Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
Coronavirus by Country: March 21 Update
I have a new chart today and an update of a prior chart. As before, I’m reporting on South Korea, Italy, Germany, France, Spain, the US, and the UK. Here is a bar chart of the average daily growth rate of reported WuFlu cases, by country, in five-day increments.
The general trend is downward, which is good news, indicating that the growth is not following an exponential curve.
The US is the exception. This should not be surprising, because the US is well behind most of the other countries listed, in the progression of the disease. The latest figures are of concern, however, so let’s all pray for a downward trend soon.
The shorter-term trend in the US is a bit more favorable. Our growth rate peaked at 48.9% on March 19, declined to 40.6% yesterday, and declined again to 37.7% today.
Here is an update of the graph that I first posted yesterday, showing total reported cases per million, with the trend line starting when each country first surpassed 10 cases per million.
Using this trend line, the US is 28 days behind S. Korea, 24 days behind Italy, 16 days behind France, 15 days behind Spain, 14 days behind Germany, and 2 days behind the UK.
My data source is Worldometer for both WuFlu cases (here) and population (here).
If you like the graphs, please “like” the post. I’m not fishing for praise, but trying to calibrate whether I’m providing useful information, or am becoming annoying. I do tend to excel at the latter.
Published in Healthcare
You’re doing a great job of informing us, Jerry, and thanks. Even when you and I have tussled over an issue of two, I’ve always had a healthy respect for your ability to analyze complex information! Keep up the good work.
Thanks, Gary. I appreciate your posts, too. This particular issue seems more up my alley.
Let’s hope that warmer weather, come May, will beat the living daylights out of this horrid little virus. Then we can get back to more enjoyable things, like your great Star Wars post a few months back.
Using this trend line, the US is 28 days behind S. Korea, 24 days behind Italy, 16 days behind France, 15 days behind Spain, 14 days behind Germany, and 2 days behind the UK.
Thanks so much for compiling these graphs, @arizonapatriot, and especially for extrapolating the info above for those of us who are trying to understand the big picture while also battening down the hatches at home. This visual representation showing the different trajectories of S Korea and Italy is excellent, and makes me wonder if we will have two distinct disease trajectories in urban/suburban vs small town America. In smaller communities our schools, gyms, and big gatherings have also been cancelled per our governor’s orders, but for once we have a distinct advantage over more densely populated areas. Our local county commissioners, law enforcement officers and health officials are much closer to the communities they serve and can respond quickly to rumors, and are also able to communicate more directly with constituents who have questions and concerns.
I loved living in and going to college in Philadelphia way back when and most of the time wish there was a big airport nearer to us in Florida, but definitely not this month!!! Thanks again for compiling this info in graph form.
I’d like to see data on testing capability for countries. The ability to run good tests seems to be an indicator of having things under control.
Diane, I don’t expect that the urban/rural divide explains the difference between Italy and S Korea, because: (1) S Korea is heavily urban, and (2) much of Italy’s problem appears concentrated is a few relatively small cities or towns. You may be right, but I don’t think that we can draw your conclusion from the data presented.
I suspect that there is a temperature aspect, but I have not analyzed it myself, and the reports on this possibility are early.
That’s what comes from me commenting late at night, lol. I wasn’t drawing conclusions from the actual data you presented. It just inspired a new line of thought when I saw the difference in curves so well represented with your graph. Sorry to go off-topic like that.
(FYI in the early days of Ricochet before I became a sandwiched caregiver I used to comment more carefully and thoughtfully here. I’ve mainly lurked for several years, though, because I was pretty self-conscious about potential mistakes I could make when commenting, or the potential abandonment of a post I’d written when the need to tend to family made me check out suddenly. My situation has stabilized some over the past year, though, so I’m trying to become more active on Ricochet again now despite my continued inability to spend a lot of time crafting careful posts and comments as so many of you do so beautifully here!)
Thanks again for your patience as I try to explain better what came to mind last night:
Looking at the graph reminded me of my own situation in rural Florida, where less populated counties seem to be more successful at accomplishing social distancing and locating potential coronavirus cases early. I’m thinking that with both the lower density of people and much more effective community self “policing” without the need (yet) for stay in place orders, we could end up more successful at achieving a South Korea-like graph here. Nothing to do with temperature, as the other counties in Florida with no cases yet also seem to be rural. I’ll double check that now.
This may not seem to be an important conclusion to many, but considering the relatively low income and education level of many of our area’s residents, and higher than average median age, I think it’s interesting that we have no cases yet. We do have 12,000 students in grades K-12 here and about 8% of those students are children of migrant workers, plus it was the peak of our tourist season and we had many drivers and crew members from abroad here early practicing for the now-cancelled 12 Hours of Sebring, and we also have a variety of armed service personnel who come and go from the Avon Park Air Force Range, so it’s not like we’ve been sitting here in splendid isolation all winter either. And we have a good number of area residents who work in surrounding counties that do have active cases so they or people driving though stopping for gas will probably transmit the virus to someone local sooner than later. It’s the complete lack of reported cases so far in the entire rural 6 county region that is interesting to me. Thanks for letting me clarify that!😁
Kozak, you are not correct about this in your definition of exponential growth. I’m working on a separate post explaining the modeling issues, which are complex.
You are correct that the UK and US are in the stage during which the growth is well-modeled by an exponential curve. In other countries, this phase has ended after 1-2 weeks (meaning 1-2 weeks after reaching 10 reported cases per million).
His definition is correct as far as I know.
I am correct. Your understanding of “exponential” incorrect.
MATH FIGHT!!!
Any word on how the Netherlands is doing? Their “herd immunity” strategy might end up paying off if they can manage to stay on the curve while not decimating their economy.
I think any discussion of case growth in the US right now must mention the rapid increase in our testing capacity.
We can probably assume that patients who present at the hospital with symptoms have been tested since tests were available. But greater availability of test kits means that more people can be tested as a result of contact tracing — people who, previously, might have been told to quarantine at home without being counted as cases.
In other words, we could reasonably expect an increase in diagnosis rate greater than the rate of increase in true infection. That will be true as long as we continue to significantly increase our testing capacity.
I believe the only number we can really trust is the mortality figure, and that one is, as discussed elsewhere, mostly representative of the spread of the infection prior to the various shutdown and social distancing efforts.
We have a reasonable handle on the number of people who are clinically ill. We don’t have a good idea of how many asymptomatic or nearly so carriers are out there (by that I mean people who have a viral infection but won’t get (noticeably) ill vs those who are infected and haven’t become sick yet.
Because of how we’re testing so far that data is going to be missing for quite some time; this contributes to the confusion.
No, guys, you’re incorrect about this. Exponential growth involves a constant rate of growth. The formula in these circumstances is y=a(1+r)^t, where r is the growth rate, a is the initial figure, and t is time. The caret (“^”) indicates exponentiation, which I use here because the Ricochet text function does not allow superscripts or subscripts.
This is the same formula used for compound interest.
When the daily growth rate is decreasing, the growth is not exponential by definition. In the exponential growth formula, r is a constant.
The Wikipedia entry on this is here.
And if you plot that with the y axis on a log scale, you get a straight line.
That’s correct. Here’s a log scale graph of the same data as the OP, adding a trendline for 33% daily exponential growth:
Update with data reported through March 22:
I’m planning to do a full post every 2-3 days. In the meantime, most of the reports for today (March 22) are favorable, with the daily growth rate in total reported cases down almost everywhere:
This shows a significant decline is the growth rate for all of these countries, which is good news by itself, and great news if it continues.
The one negative report is at home. The US reported 38.6% growth on 3-22, essentially unchanged from the 39.0% daily for the preceding 5 days.
This bears watching, but remember that the US is well behind the track of all of these other countries. The US just reached 101 cases per million, while almost all of the others are much higher (ranging from 245 per million in France to 977 per million in Italy).
Only the UK is at approximately the same point as the US, with just 83 cases per million.
On the other hand, these figures may be an artifact, as it is a weekend and we don’t know whether the medical facilities in the various countries are open to the same extent.
Exactly.
Hi Jerry: Would it be all possible for you to do a chart involving how many tests have been taken. I would agree with @HenryRacette that the US numbers look worse because we’re finally able to get the testing up and running. Another interesting graph would be # of tests and # of positives for the virus. Thanks so much from someone who will definitely NOT get into the math dispute.
Colleen, I don’t have a good source of data for this.
Unfortunately, I don’t think anyone does. We’re doing lots of tests, but we’re a big country with lots of independent groups trying to integrate hastily collected statistics.
I know: let’s put the Iowa Democratic Party in charge of aggregating the test results. Maybe they’ve got an app for that….
The Covid Tracking Project collects US testing data. See their site for data explanations & caveats. Here’s their testing data combined with the Worldometer case data for the US.
(pos = positive, neg = negative, pdg = pending tests; cases on left axis, tests on right)
(data table used to make chart here)
Thanks so much for this! This is why I like Ricochet and Ricochetti.
Those who understand write intelligent posts.
Those who don’t, post comments (I am also guilty of this myself)
Ruh Roh
I certainly hope they are using hydroxychloroquine and remdesivir for more serious cases. The idiot governor of Nevada has banned the use of hydroxychloroquine.