Scary Numbers

It continues to astound me that this country is incapable of acting in response to clear and urgent signs of danger. It does not require advanced degrees in infectious disease to realize that the rapidly climbing infections in the United States starting in early October could lead to bad things.

Lets start with the relationship between cases and deaths. The Case Fatality Rate (CFR) is a measure of people who have been found to have COVID who will die of the infection. For more details see my post on CRF. This figure has changed over time, as we have expanded our testing, and improved our hospital care for COVID-19 patients. But over the short term it is a reliable predictor of the immediate future. CFR will vary by state and the age and health status of the group of infected patients, but with a nice large sample size like the current number of daily cases you get very predictable data. Figure 1 shows the US CFR over the course of the pandemic with the rate at 2.3% as of November 14th.

Figure 1: US CFR over the course of the pandemic:

As the rate is obviously trending down, I will use a more conservative figure for predicting the near future and assume a CFR is at least 1.5% for the coming weeks.

Using 1.5% means that when 100,000 people are infected with COVID-19, sometime between the next few days to perhaps 10 weeks, at least 1,500 of those 100,000 people will die. New daily infections in the United States have averaged over 100,000 since November 6th. On Friday November 13th, there were 183,000 new infections ( That means at least 2,745 people are pretty certain to have received a death sentence that will be executed in the next few weeks. These daily infection figures are likely to continue for some time because to date the United States has done very little to slow this latest spike. During the first wave the pandemic the 7 day average number of deaths peaked about 2,200. Its simple math to see that within a very short period of time we will exceed that number of daily deaths.

Part of the decline in CFR reflects improvements in the medical system’s ability to treat COVID-19 patients. Recent research on the decline of deaths among hospitalized patients point out that the higher hospitalization deaths were at least in part likely the result of hospitalizations being overwhelmed. Its pretty basic, if you have adequate staffing and supplies, you have better outcomes.

One of these studies for a single hospital system in New York shows when adjusting for the change in ages and medical conditions in their patient population, the percent of hospitalized COVID-19 patients dying in the hospital went from 23% down to 8%. That means instead of nearly 1 in 4 hospitalized patients dying, only 1 in 12.5 patients are dying. This is of course very good news. But given these numbers we can not only predict the number people likely to die based on the number people with new infections, but we can also predict the number of hospitalizations likely to occur.

So again, if we think about 100,000 newly infected cases, we can pretty confidently predict that 1500 of them will die of COVID-19. Because the vast majority of these will die in hospital we can estimate the number of hospitalizations resulting from the 100,000 new cases as 1500 x 12.5 = 18,750 patients. Friday’s 183,000 new cases will likely generate over 34,000 hospitalizations. I have not done the research to figure the average duration of COVID-19 hospitalizations, and the overall hospital capacity in the United States. But the current COVID-19 hospitalizations are running just over 60,000, and it is beginning to put serious stress on some state systems. When we start having an average of 34,000 or more new hospital admissions on a daily basis, you can bet there will be people dying in parking lots outside of hospitals waiting for care.


None of this is rocket science. When the daily infection rate started curving upward in early October anyone with a small amount of math could see we were heading toward a crisis. Of course, not only do we have simple numbers that predict the coming crisis, months ago many public health experts predicted there would be a new rise in cases in the fall that would likely be more severe that the initial outbreak. The fact that there was no federal leadership to head off this crisis is stunning. Most state governments have been slow to address the coming crisis as well. Dark times are coming. For now, its time to roll back the amount of social interactions we all engage in, practice more social distancing, wear our masks and keep washing those hands.

2 thoughts on “Scary Numbers

  1. Thanks for your comment. Yes I have been toying with trying to figure the appropriate offset between case and deaths. It seems to vary a bit, but I think its in the 21 to 35 day range. Also the death curve is a different shape, it ramps up pretty quickly, but does not go to as big a peak, and the tail is more elongated. Also different spikes in different places have different characteristics. The summer spike in Texas and Florida was initially among younger healthier people, but I believe it gradually got more evenly distributed, but that resulted in deaths being more delayed, as they occurred more from 2nd or 3rd cycles of infection, rather than the first.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Create your website with
Get started
%d bloggers like this: