Controlling the Second Wave
Just as states return to some semblance of normal life, commentators and experts have suddenly predicted a very bleak future for the United States. The CDC, for one, has predicted that COVID-19 mortality will double to 134,000 people. It all sounds so ominous. And yet it does not withstand scrutiny.
Most of these dire predictions assume that people will forgo any preventative measures. This is not credible. People are aware of the lethality of the disease and have changed their behaviors. Government mandate will not affect that.
The Georgia mortality graph in figure 1 illustrates this phenomenon [GA Department of Health Data]. There is about a 3-week lag between onset of coronavirus disease infection and death. When Governor Kemp declared a public health emergency on March 14, the mortality curve started to bend three weeks later. It rolled over 3 weeks after his imposition of even stricter standards, a statewide shelter in place order on April 2nd. The point is that people were most likely engaging in some of the other behaviors associated with disease-prevention before governors imposed significant restrictions.
No one can deny that COVID-19 has profoundly altered human behavior. People no longer shake hands; they actually wash or disinfect their hands; sick employees stay home; many telecommute; vulnerable people are isolated; workplaces are disinfected; and they social distance like never before. These are all important measures that mitigate the spread of the coronavirus. None of these are likely change even after we have progressed through all three phases of the White House/CDC plan.
The adequacy of even mild prophylactic measures has been obscured by the media’s construal of Sweden’s epidemiological results. Contrary to the media’s obfuscation, Sweden has bent their curve as readily as its neighbors, Denmark and Norway as shown in figure 2. But Sweden has not closed restaurants, bars, elementary schools, casinos and other establishments. They have required some distancing in these different settings, and some reports from Sweden indicate that people are voluntarily restricting some of their activities. We analyze this in more detail in our Sweden Norway Denmark strategy post.
The “sunshine” effect will also dampen the spread during the summer months. Warm weather states such as California, Florida, Texas, Arizona, New Mexico, and the territory of Puerto Rico have all fared relatively well. Densely populated states should have a higher rate of infection than those less-densely populated. But Puerto Rico, California and Florida are in the top 12 of most densely-populated states. Hawaii, which is the 14th most populated, has the lowest mortality rate in the country, 1% of New York’s. It is theorized that UV light disrupts the lipid envelope of the coronavirus. Sunshine will help contain the spread of COVID-19.
Much of the predicted mortality increase is a double-hump phenomenon. Most of these analysts predict a decrease followed by an increase once society is opened. This “light switch” model is shown in figure 3. The model below assumes that we return to 100% contact for 20 and under, 80% for the general public, and contact for the vulnerable population is held at 50% relative to the pre-COVID-19 period.
If we follow a phased approach as recommended by the White House CDC recovery plan this second hump will not occur in the summer as shown in figure 4. If there is a resurgence, it will be next winter that can be dampened by re-imposing some modest measures, such as suspending large public events, such as large indoor sporting events or concerts, that carry the potential to be super-spreader events.
Although much has made of the increased infection rates in certain hotbeds, such as towns with packing plants, these infections should result in a lower mortality rate than expected because so many of these employees and their families are young and in good health. Even in these hotbeds, most of the deaths are occurring in the elderly with chronic diseases. So, an uptick in infections does not necessarily entail a significant number of deaths because most of these workers in this demanding industry are fit.
Moreover, these towns are so insular that the outbreak will have mostly subsided by mid-June. These towns will be deluged by infections over a very brief period of time because there is not much social distancing either in these plants or in the community. Nebraska ER doctor John Safranek has observed that it is not unusual to have a carload of four or five packing-plant workers coming for COVID-19 testing. If one of these is positive for the disease, then so will his buddies before long. If there are even a few infections in these plants, there will be hundreds – very swiftly. And that has been the trajectory in these towns.
We must obtain more data to learn how much risk different events cause. Do public outdoor events carry more risk than indoor events? Does a classical music concert carry less risk than a rock concert? We must acknowledge that we do not know the answers to many questions we have about COVID-19. But we will only obtain the necessary data by actually allowing liberalization of the extant restrictions.
We must acknowledge a debt of gratitude to Sweden and the states that have been early adopters of liberal restrictions. If not for Sweden, we would not know that lighter restrictions can be effective against the coronavirus. We will also obtain valuable information from the states that are returning to work. And we want these states to try different approaches: if some measures fail, better that it be in one state and not all 50 as part of a national policy.
America has changed. Whether this will be for the better or not is yet to be seen. But the epidemiologic models do not account for the changed behavior that will most likely persist for some time. States do not have to restrict individuals if most people will voluntarily engage in sufficient prophylactic behavior. It’s time to get back to work. It’s time to implement the White House CDC recovery plan.