Spirit, Speak Comfort To Me!

OK, before someone gets upset I’m not taking the current crisis seriously, don’t misunderstand: this is a serious situation. But there is no cause to lose our sense of humor or be grim. Yes, we must take action, but no, it’s not the end of the world (unless you’re a nurse or doctor, then it might feel like it for a couple of weeks).  I’ll crunch the numbers downthread and it’s not as bad as you might think if you keep perspective.  But do not doubt the sad fact that the US health care system can’t really keep up with a normal flu season; there is no way it can handle a rapid influx of respiratory patients. That is why COVID19 is so dangerous, and why everyone needs to take it seriously, following the CDC guidelines, limit interactions outside your immediate household (aka social distancing), keep strict hygiene protocols, and otherwise doing everything you can to try to slow down the rate of spread. It’s more than likely not about you. It’s about that 1% of so of the population who will get very sick, and may not get enough care because the system will be overloaded.

Here’s the latest analysis.  First, please, please, please, stop obsessing on every blip in the numbers!  They are not “skyrocketing” or whatever inflammatory phrase the media is using at the moment. Second, the absolute numbers don’t matter.  Yes, each and every one is a life, and a tragedy.  But what matters in terms of risk is the denominator: how many people are getting sick and passing away in terms of what size group?  Losing 100 people in Chatham County (pop about 290,000) is very different from losing 100 people in New York City (pop 8.6 million).  Don’t compare them.  It’s mortality per unit population that matters – and how fast that mortality happens.  Please stop feeding the beast by quoting and hyping how many deaths per day without context.  It’s not helpful, and causing people far more stress than is appropriate.

Time for some math: deaths from the virus are progressing along what is known as a logistic function.  This type of function was originally developed for use in population growth, but has found it’s way in to many other fields.  In biology, this is sometimes called a carrying capacity curve.  We are entering the scary part of that curve.  Here’s what the curve looks like with data for several areas as of 24 March 2020.  The black line is a theoretical curve that represents an estimate of how things might progress.  The grey line is for comparison, the 2017 H3N2 flu progression speeded up by a factor of 5.  The dots represent actual data as of the totals for yesterday as reported this morning.  Click to embiggen …


Of these, China (grey +) and Iran (blue o)  look weird.  I suspect those numbers are “munged.”  But Spain (red dots) and Italy (green dots), which are the farthest along of societies that might resemble the US, seem ok.  (South Korea is even further down the curve, but they took very early intervention, and have more hospital surge capacity than the US, so may not be a good analog).  New York is the cyan triangles that are hard to see because  it’s just starting to creep along the curve – about day 25 or so.  The next 20 days will be very scary – you can see that is the steepest part of the curve, and people will talk about doubling times, and extrapolation the daily rates far beyond the point where they will start to  settle down.

Where will it end?  The latest projections are that the US will see between 50 and 90 thousand deaths.  (2017 saw 61 thousand H3N2 influenza deaths – but over 6 months, not 6 weeks!).  New York will likely see upwards of 5 thousand (currently 200 or so).  Smaller communities will also see a rapid rise in deaths that, without context, will seem terrifying. Expect the health care system to be in crisis, and please do what you can to support the medical community.  This will be horrific for them.  Chatham County, Georgia hospitals, which serve about 400,000 people, will likely see nearly 1,000 respiratory cases, of which 100 may die, all in the next three weeks.  But again, by the end of April, most parts of the country should be at the upper end of the curve, with the deaths per day decreasing.

How soon will we know if that is our future, or something worse?  Italy should be at near their peak.  I expect that by early next week we will see a downward trend in their numbers, followed by Spain 4-5 days later.  If by the 1st of April Italy is still recording 700 or more per day, that will be a source of concern.  Will update the graph this weekend … meanwhile, don’t hoard TP like this guy.

15 thoughts on “Spirit, Speak Comfort To Me!

  1. Hey! Do you agree with the information Sam Wang is putting out there, and if so, am i correct in interpreting what he’s saying (generally) like this- “Since reported cases and tests might be wacky math, the one true piece of information we can look at is the rate of deaths caused by COVID19. Once the death rate stops doubling over a three day period, then it is slowing down and there is a light at the end of the tunnel.” Thanks!

  2. There is a Covid19 update on accuweather.com showing numbers for cities, states, and countries listing cases, recoveries, and deaths. The numbers in Chinese cities are extremely questionable. They are either outright lying or they have a miracle cure that the rest of the world isn’t aware of. With your research tools, could you make some sense of this? Thank you for your down to earth posts.

    • It is absurd for accuweather to be showing infectious disease numbers. That is just simply click-baiting.

      All of the numbers are really suspect because nobody is using the same metrics. Who is tested and when, etc. are different even between counties in the same state, much less between states or countries. And some countries (China, Iran) are clearly not reporting accurate numbers.

  3. A key source of data for all the major charts is the Georgia Department of Public Health daily status report….today, the report shows 1247 confirmed cases. The report goes on to break this down by county. At the tail end is a county named Unknown, which accounts for 153 cases. That is all fine, the number is less than 20% of the total and I guess could be a rounding error….but yesterday, the same line for deaths stated that 23 of the 38 deaths were in this Unknown county. That is a higher number by percentage.
    I am trying to understand the impact on my community….Chatham County, which has a low number of confirmed cases and zero deaths…but if the Unknown County is really some of our numbers, I’m concerned.
    Finally, I am concerned about testing after death….if a person is brought into the hospital from an advanced care facility, and is declared DOA….is testing done?
    Thank you for what you do….

  4. Thanks so much for the perspective! Refreshing reprieve to the inflammatory headlines. I’m a little muddy about the line for the H3N2 flu progression… you said it’s ‘speeded up by a factor of 5’. Is that because it was collected over a longer period of time than Covid data?

  5. Dr. Enki-

    Do you have any thoughts or perspective you might share on Dr. Birx’s comments from yesterday (which CNN apparently cut away from – I was watcing online).

    “I’m sure you have seen the recent report out of the U.K. about them adjusting completely their needs. This is really quite important. If you remember, that was the report that says there would be 500,000 deaths in the U.K. and 2.2 million deaths in the United States. They’ve adjusted that number in the U.K. to 20,000. Half a million to 20,000. We are looking at that in great detail to understand that adjustment.

    I’m going to say something that is a little bit complicated but do it in a way we can understand it together. In the model, either you have to have a large group of people who a-asymptomatic, who never presented for any test to have the kind of numbers predicted. To get to 60 million people infected, you have to have a large group of a-symptomatics. We have not seen an attack rate over 1 in 1,000. So either we are measuring the iceberg and underneath it, are a large group of people. So we are working hard to get the antibody test and figure out who these people are and do they exist. Or we have the transmission completely wrong.

    So these are the things we are looking at, because the predictions of the model don’t match the reality on the ground in China, South Korea or Italy. We are five times the size of Italy. If we were Italy and did all those divisions, Italy should have close to 400,000 deaths. They are not close to achieving that.

    Models are models. We are — there is enough data of the real experience with the coronavirus on the ground to really make these predictions much more sound. So when people start talking about 20% of a population getting infected, it’s very scary, but we don’t have data that matches that based on our experience.”


    • That’s seems to be reasonable. I’m assuming when she says “the model” she is referring to the Imperial College model that freaked everybody out last week. The thing is, that is just one model, and by their own admission were using a series of “worst case” assumptions based on data from very early in the outbreak. The models I’ve been using are using more realistic assumptions based on near real time data. As of today, things are still on track for US having between 22,000 at the low end and 80,000 at the high end, with the “best estimate” at 65,000 (almost exactly the same as the H3N2 outbreak in 2017 – albeit much faster, 4 weeks instead of 20). These models I have been running have been pretty stable over the last two weeks in that range.

  6. The schools are currently mandated to stay closed until April 24th. At which point we’ll be at the upper end of the curve?

    • Georgia is about three weeks from the upper inflection point of the curve, so assuming the progression stays on track, my guess is that will be about a week too early, and things will stay closed in until the end of the month. But at that point it will be more about politics than science …

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.