A VERY Simplified Comparison Between Influenza and Coronavirus Death Rates in the US · 4:40pm Mar 25th, 2020
Since it gets brought up a lot these days, I'd find it fascinating to know: how do these two viruses compare? Say, on death rates? What sort of numbers are we talking about here?
First, a few clarifications, and then the numbers themselves. Please note that all this is for casual purposes only: this is not an official analysis!
The comparison is an idle one, of raw numbers, done by me basically because I like looking at numbers to get an idea of scale and measurement. While I do have my own opinions on them, ultimately what they mean beyond that is something only a qualified statistician or epidemiologist can really say. However, since a lot of people are comparing influenza to the current coronavirus strain, I thought I'd dig up some interesting info on both. Any mistakes in the blog, I'm guilty as charged!
Also note that this is purely for recorded death rates in one country, the US. It doesn't say anything about, say, non-fatal damage caused by either disease, nor does it tell us how those death rates change relative to other factors (policy measures to counter it, for instance), nor does it cover infection rates, global incidence, hospitalization rates, correlation with other diseases, and so on. (Needless to say I would be very interested in comparing it to other countries, and to a global count).
Lastly, for those wondering: my sources for numbers are the W.H.O. Situation Reports, and the CDC FluView resources on P&I (pneumonia and influenza) mortality rates.
OK, numbers time.
Firstly, for COVID-19, the disease caused by the SARS-CoV-2 virus, the current notorious strain of coronavirus making the news:*
*(Please note this is based on the most recently available data. If I'd done it tomorrow, the weekly numbers would look quite different. Think of these as handy checkpoints, not firm comparisons, because picking numbers from an existing dataset can so easily confuse if you're not careful. I'm confident mainly because the acceleration trend is so obvious.)
As of March 3rd, according to the W.H.O. (World Health Organization), the death toll from COVID-19 in the USA was 2 deaths.
As of March 10th, according to the W.H.O., the death toll from COVID-19 in the USA was 19 deaths.
This indicates an increase of 17 deaths in one week, from March 3rd to 10th.
As of March 17th, according to the W.H.O., the death toll from COVID-19 in the USA was 58 deaths.
This indicates an increase of 39 deaths in one week, from March 10th to 17th.
As of March 24th, again according to the W.H.O., the death toll from COVID-19 in the USA was 471 deaths.
This indicates an increase of 413 deaths in one week, from March 17th to 24th.
You can find the infection and death rates in the tables in each report (the US data is usually on page 4, 5, or 6).
Now for influenza, a regular menace with a seasonal oscillating cycle:**
**(The data for this came from this CDC FluView Mortality resource: https://gis.cdc.gov/grasp/fluview/mortality.html. If you have trouble with the Disclaimer that pops up, try tapping the tab key a couple of times. That should bring up the OK button needed to get past it. Then you can pick the relevant "season" i.e. year of the influenza cycle at the top right, select a particular week within that season on the line scale at the top, and check the table of data for that week and a couple of nearby weeks at the bottom of the page for the absolute numbers. Right, off we go!)
By comparison, the worst week of flu deaths in the US according to the CDC (Center for Disease Control and Prevention) was during Week 3 of 2018. It suffered 1,626 deaths.
In fact, the nine worst weeks of flu deaths in the US (since 2013, which is the earliest available data in the dataset) surpassed 1000 deaths (the tenth worst was just under it, at 981).
Of course, most weeks don't get that high.
Let's assume for the moment that an average is informative (and one can raise doubts as to that, given the presence of more extreme outliers can skew an average, but humour me a moment). In which case, according to the CDC, from Week 40 of 2013 to Week 10 of 2020 (basically all the data available on influenza deaths included in this Excel Spreadsheet***, which was also the corroborating source for the previous paragraph)...
***Go here directly: https://www.cdc.gov/flu/weekly/weeklyarchives2019-2020/data/NCHSData11.csv. This data came indirectly from going to https://www.cdc.gov/flu/weekly/#S2 and then clicking on the "View Chart Data" link below the graph. Note that it'll open the downloadable graph as an Excel Spreadsheet, so make sure you've got the software needed. Fortunately, all the data there is also available on that CDC FluView Mortality resource: https://gis.cdc.gov/grasp/fluview/mortality.html that I mentioned earlier, so it's technically possible to access the exact same data there too; the spreadsheet just lumps it all together, so larger trends can be studied.
...then the average weekly death rate from flu is approximately 151.83 deaths per week, with a range from 1,626 deaths to just 1 death.
Please note that this average I came up with is not explicitly mentioned in the Excel Spreadsheet: I simply selected all the numbers in the "Influenza Deaths" column (that's column h) and ran a basic automatic formula to get the average. The spreadsheet also lists pneumonia deaths, if you're interested in further comparisons.
Anyway, comparing the two diseases...
Here's one way of comparing the two diseases: if we listed all the weeks of flu deaths since Week 40 of 2013, from the worst week for deaths (highest) to the best (lowest), then when lumped in with all those 336 weeks of flu deaths, the coronavirus's most deadly week (reminder: that's 413 deaths from coronavirus in a week)...
...would place 43rd out of 337 weeks (the 336 flu weeks plus the 1 coronavirus week).
To reiterate: that's 43rd out of 337 weeks.
To be fair, this is pretty cheeky as a comparison, as we've got hundreds of weeks' worth of data on influenza death rates and not so many on the current coronavirus. Technically, a fairer comparison would involve hundreds of weeks' worth of data on that too.
Of course, in the US at least, in order for the SARS-COV-2 virus to surpass, in one week, the worst week of flu deaths, then in the span of one week it would have to surpass the "target" of 1,626 deaths from flu. Basically, it's got another 42 competitors to beat before it places 1st.
To put that into perspective, 1,626 deaths in a week would require approximately 232.29 deaths on average per day. Or nearly ten every hour.
Thus far, the highest death count recorded in a single day in the US for the current coronavirus was on March 23rd, at 201 deaths in a day. At the time, that boost brought the total up to 402 deaths, basically doubling it overnight.
That was two days ago, and while the next day (March 24th) yielded a lower 69 new deaths in one day, fluctuations like that are expected in long-term statistical trends, so there's potential for further rises and new record-breakers.
Of course, all this will become obsolete once more information arrives, and will need updating. So we'll have to see how the coronavirus behaves over the coming days, weeks, months, or however long it takes to stop accelerating.
So, fellow users... any predictions for March 31st?
That's all for now. Impossible Numbers, out.
Honestly, the 50 States are each so different from each other that trying to compare numbers for the US as a whole doesn't seem very informative to me. As of today, New York has over 30,000 confirmed cases of SARS-CoV-2 and 285 deaths (which makes sense, given how densely populated New York City is), while over here in Minnesota, we have 279 confirmed cases and only 1 death, and going on down to West Virginia, only 39 confirmed cases with no deaths (source: Johns Hopkins University Coronavirus Resource Center).
Does anyone know if there are statistics for influenza deaths broken down at the State level?
5228317
That'd be this one, correct?
Happily, you can find those in the same source as for the national count in my original post: https://gis.cdc.gov/grasp/fluview/mortality.html
The page defaults to "National", as indicated at the top of the screen. All you have to do is select "State" instead, and then use the map on the left to pick a state. Haven't tried it myself yet, but it sounds like what you're looking for.
5228317
Understandable. That's a problem internationally too, most obviously with the criticisms that China has been underreporting on that front. I'm just relying on what official sources report, so far the W.H.O. but also checking against others as I become aware of them.
Generally, I find it useful to check multiple sources and see if they're within the same ballpark. I only recently found out, for instance, the CDC have their own tally for total infection and death counts.
Currently, their total death count for the US is 737, but according to Georg it was 400 yesterday. I'll be seeing how this compares to the WHO as soon as Situation report - 65 comes out.
5228325
That's the one.
You, my friend, are a font of information.
5228342
Heh, thanks.
But it's really the other sites which are responsible for the data. I'm just a signpost, myself.
5228349
A FONT OF INFORMATION
And... UPDATE!
As of March 25th, according to the W.H.O., the death toll from COVID-19 in the USA was 673 deaths.
That now, according to that source, makes today the new record-holder for highest death count recorded in a single day in the US for COVID-19 (202 deaths in one day, compared with 201 deaths on March 23rd).
EDIT: It's worth noting for fairness's sake that other sources provide different counts at the moment. There is the Centers for Disease Control and Prevention count here, for instance: they currently put the death toll in the US at 737. Meanwhile, the John Hopkins University and Medicine Coronavirus Resource Center interactive map currently puts the death toll in the US at 894.
I'm not exactly sure what factors are present to make each count differ from the others, but in any case, the broad gist is that the number of deaths is well into the hundreds and should be approaching its first thousand soon, if current trends continue.
5228488
As of March 26th, according to the W.H.O., the death toll from COVID-19 in the USA was 884 deaths. (211 deaths in one day).
As of March 27th, according to the W.H.O., the death toll from COVID-19 in the USA was 991 deaths. (107 deaths in one day).
At last count, the Centers for Disease Control and Prevention put the death toll in the US at 1,246.
The John Hopkins University and Medicine Coronavirus Resource Center interactive map currently puts the death toll in the US at 1,478.
(PLEASE NOTE: These last two sources update regularly: these counts will become obsolete within the next 24 hours).
5229980
As of March 28th, according to the W.H.O., the death toll from COVID-19 in the USA was 1,243 deaths. (252 deaths in one day).
As of March 29th, according to the W.H.O., the death toll from COVID-19 in the USA was 1,668 deaths. (425 deaths in one day).
At last count, the Centers for Disease Control and Prevention put the death toll in the US at 2,112.
The John Hopkins University and Medicine Coronavirus Resource Center interactive map currently puts the death toll in the US at 2,513.
(PLEASE NOTE: These last two sources update regularly: these counts will become obsolete within the next 24 hours).