Early COVID-19 outbreaks are likely already in America: we should do more to stop them

Finding the first few person-to-person cases of COVID-19 in the U.S. will be like finding needles in a haystack. (Photo by freestocks on Unsplash)

In the past week, we’ve seen early stages of COVID-19 outbreaks in Singapore, Japan, and South Korea, and over 500 passengers have now tested positive for the virus on the Diamond Princess cruise ship off the coast of Japan. However, only 15 people in the United States have tested positive so far, putting many Americans at ease that we have successfully contained COVID-19 in this country.

I’m afraid that the low number of reported cases in the United States is giving us a false sense of security. I am writing this post as a follow up to my last post examining the viral growth data metrics behind COVID-19 because I think it is highly likely that we have a small but exponentially growing number of undetected cases in this country, and I believe there are things we should be doing right now to mitigate the spread of this virus before it’s too late. Doing so will buy us precious time as we develop a vaccine, which could take another 12-18 months.

China gives us a glimpse into what the United States could look like in a couple of months if we do not act now. We are fortunate to have this advanced warning, so let’s not ignore it.

How many people in U.S. might be infected?

States with Confirmed COVID-19 Cases. https://www.cdc.gov/coronavirus/2019-ncov/cases-in-us.html

In early February, shortly after the CDC detected the 11th case in the U.S., Professor Neil Ferguson estimated that outside of China “we may be detecting only a quarter of all infections.” Asymptomatic cases and testing accuracy are two factors that contribute to this low detection rate. Professor Ferguson’s estimate implies there may have been as many as 33 undetected cases in the U.S. in early February, in addition to the 11 detected cases.

Multiple outbreaks could silently be happening right now in the United States as a result of these undetected cases. We don’t yet know how viral COVID-19 will be in the U.S., but many signs point towards the virus being extremely contagious with an R0 significantly above 1 both inside and outside of China. Within China, COVID-19 has sickened over 1,700 health workers, and China has even started disinfecting cash with ultraviolet light and high temperatures in an effort to mitigate the spread of the virus. Outside of China, health workers in Japan, France and the UK have all contracted the virus, and even a fire department employee in Japan who was wearing protective gear while transporting a patient off the quarantined Diamond Princess cruise ship later tested positive as well. Michael Osterholm, director of the Center for Infectious Disease Research and Policy at the University of Minnesota, says, “This is influenza-like transmission. It’s like trying to stop the wind.”

According to Professor Ferguson, the epidemic in China was “roughly doubling in size about every 5 days” during the month of January. If we assume that the U.S. had 33 undiagnosed cases on February 2, and further assume that this number doubles every 5 days, that implies we could have roughly one thousand infections in the U.S. by the end of February, tens of thousands of infections by the end of March, millions of infections by the end of April, and possibly over 100 million infections by the end of May. The timing of when we hit these numbers is extremely sensitive to the “doubling time” in the U.S. — with a lower R0, it could take twice as long, and with a higher R0, it could happen even more quickly — but either way we will ultimately end up in the same place if R0 does not drop below 1.

Why haven’t we detected more cases in the U.S.?

If we already have anywhere close to one thousand infections in the United States, wouldn’t we know it by now? Not necessarily.

First of all, according to the WHO, over 80% of COVID-19 cases are mild. So even if 1,000 people were infected, fewer than 200 of them would likely develop severe symptoms. This would be hard to notice when tens of thousands of hospitalizations due to the seasonal flu are happening on a weekly basis in the U.S.

Secondly, the criteria for being eligible for a COVID-19 test in the U.S. have been very strict so far:

Criteria to Guide Evaluation of Persons Under Investigation (PUI) for 2019-nCoV. https://www.cdc.gov/coronavirus/2019-ncov/hcp/clinical-criteria.html

If 33 undiagnosed carriers of COVID-19 unknowingly were to pass the virus along to Americans who have never traveled to China, none of those people would be eligible for a COVID-19 test no matter how sick they were, because they would not have had “close contact with a laboratory-confirmed 2019-nCoV patient.” When my wife had a fever of 104 degrees a couple of weeks ago, no one suggested testing her for COVID-19 because she had not recently traveled to China (it turned out she had the flu, and she is now fully recovered).

In fact, only 479 people in the U.S. have been tested for COVID-19 so far:

COVID-19 PUI in the U.S. https://www.cdc.gov/coronavirus/2019-ncov/cases-in-us.html

This low number of tests does not give me comfort that we have identified everyone in the United States who carries the virus. The CDC has stated that they believe the spread is mainly occurring between close contacts. However, as mentioned above, anecdotes such as the infection of so many health workers, the viral spread on the Diamond Princess cruise ship, and China’s cash sterilization efforts all point toward COVID-19 being extremely contagious, and not something that spreads only among close contacts.

We should be testing more people

At a press briefing on February 14, Dr. Nancy Messonnier, director of the CDC’s National Center for Immunization and Respiratory Diseases, said that the agency is planning to rollout a surveillance system in Los Angeles, San Francisco, Seattle, Chicago, and New York City to test people for COVID-19 if they have flu-like symptoms but test negative for the flu.

Testing all people with flu-like symptoms for COVID-19 in these five cities is a good start, but since there are over one million Americans per week who develop flu systems at this time of year, we will need to ramp up testing as quickly as possible to cover more cities in our search for the first person-to-person transmitted cases of COVID-19. In these early days of the epidemic, it will feel like we are looking for needles in a haystack — over 99.9% of COVID-19 tests in the surveillance system will likely be negative — but if we test enough people, I believe we’ll find the first cases of undetected person-to-person transmission in the U.S.

In addition to testing patients who have flu-like symptoms, I hope that the CDC will begin testing people who are most likely to be exposed to the virus, including doctors, nurses, flight attendants, and cruise staff.

We should be doing more to contain and mitigate

Earlier this week, Japan’s health minister urged the public to “avoid crowds and non-essential gatherings.” For example, on Monday, the Tokyo Marathon was canceled for all non-elite runners. Events in other parts of the world have recently been canceled as well, including Mobile World Congress in Barcelona and Facebook’s Global Marketing Summit in San Francisco.

Other events, however, such as Game Developers Conference and RSA Conference, have opted not to cancel but instead put extra precautions in place, including additional hand sanitization stations and more frequent wipedowns of common touch areas. The tragedy on the Diamond Princess cruise ship has shown us that such precautionary measures to contain a virus in a crowded environment can go wrong.

I believe we should be canceling conventions more broadly as a matter of public policy, rather than relying on individuals or firms to pull out. We should also start developing plans for how people can do more from home, whether it’s attending school online or working remotely, whenever possible. We should be willing to take some economic cost now in order to prevent a future disaster.

Finally, even though the COVID-19 seems to be contained in the U.S., I wanted to reiterate that we should all continue doing the simple stuff, which basically has no economic cost, including washing hands more frequently, greeting others without handshakes, getting a flu shot (if you haven’t already), and cleaning and disinfecting frequently touched objects. My friend, Adam Gries, wrote a helpful post this week suggesting a number of additional precautionary measures that can be taken by all of us at this time. If we all accept a little inconvenience this year, we can make COVID-19 go away. If we don’t, we may be stuck in fear of it for years to come.