Coronavirus Scenarios
Data-Driven Analysis Suggests We Remain Calm
I hope you had a chance to read The Greatest Time to Be Alive last week. I continue to see tremendous opportunities for all of us, especially those who see through the noise to appreciate the unique gifts available to us this season ranging from extra time with family if working from home to the generational wealth that thoughtful investors will create. In this post, I explore a number of coronavirus scenarios for the US based on the available data.
In short, reasonable estimates indicate that this will be no worse than the typical flu season if we follow CDC measures and in even the most aggressive scenarios the number of deaths will be roughly the same as the annual quantity of deaths due to heart disease. While of course any loss of human life is tragic and we should strive to prevent it, current reactions of panic appear out of proportion to the actual threat and may in fact be worsening the situation given that fear and depression weaken the immune system and leave us more susceptible to the spread of disease.
I recommend to stay calm, assess the situation objectively, and act from a place of informed intelligence. Of course this includes following CDC guidelines. Additionally, I suggest that you exercise, eat healthy, cultivate a beautiful psychological state, and generally leverage this as a reminder to care for yourself and your loved ones because life and its many wonders are generally far more precious than we remember day-to-day.
Please note that this post focuses on the data using back of the envelope calculations given the limited information available and comparisons to other threats to human life so that we can gain fact-based perspective. Please don’t take the emphasis on analysis as callous, dismissive or demonstrating lack of compassion. In the past 72 hours, I have reached out to family, portfolio companies, friends and everyone I can reach in our community offering support and advice on how to handle this situation because I care deeply about ending human suffering and our mission at Prime Movers Lab to transform billions of lives.
Relevant Data
At the moment, the media is citing many numbers about coronavirus mortality rates that inspire fear but do not set context. First, let’s look at general mortality rates independent of coronavirus.
General Mortality By Age
Here are the general mortality rates by age in 2018 in the US caused by all conditions such as heart disease, cancer, injury, chronic respiratory disease, stroke, Alzheimer’s, diabetes, influenza, pneumonia, etc…. These numbers are largely similar to 2017, as are the breakdowns by cause.
The key takeaway is that older individuals are already far more likely to die independent of any pandemic and that the oldest populations face significant mortality rates regardless of any particular disease or outbreak.
General Mortality by Cause
Here are mortality rates for the primary causes of death in 2018 in the US.
In 2018 in the US, the flu caused over 30,000 deaths, chronic respiratory diseases caused 132,000 deaths, and cancer caused 494,000 deaths.
Coronavirus vs. General Mortality
First, it is important to note that coronavirus mortality data has a much smaller sample size compared to the data above on general mortality. Second, note that these mortality rates by age group for coronavirus are based on confirmed cases in China, which are likely far less than the true number of cases since many cases are occurring without severe symptoms or detection. This means that actual coronavirus mortality rates are likely lower than the data below suggest. With all those caveats, here is the available data for coronavirus mortality in China (since that is the only country where there is anywhere close to sufficient data to be significant).
The main thing to notice is that mortality rates for each age range due to coronavirus in China are comparable to general mortality rates in the US!
For the 80+ year old population, mortality rates are similar given statistical uncertainty. For the 50–80 year old population, Chinese coronavirus mortality rates are roughly double general American mortality rates. For those <50 years old, there is no significant difference in mortality rates.
That is a key fact that I am not seeing in many places in the media that certainly reassured me (or it might cause one to wake up from our general acceptance of death as a civilization — check out Gordian Bio, one of our portfolio companies making strides to address that).
Note that given how the data is segmented by age slightly differently, that of course has an impact on this comparison. That being said, the fact that the Chinese coronavirus mortality data is based on only confirmed cases significantly outweighs the impact of the mismatched age ranges.
Geographical Data
Let’s look at the populations, number of confirmed cases, number of deaths, percentage of the population with a confirmed case, percentage of the population dead, and percentage of the cases resulting in death for various geographies across China, Italy, Iran, and South Korea as of March 14. Note also that testing rates have varied widely and so these comparisons certainly have some inaccuracy built into them.
Top takeaways from this data are:
- Outside of Hubei province (Wuhan is the capital of Hubei) where the virus started in China, mortality rates were 0.9% of known cases (likely lower given that known cases are only a fraction of total cases).
- Nowhere has more than 0.12% of the population been confirmed with the disease and this is at the least-prepared center of origin of the disease. While this percentage is almost certainly low, even if it is 10x too low, that would mean only 1% of the population in the hotspot origin of the outbreak caught the virus. As we will get to below, a typical American flu season sees 10% of the population catch the flu.
- Even in the location of the highest mortality rates, Italy at 7.2% (due to a more elderly population with a long history of smoking), that still means only .0029% of the population has died of coronavirus.
Flu Data
Here is the data on cases, hospitalizations, and deaths for the last decade of flu seasons in the US.
Note that 37,000 people die from the flu in the US each year on average and 446,000 are hospitalized due to the flu in the US in a typical year.
In 2017–2018, flu cases rose to 45 million, hospitalizations reached 810,000, and deaths soared to a total of 61,000. This suggests that claims about lack of hospital beds are likely being overstated in the media. More on this below.
Finally, it is important to also note the flu incidence rates by age group to set us up for our our scenario analysis below. For ages 5–17, 7.9% contract the flu on average. For ages 18–49, 7.4% contract the flu on average. For ages 50–64, 12.0% contract the flu on average. For ages 65+, 3.9% contract the flu on average. Now, let’s turn to scenario analysis based on the all data above.
Scenario Analysis
US Cases and Mortality Mirror Other Nations
In the first scenario, we consider what it would mean for the US if the rate of coronavirus infection and mortality rate mirror other countries including Hubei province, China, Korea, Italy, and Iran.
Here is the punchline! If the US coronavirus spread mirrors Hubei, the worst outbreak in the world so far, the US will have 379,000 cases and 17,000 deaths. This is less than 10% of the worst recent flu season case volume and less than 1/3 of the deaths from that year due to the flu.
Even if every case required hospitalization (which most do not), it would be half the volume of hospitalizations due to the flu in the worst year recently. Not only is the coronavirus unlikely to surpass a typical flu season’s case volume and number of deaths, it is also unlikely to overwhelm our hospitals as long as we stay calm and do not overreact in panic. Please see the section below on flattening the curve and strain on medical resources for more.
A More Aggressive Scenario
Let’s also consider a more aggressive scenario. Again, let’s assume the incident rate matches Hubei such that there are 379,000 cases. Now, let’s use the highest mortality rate so far recorded, which is Italy at 7.2% largely due to the fact that it is a more elderly population and many are smokers.
In this scenario, there would be approximately 27,000 deaths, still placing the volume of deaths at roughly 2/3 of a typical flu season. Does this seem to warrant the level of concern in the media that is currently happening across the country?
Worst Case Scenario
Now, let’s consider the worst case, not because it is likely to happen, but just to evaluate if it justifies the current level of media frenzy. To do this, let’s assume that each age group contracts the coronavirus at the same rate as they have historically caught the flu (i.e. with roughly 75x+ frequency compared to Hubei infection rates for coronavirus). Second, let’s use the mortality rates by age group from the data in China to compute total deaths.
So in the worst case scenario, there would be 21 million US cases (less than each of the last 8 years of flu seasons) and 295 thousand deaths. While that volume of deaths is tragic, it is less than 7 percent of the 4.1 million deaths that occurred in the US in 2018 and only twice the volume of deaths due to chronic respiratory disease and flu.
Note that the worst case scenarios being covered in the media are relying purely on epidemiologist modeling given transmission rates and do not account for any of the substantial measures already taken in the last week. Additionally there is nowhere in the world where those worst case scenarios have happened.
Even in the 1918 Spanish flu pandemic being cited, only 27% of the population was infected and that was over the course of three separate flu seasons spread out over nearly a year. There is effectively no chance of 40–70% of the population getting infected in a matter of weeks (or even months) as the current news coverage would suggest, which is a key assumption of the suggestion that the medical system will be overwhelmed shortly.
Flattening the Curve
Many of us have read about the need to flatten the curve (i.e. slow down the rate of spread of the disease) in order to minimize strain on medical resources, in particular ICU beds and ventilators in hospitals for the most severe cases at risk of death. Most of that coverage is relying on the unlikely 40–70% infection rate projection mentioned above and significant underestimates of the available ventilators. Let’s look at the data on availability and what the scenarios suggest about the possibility of overwhelming the medical system.
Note that in China 80.9% of cases were mild, 13.8% severe (requiring an acute care bed), and 4.7% critical (requiring a ventilator). Here we compare the number of severe cases to the number of acute care beds and the number of critical cases to the number of ventilators. We examine this both for a situation in which the US has a similar percentage of cases as Hubei and the situation in which the US has flu-like levels of infection.
Also note that for the flu-like levels of infection, we assume these occur over a three-month period similar to a typical flu season (given that is already extremely aggressive to assume that flu-like levels of infection occur since the highest rate of population infection so far is <75x that). We also display the projected deaths for both scenarios and an adjusted computation of projected deaths accounting for the potentially limited quantity of ventilators.
This is all based on data suggesting total capacity of 200,000 ventilators (many publications are under-reporting this number) and 534,000 acute beds in the US. Note that many of these are occupied (although anecdotal reports suggest current occupancy is lower than usual as many people are avoiding hospitals as potential hotspots of transmission) and on average 33% of ventilator capacity is being used at any time.
The takeaway is that if infection rates are comparable to Hubei, then there will not be any strain on the hospitals or ventilator supply. If infection rates are comparable to the flu (and similarly spread out over a few months), we have on the order of 50% of the acute beds and ventilator supply needed. So an aggressive projection would be a doubling in the mortality rate (though that is way too aggressive because the Chinese death rates already reflect far worse shortages of beds and ventilators).
Even in this extremely unlikely scenario of flu-like spread and twice the mortality rate of China, total deaths would be 590,000. While this is of course horrible to contemplate, we must hold the context that such a death toll would be roughly equivalent to deaths in the US due to heart disease.
Conclusion
To conclude, the data suggests that if coronavirus takes the same course in the US as the Hubei province of China (by far the worst outbreak so far), it will result in less than 1/3 of the deaths in the worst US flu season from the last few years. Even in the worst case scenario, it would lead to only twice the volume of deaths in 2018 due to chronic respiratory disease and flu. Of course we should follow the CDC guidelines and take care of ourselves with proper exercise, nutrition, and rest; however, we shouldn’t panic and make the situation worse. Let’s stop buying into the media-driven panic, remain calm, and find the beauty in this opportunity to enjoy time with loved ones at home.
Prime Movers Lab invests in breakthrough scientific startups founded by Prime Movers, the inventors who transform billions of lives. We invest in seed-stage companies reinventing energy, transportation, infrastructure, manufacturing, human augmentation and computing.
Sign up here if you are not already subscribed to our blog.
