This is Going to be Bad Ep. 04
Something is Wrong with the Numbers
One of the big problems with trying to figure out how bad Covid-19 is going to be, is that we don’t have a clue about how many people are really infected. Last Thursday night around this time there were 471,000 confirmed cases with a reported death total of 21,287 (Johns Hopkins Coronavirus Resource Center). Tonight, there are 722,289 confirmed cases with a death toll of 33,984. Clearly the virus is not under control. The big questions though, are “what is the number of undetected infected” and “why is the US mortality rate so low”? The answers to those questions are crucial to figure out, because they will tell us what Covid-19’s true mortality rate is and just how bad things are going to get.
So, right now there are four ways to look at Covid-19’s mortality rate. There are the:
Observed Global Mortality Rate (OGMR) which is about 4.5%
Projected Global Mortality Rate (PGMR) which is expected to be about 2.25%
Observed US Mortality Rate (OUMR) which is about 1.75%
Projected US Mortality Rate 1 (PUMR1) which is estimated as being about 1.1%
Projected US Mortality Rate 2 (PUMR2) which is estimated to be about 0.1% (about the same as the flu)
This is a crazy degree of uncertainty with vastly different death tolls. For example, in the US the number of possible fatalities ranges from a low of about 200,000, which was the number that was given today (Sunday the 29th), to the 2.2 million estimate given last Thursday by Dr. Fauci, to a worst case high of almost 8.9 million depending on which mortality rate you use*. To reduce that uncertainty, it is critical that we get a handle on the actual mortality rate of Covid-19.
The problem, is that we don’t know the number of undetected infections. This lack of knowledge is crippling our ability to know just how deadly Covid-19 really is. Let’s start with a hard number, the OGMR which is obtained by dividing the reported number of deaths by the reported number of cases. It is 4.5% and it represents the “worst case” mortality rate because it makes no allowance for undetected cases. It assumes that every case is detected, which makes the “observed mortality rate” the actual mortality rate. Under this case the US death toll could be as high as 8.9 million. This is the worst possible case, but no one thinks it is going to be this bad.
The reason, is that no one thinks that the number of reported cases is the number of actual cases. Everyone knows, that for every confirmed case, there are “x” number of undetected cases. These are the infections that are asymptomatic or so mild that they get shrugged off. The cases that a policy of, “if you don’t have symptoms, you don’t need to get tested” results in not being counted. The question is, how many of these are there?
If we make a pessimistic guesstimate, that for every global confirmed case there is only one other undetected case, the total number of cases jumps to roughly 1.4 million and the mortality rate drops to about 2.25%. This is about the mortality rate of the Spanish Flu of 1918 and it seems like a reasonable number based on the speed that the infection is spreading and the number of asymptomatic people who have been detected when mass testing (groups where everyone got tested) has occurred. If you think this rate is too high, remember that the mortality rate in Italy is currently just over 10%. Given the observed mortality rates being reported on the ground, this is probably a conservative estimate. In this case the US death toll would be about 4.46 million.
This is not a number that anyone wants to see. It is a devastating number that would represent a total failure of the federal and state health services to control the spread of Covid-19. It is the kind of number that could spark real panic in the country if the White House Coronavirus Task Force admitted it was a possible outcome. So, they haven’t used this number and have gone with several much lower numbers.
The first number was given in the daily briefing by Dr. Fauci last Thursday when he asked about how high the death total might go. He stated that 2.2 million deaths were possible. This works out to a projected mortality rate of about 1.1%, slightly lower that the current OUMR of 1.75%, but not out of the realm of possibility. The problem is that this number makes no sense.
Our testing is so bad that everyone knows we are missing a lot of cases, particularly cases that are asymptomatic or mild. In an attempt to sooth people it has been endlessly pointed out that Covid-19 is not as lethal as it looks because, “when you factor in all the cases we are not detecting, the mortality rate will drop way down”. Which is true. However, in this case we can calculate what the under count estimate the White House team is using and know how many “extra cases” they need to make their projections work.
Tonight (March 29th) there are 142,300 confirmed cases in the US with 2,493 dead. This makes the OUMR 1.7%. When Fauci cited the 2.2 million fatalities estimate, he was using a projected mortality rate of 1.1%. In order to get from 1.7% to 1.1% you must assume that there are now about 70,000 additional undetected cases for an under count of about 30%. This is a ridiculously low estimate. In order to get from the OGMR of 4.5% to the PGMR of 2.25%, which is what the WHO is using as it’s projection, assumes an under count of 100%. Knowing the problems with testing in the US, it is absurd to think that we are catching 2/3 of all the infections.
We know that the number of undetected infections must be much larger than this. Which means, that to get a PUMR1 rate of 1.1% that isn’t based on “under count”, you must believe that 1.1% is the actual mortality rate of the Covid-19 virus. The numbers don’t support it but this might be true. The mortality rate of the Princess cruises that were severely infected with Covid-19 was roughly 1%. This may indicate that the final mortality rate for the virus will be about 1% when everything is said and done. Still it seems very early to to be projecting this as the “worst case”.
Which leads us to the PUMR2 of 0.1%, about the same as the annual flu. This is the mortality rate that the President is using when he, and the Task Force, talk about 200,000 people dying. In order to get to this mortality rate, based on the observable numbers, you would have to assume that there are over 2 million additional undetected cases in the US. Or that 9 out of 10 who get Covid-19 are not getting a test because their symptoms are so mild.
While I am willing to believe that testing in the US is only detecting 10% of the cases, it seems unlikely. Our infection curve simply isn’t high enough for there to be that many infected in the population right now. If there were that many people out there spreading it around, the number of people we would be seeing in the healthcare system would be much higher. There simply cannot be that many infected in the country right now. Which makes this mortality estimate of 0.1% inexplicable. Unless you remember what President Trump, has said and what he wants to hear.
President Trump has repeatedly stated that Covid-19 is no worse than the flu. The commentary on FOX News has repeated this statement as well. Covid-19 they have assured the public is “like the flu”. We are told that the elderly and the frail are at risk, but for the rest of the population the virus is “no worse than the flu”. For this to be true, the true mortality rate of Covid-19 would have to be about the same as the flu, which coincidentally is about 0.1%. Assuming a mortality rate for Covid-19 of 0.1% is the only way that you can get to a death toll of about 200,000.
There is no observable evidence that the mortality rate for the Covid-19 virus is equivalent to that of the flu. All the evidence indicates that the final mortality rate is likely to be between 1.5% and 2.5% which means that US fatalities are probably going to be in the three to five million range if the progression if the virus is not slowed down significantly in the next 3–5 weeks.
At this point, where things are going is difficult to tell. There is about a two-week lead between the average time of infection and when someone goes to the hospital (which is when most people are being tested). This means that the numbers we are seeing today are where the virus was two to three weeks ago. If the voluntary social distancing that the President put in place 13 days ago has had any impact, the day over day numbers of new infections should begin to drop soon. If they do not, it will indicate that the policy was a failure and that more drastic steps, like a nationwide lockdown, will be necessary. The question is, will the President have the moral courage to take the necessary steps? Based on what we have seen so far, and on the way that the Task Force is trying to play down the likely number of fatalities by using very questionable numbers, it seems unlikely that we are going to shut this down the way China claims to have done.
This is going to be very bad this Spring and probably into the Summer.
*Note: All US mortality rates are calculated by estimating 60% infection penetration of a population of 330 million for an infected population of 198 million over the next 2 years. This is worst case, in the event that no vaccine is developed in the next 24 months and social distancing fails to significantly slow down the dispersion of the virus.