for the best. But policymakers must prepare for the worst. And that means that we must consider options for reopening the economy in a world in which we have not completely controlled the COVID-19 pandemic.
Time is of the essence
Moody’s Analytics estimates that in only a few weeks’ time, stay-at-home orders have reduced U.S. economic output by 29 percent. Congress has spent over $2 trillion, and the Federal Reserve even more, to prop up the economy merely through the month of April.
Small businesses employ nearly half of all U.S. workers. A 2016 study by the JPMorgan Chase Institute found that “most small businesses hold a level of cash reserves that would provide an insufficient cushion in the face of a significant economic downturn or other disruption,” with the median small business holding 27 days’ worth of cash in reserve.
A study by researchers at the University of Illinois, Harvard, and the University of Chicago estimates that 100,000 small businesses have already permanently closed, with more closures to come. The Paycheck Protection Program, a Congressional relief package passed in late March, helped some small businesses, but most were either not savvy enough or not qualified to draw down federal funds. For example, businesses are not eligible for PPP loan forgiveness unless they maintain their payroll; but for many businesses, maintaining their payroll means they will run out of cash.
The lockdown has profoundly impacted hourly-wage workers, the most vulnerable workers in our economy. An analysis of timecard data from Homebase by researchers at the University of Chicago’s Rustandy Center for Social Sector Innovation has found that hourly workers in all sectors have seen significant declines in available work, with beauty, personal care, leisure, and entertainment nearing zero.
The good news is that there are ways to get more of the economy back on line while we continue to fight to flatten the COVID-19 curve.
Some commentators have propagated a dichotomy in which we can only improve public health through a total shutdown of the economy, and we can only improve the economy—in the near term—by sacrificing public health. This dichotomy is false. New technologies and emerging evidence regarding the epidemiology of COVID-19 will enable Americans to adopt a new strategy which can restore much of the U.S. economy without overly endangering public health.
Some commentators have propagated a dichotomy in which we can only improve public health through a total shutdown of the economy, and we can only improve the economy — in the near term — by sacrificing public health. This dichotomy is false.
Healthy Americans under 65 face lower mortality risks
While people of all ages have succumbed to COVID-19, there is considerable preliminary evidence that infection by SARS-CoV-2 is especially lethal in the elderly, especially in those who have preexisting conditions, such as cardiovascular disease, high blood pressure, diabetes, obesity, kidney failure, and immunodeficiency. Those between the ages of 40 and 65 appear to also be at risk if they have preexisting conditions.
According to the Centers for Disease Control and Prevention, the risk of death from COVID-19 is far greater in the elderly than it is for younger individuals. As of May 8, the CDC tallies 329 deaths associated with COVID-19 among Americans below the age of 35, and only 9 deaths among those below the age of 15. By contrast, there have been 29,661 deaths associated with COVID-19 among those over 65: 79 percent of the U.S. total.
The CDC counts are lower than other official counts due to different reporting criteria. Adjusting for the probability that we reach 120,000 deaths in the U.S., the odds of death for someone under age 35 are roughly 7 per million, or 1 per 140,000.