Over the weekend, a research group led by a team at Imperial College in the UK published a paper looking at the effectiveness of various efforts taken by governments to stem the spread of the COVID-19 outbreak. It has been credited with convincing UK and US government officials to get serious about moving quickly and decisively — but it also points to the rough road ahead of us.
Using the latest data from China, Italy, Iran and other hotspots for the virus, they modeled the epidemic curve for COVID outbreaks in the US and UK with no interventions, certain single interventions, and combinations of interventions. The five specific ones they looked at are:
Their model predicted that without any mitigating interventions, 81% of the U.S. population would eventually be infected, leading to about 2.2 million deaths here; keep in mind that this is purely theoretical, because now there are interventions in place nearly everywhere in the U.S. As with other models, theirs “flattens the curve” when interventions are applied. They found that the most effective approach was a combination of case isolation, home quarantine, and social distancing of those most at risk (the elderly and health-compromised); it reduced the demand on critical-care resources by 2/3 and halved the number of deaths. On the other hand, they found that closing schools and stopping short-term mass gatherings had little impact.
They specifically note that social distancing of the elderly doesn’t reduce transmission, but it makes a big difference to hospitalization, ICU and mortality rates since those are the folks who are hit hardest by the virus. The paper includes updated calculations on the hospitalization, ICU and fatality rates for COVID-19 by age:
I used this data, in combination with U.S. Census Bureau demographics on Seattle and King County, to calculate what we can expect locally given the age of the local population:
- We would likely see an infection fatality rate (IFR) of 0.91%; that is 0.91% of the people who become infected will die. This is also based on a finding in the Imperial College paper that about 50% of the infected population become symptomatic.
- About 6.7% of the symptomatic population will be hospitalized; 1.9% will require intensive care.
- About 32% of King County’s population is age 50 or over; however, they will represent 78.3% of the COVID-19 hospitalizations, 95.7% of the ICU patients, and 95.7% of the deaths.
- Likewise, while only 8.9% of the King County population is 70 or older, they will represent 33.9% of the hospitalizations, 63.4% of the ICU patients, and 63.4% of the deaths.
You can download my Excel spreadsheet here, and use it to model the total number of hospitalizations, ICU patients, and fatalities in Seattle and King County depending on the percentage of the population that becomes infected. Since we still don’t have good predictions on where our “curve” will peak and what percentage of the population will eventually contract the virus, it’s hard to predict with any certainty the ultimate number of hospitalizations, ICU patients, and fatalities.
Despite that, the modeling can still predict some of the downstream implications — in particular, the likelihood of a second (or third) “wave” once social-distancing restrictions are lifted. Here’s what the Imperial College research team predicted for the U.S. assuming five months of restrictions that are then lifted, in terms of the need for ICU beds:
It’s worth pointing out that King County is a bit better off than the country as a whole for ICU beds; we have about 650, or 29 per 100,000 people. And that might turn out to be enough for the first wave — until the restrictions are lifted and the inevitable second wave hits.
This really gets to the heart of the problem for Seattle, King County, and the rest of the country. We are putting social-distancing measures in place (fairly strong ones locally, though it varies across regions of the country), but the economic cost of those measures is not sustainable. We need to keep that curve flat until effective treatments are developed, or a vaccine arrives in 12-18 months. But by some estimates keeping the current measures in place continuously until then will lead to 20% unemployment and entire business sectors collapsing — not just locally, but globally. The U.S. government is spending about $1 trillion dollars in relief to try to keep things afloat for the next two months; the cost to do that kind of deficit spending for an additional 10-14 months is astronomical, and will make Bernie Sanders’ “Medicare for All” plan look cheap by comparison. China is in the same bind: they clamped down hard to stop the outbreak at an enormous economic cost, and are now just starting to lift their restrictions. They will inevitably see a second wave as well, and will once again face the awful choice between saving lives and saving their economy (and their citizens’ livelihoods).
There is a different option though, the one that South Korea took. Rather than impose super-restrictive social distancing measures, South Korea opted for widespread testing matched with contact tracing. If someone tests positive, then they self-quarantine and their close contacts are notified to go get tested as well. In this way, clusters can be identified and isolated quickly — in many cases before they become contagious and have an opportunity to infect others. The danger of COVID-19 is that many infected persons are contagious for a day or two before they become symptomatic; but the flip side is that a day or two before they are contagious, they have enough virus present in their body to test positive. South Korea’s system takes full advantage of that, and it’s working. To be clear, they have social distancing and hygiene measures in place too, but nothing as dramatic as what China, Italy and now the United States have been forced to do.
Of course, the United States dropped the ball on testing and even now is still scrambling to make testing available widely. Even if and when testing is available on demand and at scale here, the country doesn’t have the public health infrastructure in place to do contact tracing at the level required to emulate South Korea’s model (though UW virologist Trevor Bedford thinks we can make it work). So we’re stuck with Plan A: draconian social distancing while the economy tanks.
The Imperial College team also took a look at whether there were other options for imposing social-distancing that wouldn’t have such an awful economic impact. One approach they modeled was “adaptive triggering,” where certain restrictions (such as school closures and social distancing in public) would be lifted when ICU cases went below a certain threshold, and re-imposed when they rose back above a trigger level. They estimated that in the UK this might lead to a situation where the full set of restrictions were in force about 2/3 of the time, on a fairly regular cycle. They are not economists, however, and didn’t model the economic impacts such a model might have — or how much better it might be than continuous restrictions.
Today a coalition of Seattle-area business associations released a preliminary study of the economic impacts of the COVID-19 outbreak and the interventions put in place to slow its spread. Have a good, stiff drink before you read it.
In a nutshell: there are two approaches (so far) to stopping COVID-19: the China model of draconian social-distancing measures that have dramatic economic consequences and will most likely lead to a “second wave” when they are lifted; and the South Korea model of expansive, early testing on demand coupled with contact tracking, which the U.S. doesn’t have the public health infrastructure to effectively implement. But there are other strategies that offer a hope of easing this no-win scenario:
Strategy 1: A new test. There are actually two kinds of tests: the one that everyone is using now, which tests for the presence of the virus itself and tells you if you are infected; and one that tests for coronavirus antibodies in your bloodstream and tells you if you have been infected in the past and are now likely immune. This is important information, since according to the latest data only about half of those infected with the coronavirus will show symptoms but they still develop antibodies and immunity. People with immunity can transfer the virus on their hands if they touch an infected surface or droplets from a contagious person (so they still need to practice good hygiene), but they aren’t contagious or sick themselves. That means that they can perform critical services in our healthcare system as well as be the backbone for maintaining and rebooting parts of our economy. Even more important, as the number of immune people grows in the population, we gain “herd immunity” where it becomes more difficult for the virus to be passed on from person to person. Antibody tests are common for other viruses, and several organizations are already developing them for COVID-19. We will still need the test for the virus itself to know who is contagious and must be quarantined, but an antibody test will be a critical tool to forge a path forward.
Strategy 2: Extreme quarantine for the elderly. Remember the statistics I presented earlier: the eldest segment of the population is hit hardest by COVID-19 and represents a wildly disproportionate share of the hospitalizations, ICU cases, and deaths from the virus. If we can protect and isolate them from the general population over the next 12-18 months, we can dramatically reduce the impact of the disease on them and on the healthcare system as a whole — and potentially reduce (though not eliminate) the need for other draconian measures. Now there are lots of ways to do that wrong: we could easily violate their civil liberties and human rights, and just plain make them miserable. We could also inadvertently create conditions such as at the Kirkland long-term care facility that became a major hotspot for the virus, such that if the virus does make its way in, it has devastating effects. But it is well worth thinking through options for humanely increasing protections and isolation for our most vulnerable neighbors until we have a vaccine that gives them lasting protection. If we can do that, it will be much easier to find a path out of our current no-win scenario. And having a known group of workers with immunity to the virus will make doing this much easier (see Strategy 1 above).
That said, in the last 48 hours a new study has cast some some doubt on the conventional wisdom that younger people are not at risk from COVID-19. The CDC’s COVID-19 Response Team has published a first look at the COVID patients in the U.S. between February 12 and March 16. While they generally found results consistent with earlier findings in China related to increased vulnerability by age, they found a higher rate of hospitalization, ICU use and fatality among adults aged 20 through 65 than in earlier studies — suggesting that severe outcomes can occur for any adult, not just the oldest ones. However, there are many issues and limitations with the study, including lots of missing data, many cases still pending, the limited testing of patients to confirm which ones had the COVID-19 virus, and lack of knowledge about other underlying medical issues of the patients. The data coming out of China is far more comprehensive, if for no reason other than that there are so many more cases to-date and many of them are now fully recovered. So while this new study is interesting and worth follow-up, it’s not reason alone to think that the outcomes for COVID-19 will be significantly different in the U.S. than in China.
Strategy 3: Double down on treatment development. There are some interesting efforts going on, including evaluation of whether an existing Cuban medicine, Interferon Alpha 2B, might be effective. An effective treatment, particularly an outpatient treatment program based on a pharmaceutical, would be a game-changer. There is currently no timeline for expected treatments to become available.
Strategy 4: Put a premium on accelerating vaccine development. As with isolating the elderly, there are all sorts of wrong and dangerous ways to do this, and it absolutely should not be done haphazardly. But this is a global pandemic, and millions of lives are at stake. And let’s be honest: this is not the last global pandemic we will see in our lifetimes. Since a virus can be transmitted around the globe in hours or days, spending 18 months to develop a vaccine isn’t going to get the job done. We have to find a better way that dramatically cuts the time to develop and test a safe vaccine for an emerging virus. This is an imperative: it’s the most important thing we can do to mitigate the worst outcomes now and to ensure that they never come to pass again.
While the path we’re on makes sense today given the options available — it is the only viable option we have — the inescapable conclusion is that it is not sustainable, it’s postponing but not avoiding an inevitable spike in cases, and it is leading us into an economic pit of despair. Staying the course and applying economic band-aids for 12 to 18 months is not going to work; we need to be working in earnest on some new exit strategies now.
Fabulous, non-hysterical analysis, thank you.