It is a serious disease, and we should treat it like one. But the shutdown approach may be imperfect.
Last week, my daughter, a freshman at a Midwestern university, was informed that the spring quarter would be replaced by a remotely conducted quarter. Hundreds of other schools have canceled what remained of their semesters. And there is a long and expanding list of canceled conferences, concerts, shows, parades, and even professional sports leagues. The cause is a new disease, COVID-19, caused by a new virus, SARS-CoV-2. At the moment, it is hard to know whether these responses are warranted. How many people are infected and how many of them will die remain unknown. But enough is known to suggest that fear of the unknown, risk aversion, and acute anxiety are influencing these actions.
COVID-19 was first identified in China in December 2019. The number of cases in China rose rapidly toward 80,000 but, per Chinese-government figures, stabilized by the end of February. The number of new cases began to fall in response to draconian public-health measures, including aggressive identification and isolation of cases and their contacts, quarantines of large cities, and travel restrictions. A similar pattern unfolded in South Korea, which identified its first case on January 20, had an initial rapid rise in cases, and now, following an aggressive program of testing and isolating cases, has seen its number of new cases fall rapidly. Now Europe (especially Italy), Iran, and the U.S. are in the expansion phase. The first U.S. case, a 35-year-old man who had returned from China five days earlier, presented on January 20. As of March 20, 2020, there are about 14,000 reported cases in the U.S. and about 215,000 confirmed COVID-19 cases worldwide. The World Health Organization (WHO) officially declared a worldwide pandemic on March 11.
How scared should we be of COVID-19? The answer depends on determining how many people are actually infected and what the case fatality rate — the number of deaths from the disease divided by the number of diagnosed cases — is. The COVID-19 virus appears to be readily transmissible between people. But determining the number of cases is difficult because the typical COVID-19 symptoms — cough, fever, fatigue — are similar to other illnesses, such as the flu, that are prevalent this time of year. Moreover, 80 percent of cases have only mild to moderate symptoms and do not attract medical attention. Even in South Korea, which has the world’s most aggressive testing regime, only a small fraction of the population has undergone laboratory testing for the virus. Deaths from COVID-19 (the numerator) are far easier to count than cases (the denominator). If the number of cases in the denominator is too low, the case–fatality ratio will be too high.
While early estimates of the COVID-19 fatality rate were quite high, a recent Chinese report put the fatality rate at 1.4 percent of hospitalized patients. These are more severely ill than the general patient who is not sick enough to be hospitalized or even need medical attention, so the number for all patients should be lower. The latest fatality estimate from South Korea is 0.9 percent. But this too is likely to fall as the outbreak evolves and a more accurate account of Korean cases is obtained. That is the usual pattern in epidemic infections. In the 2009 H1N1 pandemic influenza outbreak, for instance, the initial case–fatality estimate was 1 percent. As time passed and we obtained a more accurate case count, the U.S. rate fell to 0.02 percent. At this still early stage of the pandemic, uncertainty remains. The fatality rates in Italy and Japan, both countries with elderly populations, are reported to be near 8 percent and 3 percent, respectively. Germany, in contrast, has a rate of 0.2 percent.
Dr. Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases at the National Institutes of Health, wrote that because of the high number of unreported cases, “the case fatality rate may be considerably less than 1% . . . ultimately more akin to those of a severe seasonal influenza (which has a case fatality rate of approximately 0.1%).” (This is from Fauci, Lane, and Redfield, “Covid-19 — Navigating the Uncharted,” NEJM (2020 February 28): 1–2.) Yet at other times, Dr. Fauci has suggested that the fatality rate could be higher. Only time will tell.
What is known is that COVID-19 is much less lethal than Ebola (50 percent average case–fatality rate over multiple outbreaks) or the two other deadly coronavirus diseases SARS (10 percent) and MERS (36 percent) that made the jump from animals to humans, in 2002 and 2012, respectively. Most COVID-19 cases are mild or even asymptomatic. Nearly all the severe cases and deaths have happened among either elderly people or people who have underlying medical conditions such as lung and heart disease or diabetes. Unlike the case with other respiratory viruses, clinical attack rates — the percentage of people exposed to the virus who will become ill — are low among individuals under 19 years old; if they fall ill, the disease is almost always (97.3 percent of cases) mild.
To put COVID-19 in proper context, it is worth considering another respiratory virus, influenza, that causes the annual, seasonal flu epidemic to which Dr. Fauci referred. Small changes in the genetic makeup of circulating influenza viruses from year to year make the viruses less recognizable to humans’ immune systems and result in illness. In a typical flu season, 27 million Americans will get sick, 59,000 will die, and another 368,000 will be hospitalized but recover.
In the current flu season, which has a month or more to run, the CDC estimates that 36 to 51 million Americans have become ill, with 360,000 to 670,000 hospitalized, and 22,000 to 55,000 deaths. As in previous years, the rate of hospitalization and death is highest among adults over 65. But this influenza season is uncharacteristically severe for younger age groups. Hospitalization rates for children 0–4 and adults 18–49 are the highest the CDC has on record in those age groups. Hospitalization rates for school-age children are the highest they have been in any recent flu season.
Over the past 100 years, there have been four years (1918, 1957, 1968, and 2009) when new influenza viruses with more-significant genetic changes emerged. People had little or no immunologic protection against these new viruses, which therefore spread efficiently from person to person and caused influenza pandemics with higher rates of illness, hospitalizations, and deaths than seasonal influenza does. In a recent paper, the Council of Economic Advisors estimated, from historical pandemic data, that in the most severe pandemic scenario, where a highly contagious virus causes a severe level of illness, more than 500,000 Americans could die.
Some epidemiologists are concerned that, since COVID-19 is caused by a new virus that our immune systems have not encountered before, it will look like a severe pandemic influenza with widespread transmission of the virus, serious illness, and a high number of deaths. Yet it does not appear to have taken that course in China or South Korea. To be sure, both countries applied aggressive containment and mitigation strategies. But it is doubtful — especially in China, where the government delayed acknowledging the disease and starting public-health interventions — that these strategies alone would have led to such a rapid peak and decline in new cases. Unless these countries experience a “second wave” of infections, the total numbers of cases and deaths seem low for countries of 1.4 billion and 51 million respectively.
The fatality rates in COVID-19 may prove to be far lower than the highest rates seen in historical influenza pandemics. We know that COVID-19 seems to spare younger persons and, like the seasonal flu, targets the elderly and those with underlying medical problems. In contrast, the worst of the influenza pandemics, the so-called Spanish Flu in 1918, with a reported case fatality rate of 2.5 percent in the U.S., had a uniquely high propensity to kill healthy young adults and was relatively less fatal for older patients. In addition, projections based on historical data come from pandemics that mostly occurred before modern drugs (e.g., antibiotics to treat often lethal secondary bacterial infections) and critical care were available. Death rates should be lower now, especially in countries with advanced medical systems. Finally, there is preliminary information that treatments for this new pandemic may be available. Chloroquine, an old, relatively safe, and cheap drug primarily used for treating malaria, is being evaluated for COVID-19 treatment and prophylaxis. Some existing antiviral drugs, most notably the as yet unapproved remdesivir, are being evaluated in clinical trials for COVID-19 treatment.
The early imposition of travel bans in this country likely did much to limit the initial spread of disease. The approval of rapid diagnostic tests for COVID-19 that are increasingly available at thousands of public and private labs and through retail outlets such as Walmart, Target, and CVS will facilitate the identification and isolation of those who are infected. Targeting these people while encouraging those who are most vulnerable to COVID-19 — the elderly and those with underlying medical problems — to take proper precautions should interrupt the train of transmission, curtail the spread of the virus, and lower the number of deaths. But the authorities are doing much more. Perhaps too much more.
The large-scale social-distancing and isolation measures being taken to combat COVID-19 — closing schools, banning large gatherings, nearly worldwide travel limitations, and, in many cities, closing restaurants and cafes — are intended to stop transmission of the virus and the spread of illness. However, they are blunt measures that are already having enormous economic and social consequences. Cancellation of sports seasons and tournaments will put everyone who works in concessions, cleaning, and maintenance at sports venues out of work. The same is true for the surrounding businesses that sell souvenirs and meals to fans. Closed theaters and concerts present the same issue. In New York City, closing the Broadway theaters for four weeks will cost an estimated $100 million in lost ticket revenues. And that does not factor in the losses to the surrounding restaurants and businesses. Canceling the South by Southwest (SXSW) festival in Austin, Texas, will have an estimated $356 million economic impact. When things return to normal, workers in large businesses will probably have jobs to go back to. But that will not be the case for workers at the surrounding restaurants and establishments that go out of business.
Important social interactions will also be lost as houses of worship, schools, and bowling leagues close and cancel regular events. As Robert Putnam and Charles Murray have documented, these organizations provide the social fabric that supports our democratic society. The Internet makes social distancing possible while carrying on a semblance of normalcy, but it is only a semblance. Taking classes remotely over the Internet is a sterile substitute for a full undergraduate experience that includes face-to-face interactions in and outside the classroom.
Directives to maintain social distance and to isolate if infected or exposed to someone sick violate the basic human instinct to affiliate, an instinct that becomes more insistent in times of crisis. Such measures threaten to heighten anxiety in this time of uncertainty, especially among older adults and people with disabilities, many of whom already struggle with social isolation. After 9/11, the nation came together in rallies and concerts. Now American are instructed to set themselves apart.
Remarkably, our society does not adhere to the basic public-health measures now being used for COVID-19 when dealing with the annual seasonal flu. Few people take care to wash their hands more frequently during flu season, even though this has been demonstrated to limit viral transmission. People grudgingly acknowledge that handshakes can spread disease, but continue to shake hands in flu season. Society does little to make it socially acceptable and economically feasible to stay home from work if you are sick with the flu. Most tellingly, despite the existence of an influenza vaccine that is known to decrease the chance of contracting the flu after exposure to the influenza virus and to decrease the severity of the disease if one does get sick, less than half the population gets vaccinated.
What can explain the disparate responses to COVID-19 and influenza? Fear of the unknown is certainly at play. Flu is familiar; coronavirus is not.
Humans are also risk-averse. As Nobel-prize winner Daniel Kahneman and other behavioral economists and psychologists have pointed out, loss aversion is one of the most powerful behavioral motivators people have. No public official, especially in an election year, wants to be accused of doing too little. No business owner, educator, or religious leader wants to be blamed for not being concerned enough about the welfare of their customers, students, or parishioners. Moreover, as Kahneman points out, we tend to overweight terrible but low-probability outcomes like a full-blown infectious-disease pandemic without comparing it with the bad, but less terrible, certain outcome that we face every year with the seasonal flu.
Fear and anxiety may also be impairing our judgment. Our leaders may take more-extreme measures than the circumstances warrant. Healthy younger people can’t be bothered to wash hands, forgo handshakes, stay home if they are sick, and get the influenza vaccine during the flu season. But now they hoard, and avoid going outside even though their objective risk of contracting COVID-19 and having a bad outcome is extremely small.
Clearly, earlier availability of widespread, rapid testing would have allowed the identification of cases so that isolation measures could be more targeted. Despite this failure, it is possible that ingenious measures, less extreme than those currently being imposed, could have been devised to accomplish most of the public-health benefits with less economic and social cost. Could airlines have developed rapid screening procedures to discern infected passengers at less cost than canceling most of their flights and flying empty on the remaining ones? Could theaters, concert venues, and places of worship have stayed open at lower capacity by cleaning their interiors before each opening, screening attendees, and spacing them farther apart? Could restaurants have been allowed to stay open at reduced capacity to preserve jobs? Could older, more medically vulnerable members of the population, along with people who are already sick, be counted on to self-quarantine by staying home, especially if provisions were made to supply them with food and medicine? Could schools have closed for briefer periods so they could reassess the situation after a few weeks instead of closing for entire semesters and quarters? We may never know.
And could sending healthy young college students, who are more resistant to COVID-19 infection and suffer less-severe disease if they are infected, back to their homes, where they are more likely to infect older family members with underlying medical conditions, prove to have been the wrong decision? (Given that influenza appears to be more severe among young adults this year than usual, perhaps not.)
As I write, we simply do not know how the COVID-19 pandemic will play out in the U.S. When the dust settles, it may appear that this was an enormous overreaction, or that the measures taken were entirely appropriate, and perhaps even inadequate. I lean toward the former. Nevertheless, I acknowledge that policymakers have to make decisions with incomplete information. Regardless of how it turns out, the COVID-19 experience will hopefully focus peoples’ attention on the commonsense measures they can take in future years to lessen the impact of seasonal flu — and focus public attention on the need to be prepared for future pandemics.